Search results for: convolutional architectures

Authors: Sara A. Ben Lashihar

Abstract:

The heritage building information modelling (HBIM) of the historical masonry buildings has expanded lately to meet the urgent needs for conservation and structural analysis. The masonry structures are unique features for ancient building architectures worldwide that have special cultural, spiritual, and historical significance. However, there is a research gap regarding the reliability of the HBIM modeling process of these structures. The HBIM modeling process of the masonry structures faces significant challenges due to the inherent complexity and uniqueness of their structural systems. Most of these processes are based on tracing the point clouds and rarely follow documents, archival records, or direct observation. The results of these techniques are highly abstracted models where the accuracy does not exceed LOD 200. The masonry assemblages, especially curved elements such as arches, vaults, and domes, are generally modeled with standard BIM components or in-place models, and the brick textures are graphically input. Hence, future investigation is necessary to establish a methodology to generate automatically parametric masonry components. These components are developed algorithmically according to mathematical and geometric accuracy and the validity of the survey data. The main aim of this paper is to provide a comprehensive review of the state of the art of the existing researches and papers that have been conducted on the HBIM modeling of the masonry structural elements and the latest approaches to achieve parametric models that have both the visual fidelity and high geometric accuracy. The paper reviewed more than 800 articles, proceedings papers, and book chapters focused on "HBIM and Masonry" keywords from 2017 to 2021. The studies were downloaded from well-known, trusted bibliographic databases such as Web of Science, Scopus, Dimensions, and Lens. As a starting point, a scientometric analysis was carried out using VOSViewer software. This software extracts the main keywords in these studies to retrieve the relevant works. It also calculates the strength of the relationships between these keywords. Subsequently, an in-depth qualitative review followed the studies with the highest frequency of occurrence and the strongest links with the topic, according to the VOSViewer's results. The qualitative review focused on the latest approaches and the future suggestions proposed in these researches. The findings of this paper can serve as a valuable reference for researchers, and BIM specialists, to make more accurate and reliable HBIM models for historic masonry buildings.

Keywords: HBIM, masonry, structure, modeling, automatic, approach, parametric

Procedia PDF Downloads 137

Authors: Mohamad R. Moshtagh, Ahmad Bagheri

Abstract:

Accurate fault detection in rotating machinery is of utmost importance to ensure optimal performance and prevent costly downtime in industrial applications. This study presents a robust fault detection system based on vibration data collected from rotating gears under various operating conditions. The considered scenarios include: (1) both gears being healthy, (2) one healthy gear and one faulty gear, and (3) introducing an imbalanced condition to a healthy gear. Vibration data was acquired using a Hentek 1008 device and stored in a CSV file. Python code implemented in the Spider environment was used for data preprocessing and analysis. Winner features were extracted using the Wiener feature selection method. These features were then employed in multiple machine learning algorithms, including Convolutional Neural Networks (CNN), Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN), and Random Forest, to evaluate their performance in detecting and classifying faults in both the training and validation datasets. The comparative analysis of the methods revealed the superior performance of the Wiener-CNN approach. The Wiener-CNN method achieved a remarkable accuracy of 100% for both the two-class (healthy gear and faulty gear) and three-class (healthy gear, faulty gear, and imbalanced) scenarios in the training and validation datasets. In contrast, the other methods exhibited varying levels of accuracy. The Wiener-MLP method attained 100% accuracy for the two-class training dataset and 100% for the validation dataset. For the three-class scenario, the Wiener-MLP method demonstrated 100% accuracy in the training dataset and 95.3% accuracy in the validation dataset. The Wiener-KNN method yielded 96.3% accuracy for the two-class training dataset and 94.5% for the validation dataset. In the three-class scenario, it achieved 85.3% accuracy in the training dataset and 77.2% in the validation dataset. The Wiener-Random Forest method achieved 100% accuracy for the two-class training dataset and 85% for the validation dataset, while in the three-class training dataset, it attained 100% accuracy and 90.8% accuracy for the validation dataset. The exceptional accuracy demonstrated by the Wiener-CNN method underscores its effectiveness in accurately identifying and classifying fault conditions in rotating machinery. The proposed fault detection system utilizes vibration data analysis and advanced machine learning techniques to improve operational reliability and productivity. By adopting the Wiener-CNN method, industrial systems can benefit from enhanced fault detection capabilities, facilitating proactive maintenance and reducing equipment downtime.

Keywords: fault detection, gearbox, machine learning, wiener method

Procedia PDF Downloads 49

Authors: M. Maleczek, Leszek Bogdan, Kazimierz Drabczyk, Agnieszka Iwan

Abstract:

Crystalline silicon (Si) solar cells are the main product in the market among the various photovoltaic technologies concerning such advantages as: material richness, high carrier mobilities, broad spectral absorption range and established technology. However, photovoltaic technology on the stiff substrates are heavier, more fragile and less cost-effective than devices on the flexible substrates to be applied in special applications. The main goal of our work was to incorporate silicon solar cells into various fabric, without any change of the electrical and mechanical parameters of devices. This work is realized for the GEKON project (No. GEKON2/O4/268473/23/2016) sponsored by The National Centre for Research and Development and The National Fund for Environmental Protection and Water Management. In our work, the polyamide or polyester fabrics were used as a flexible substrate in the created devices. Applied fabrics differ in tensile and tear strength. All investigated polyamide fabrics are resistant to weathering and UV, while polyester ones is resistant to ozone, water and ageing. The examined fabrics are tight at 100 cm water per 2 hours. In our work, commercial silicon solar cells with the size 156 × 156 mm were cut into nine parts (called single solar cells) by diamond saw and laser. Gap and edge after cutting of solar cells were checked by transmission electron microscope (TEM) to study morphology and quality of the prepared single solar cells. Modules with the size of 160 × 70 cm (containing about 80 single solar cells) were created and investigated by electrical and mechanical methods. Weight of constructed module is about 1.9 kg. Three types of solar cell architectures such as: -fabric/EVA/Si solar cell/EVA/film for lamination, -backsheet PET/EVA/Si solar cell/EVA/film for lamination, -fabric/EVA/Si solar cell/EVA/tempered glass, were investigated taking into consideration type of fabric and lamination process together with the size of solar cells. In investigated devices EVA, it is ethylene-vinyl acetate, while PET - polyethylene terephthalate. Depend on the lamination process and compatibility of textile with solar cell an efficiency of investigated flexible silicon solar cells was in the range of 9.44-16.64 %. Multi folding and unfolding of flexible module has no impact on its efficiency as was detected by Instron equipment. Power (P) of constructed solar module is 30 W, while voltage about 36 V. Finally, solar panel contains five modules with the polyamide fabric and tempered glass will be produced commercially for different applications (dual use).

Keywords: flexible devices, mechanical properties, silicon solar cells, textiles

Procedia PDF Downloads 150

Authors: Sulemana Ibrahim

Abstract:

Food security remains a paramount global challenge, with vulnerable regions grappling with issues of hunger and malnutrition. This study embarks on a comprehensive exploration of data-driven strategies aimed at ameliorating food security in such regions. Our research employs a multifaceted approach, integrating data analytics to predict crop yields, optimizing supply chains, and enhancing food distribution networks. The study unfolds as a multi-dimensional analysis, commencing with the development of robust machine learning models harnessing remote sensing data, historical crop yield records, and meteorological data to foresee crop yields. These predictive models, underpinned by convolutional and recurrent neural networks, furnish critical insights into anticipated harvests, empowering proactive measures to confront food insecurity. Subsequently, the research scrutinizes supply chain optimization to address food security challenges, capitalizing on linear programming and network optimization techniques. These strategies intend to mitigate loss and wastage while streamlining the distribution of agricultural produce from field to fork. In conjunction, the study investigates food distribution networks with a particular focus on network efficiency, accessibility, and equitable food resource allocation. Network analysis tools, complemented by data-driven simulation methodologies, unveil opportunities for augmenting the efficacy of these critical lifelines. This study also considers the ethical implications and privacy concerns associated with the extensive use of data in the realm of food security. The proposed methodology outlines guidelines for responsible data acquisition, storage, and usage. The ultimate aspiration of this research is to forge a nexus between data science and food security policy, bestowing actionable insights to mitigate the ordeal of food insecurity. The holistic approach converging data-driven crop yield forecasts, optimized supply chains, and improved distribution networks aspire to revitalize food security in the most vulnerable regions, elevating the quality of life for millions worldwide.

Keywords: data-driven strategies, crop yield prediction, supply chain optimization, food distribution networks

Procedia PDF Downloads 34

Authors: Mohammadreza Hedyehzadeh, Mahdi Yousefi

Abstract:

Colon cancer is one of the most common cancer, which predicted to increase its prevalence due to the bad eating habits of peoples. Nowadays, due to the busyness of people, the use of fast foods is increasing, and therefore, diagnosis of this disease and its treatment are of particular importance. To determine the best treatment approach for each specific colon cancer patients, the oncologist should be known the stage of the tumor. The most common method to determine the tumor stage is TNM staging system. In this system, M indicates the presence of metastasis, N indicates the extent of spread to the lymph nodes, and T indicates the size of the tumor. It is clear that in order to determine all three of these parameters, an imaging method must be used, and the gold standard imaging protocols for this purpose are CT and PET/CT. In CT imaging, due to the use of X-rays, the risk of cancer and the absorbed dose of the patient is high, while in the PET/CT method, there is a lack of access to the device due to its high cost. Therefore, in this study, we aimed to estimate the tumor size and the extent of its spread to the lymph nodes using MR images. More than 1300 MR images collected from the TCIA portal, and in the first step (pre-processing), histogram equalization to improve image qualities and resizing to get the same image size was done. Two expert radiologists, which work more than 21 years on colon cancer cases, segmented the images and extracted the tumor region from the images. The next step is feature extraction from segmented images and then classify the data into three classes: T0N0، T3N1 و T3N2. In this article, the VGG-16 convolutional neural network has been used to perform both of the above-mentioned tasks, i.e., feature extraction and classification. This network has 13 convolution layers for feature extraction and three fully connected layers with the softmax activation function for classification. In order to validate the proposed method, the 10-fold cross validation method used in such a way that the data was randomly divided into three parts: training (70% of data), validation (10% of data) and the rest for testing. It is repeated 10 times, each time, the accuracy, sensitivity and specificity of the model are calculated and the average of ten repetitions is reported as the result. The accuracy, specificity and sensitivity of the proposed method for testing dataset was 89/09%, 95/8% and 96/4%. Compared to previous studies, using a safe imaging technique (MRI) and non-use of predefined hand-crafted imaging features to determine the stage of colon cancer patients are some of the study advantages.

Keywords: colon cancer, VGG-16, magnetic resonance imaging, tumor size, lymph node metastasis

Procedia PDF Downloads 33

Authors: Mustafa Alhamdi

Abstract:

Industrial application to classify gamma rays and neutron events is investigated in this study using deep machine learning. The identification using a convolutional neural network and recursive neural network showed a significant improvement in predication accuracy in a variety of applications. The ability to identify the isotope type and activity from spectral information depends on feature extraction methods, followed by classification. The features extracted from the spectrum profiles try to find patterns and relationships to present the actual spectrum energy in low dimensional space. Increasing the level of separation between classes in feature space improves the possibility to enhance classification accuracy. The nonlinear nature to extract features by neural network contains a variety of transformation and mathematical optimization, while principal component analysis depends on linear transformations to extract features and subsequently improve the classification accuracy. In this paper, the isotope spectrum information has been preprocessed by finding the frequencies components relative to time and using them as a training dataset. Fourier transform implementation to extract frequencies component has been optimized by a suitable windowing function. Training and validation samples of different isotope profiles interacted with CdTe crystal have been simulated using Geant4. The readout electronic noise has been simulated by optimizing the mean and variance of normal distribution. Ensemble learning by combing voting of many models managed to improve the classification accuracy of neural networks. The ability to discriminate gamma and neutron events in a single predication approach using deep machine learning has shown high accuracy using deep learning. The paper findings show the ability to improve the classification accuracy by applying the spectrogram preprocessing stage to the gamma and neutron spectrums of different isotopes. Tuning deep machine learning models by hyperparameter optimization of neural network models enhanced the separation in the latent space and provided the ability to extend the number of detected isotopes in the training database. Ensemble learning contributed significantly to improve the final prediction.

Keywords: machine learning, nuclear physics, Monte Carlo simulation, noise estimation, feature extraction, classification

Procedia PDF Downloads 121

Authors: Tingwei Shu, Dong Zhou, Chengjun Guo

Abstract:

Semantic communication is an emerging form of communication that realize intelligent communication by extracting semantic information of data at the source and transmitting it, and recovering the data at the receiving end. It can effectively solve the problem of data transmission under the situation of large data volume, low SNR and restricted bandwidth. With the development of Deep Learning, semantic communication further matures and is gradually applied in the fields of the Internet of Things, Uumanned Air Vehicle cluster communication, remote sensing scenarios, etc. We propose an improved semantic communication system for the situation where the data volume is huge and the spectrum resources are limited during the transmission of remote sensing images. At the transmitting, we need to extract the semantic information of remote sensing images, but there are some problems. The traditional semantic communication system based on Convolutional Neural Network cannot take into account the global semantic information and local semantic information of the image, which results in less-than-ideal image recovery at the receiving end. Therefore, we adopt the improved vision-Transformer-based structure as the semantic encoder instead of the mainstream one using CNN to extract the image semantic features. In this paper, we first perform pre-processing operations on remote sensing images to improve the resolution of the images in order to obtain images with more semantic information. We use wavelet transform to decompose the image into high-frequency and low-frequency components, perform bilinear interpolation on the high-frequency components and bicubic interpolation on the low-frequency components, and finally perform wavelet inverse transform to obtain the preprocessed image. We adopt the improved Vision-Transformer structure as the semantic coder to extract and transmit the semantic information of remote sensing images. The Vision-Transformer structure can better train the huge data volume and extract better image semantic features, and adopt the multi-layer self-attention mechanism to better capture the correlation between semantic features and reduce redundant features. Secondly, to improve the coding efficiency, we reduce the quadratic complexity of the self-attentive mechanism itself to linear so as to improve the image data processing speed of the model. We conducted experimental simulations on the RSOD dataset and compared the designed system with a semantic communication system based on CNN and image coding methods such as BGP and JPEG to verify that the method can effectively alleviate the problem of excessive data volume and improve the performance of image data communication.

Keywords: semantic communication, transformer, wavelet transform, data processing

Procedia PDF Downloads 46

Authors: Paolo Grasso, Sai Kalyan Yelike, Federico Benzi, Mathieu Le Cam

Abstract:

The cabin air quality (CAQ) in commercial aircraft is of prime interest, especially in the context of the COVID-19 pandemic. Current Environmental Control Systems (ECS) rely on a prescribed fresh airflow per passenger to dilute contaminants. An adaptive ECS strategy is proposed, leveraging air sensing and filtration technologies to ensure a better CAQ. This paper investigates the CAQ level achieved in commercial aircraft’s cabin during various flight scenarios. The modeling and simulation analysis is performed in a Modelica-based environment describing the dynamic behavior of the system. The model includes the following three main systems: cabin, recirculation loop and air-conditioning pack. The cabin model evaluates the thermo-hygrometric conditions and the air quality in the cabin depending on the number of passengers and crew members, the outdoor conditions and the conditions of the air supplied to the cabin. The recirculation loop includes models of the recirculation fan, ordinary and novel filtration technology, mixing chamber and outflow valve. The air-conditioning pack includes models of heat exchangers and turbomachinery needed to condition the hot pressurized air bled from the engine, as well as selected contaminants originated from the outside or bled from the engine. Different ventilation control strategies are modeled and simulated. Currently, a limited understanding of contaminant concentrations in the cabin and the lack of standardized and systematic methods to collect and record data constitute a challenge in establishing a causal relationship between CAQ and passengers' comfort. As a result, contaminants are neither measured nor filtered during flight, and the current sub-optimal way to avoid their accumulation is their dilution with the fresh air flow. However, the use of a prescribed amount of fresh air comes with a cost, making the ECS the most energy-demanding non-propulsive system within an aircraft. In such a context, this study shows that an ECS based on a reduced and adaptive fresh air flow, and relying on air sensing and filtration technologies, provides promising results in terms of CAQ control. The comparative simulation results demonstrate that the proposed adaptive ECS brings substantial improvements to the CAQ in terms of both controlling the asymptotic values of the concentration of the contaminant and in mitigating hazardous scenarios, such as fume events. Original architectures allowing for adaptive control of the inlet air flow rate based on monitored CAQ will change the requirements for filtration systems and redefine the ECS operation.

Keywords: cabin air quality, commercial aircraft, environmental control system, ventilation

Procedia PDF Downloads 72

Authors: Suruchi Saini, Hitender Kumar Tyagi

Abstract:

With the rapid progress of the nanotechnology industry, it is becoming increasingly important to have compact semiconductor devices to function and offer the best results at various technology nodes. While performing the scaling of the device, several short-channel effects occur. To minimize these scaling limitations, some device architectures have been developed in the semiconductor industry. FinFET is one of the most promising structures. Also, the double-gate 2D Fin field effect transistor has the benefit of suppressing short channel effects (SCE) and functioning well for less than 14 nm technology nodes. In the present research, the MuGFET simulation tool is used to analyze and explain the electrical behaviour of a double-gate 2D Fin field effect transistor. The drift-diffusion and Poisson equations are solved self-consistently. Various models, such as Fermi-Dirac distribution, bandgap narrowing, carrier scattering, and concentration-dependent mobility models, are used for device simulation. The transfer and output characteristics of the double-gate 2D Fin field effect transistor are determined at 10 nm technology node. The performance parameters are extracted in terms of threshold voltage, trans-conductance, leakage current and current on-off ratio. In this paper, the device performance is analyzed at different structure parameters. The utilization of the Id-Vg curve is a robust technique that holds significant importance in the modeling of transistors, circuit design, optimization of performance, and quality control in electronic devices and integrated circuits for comprehending field-effect transistors. The FinFET structure is optimized to increase the current on-off ratio and transconductance. Through this analysis, the impact of different channel widths, source and drain lengths on the Id-Vg and transconductance is examined. Device performance was affected by the difficulty of maintaining effective gate control over the channel at decreasing feature sizes. For every set of simulations, the device's features are simulated at two different drain voltages, 50 mV and 0.7 V. In low-power and precision applications, the off-state current is a significant factor to consider. Therefore, it is crucial to minimize the off-state current to maximize circuit performance and efficiency. The findings demonstrate that the performance of the current on-off ratio is maximum with the channel width of 3 nm for a gate length of 10 nm, but there is no significant effect of source and drain length on the current on-off ratio. The transconductance value plays a pivotal role in various electronic applications and should be considered carefully. In this research, it is also concluded that the transconductance value of 340 S/m is achieved with the fin width of 3 nm at a gate length of 10 nm and 2380 S/m for the source and drain extension length of 5 nm, respectively.

Keywords: current on-off ratio, FinFET, short-channel effects, transconductance

Procedia PDF Downloads 39

Authors: Edgar Santoyo, Daniel Perez-Zarate, Agustin Acevedo, Lorena Diaz-Gonzalez, Mirna Guevara

Abstract:

Four new gas geothermometers have been derived from a multivariate geo chemometric analysis of a geothermal fluid chemistry database, two of which use the natural logarithm of CO₂ and H2S concentrations (mmol/mol), respectively, and the other two use the natural logarithm of the H₂S/H₂ and CO₂/H₂ ratios. As a strict compilation criterion, the database was created with gas-phase composition of fluids and bottomhole temperatures (BHTM) measured in producing wells. The calibration of the geothermometers was based on the geochemical relationship existing between the gas-phase composition of well discharges and the equilibrium temperatures measured at bottomhole conditions. Multivariate statistical analysis together with the use of artificial neural networks (ANN) was successfully applied for correlating the gas-phase compositions and the BHTM. The predicted or simulated bottomhole temperatures (BHTANN), defined as output neurons or simulation targets, were statistically compared with measured temperatures (BHTM). The coefficients of the new geothermometers were obtained from an optimized self-adjusting training algorithm applied to approximately 2,080 ANN architectures with 15,000 simulation iterations each one. The self-adjusting training algorithm used the well-known Levenberg-Marquardt model, which was used to calculate: (i) the number of neurons of the hidden layer; (ii) the training factor and the training patterns of the ANN; (iii) the linear correlation coefficient, R; (iv) the synaptic weighting coefficients; and (v) the statistical parameter, Root Mean Squared Error (RMSE) to evaluate the prediction performance between the BHTM and the simulated BHTANN. The prediction performance of the new gas geothermometers together with those predictions inferred from sixteen well-known gas geothermometers (previously developed) was statistically evaluated by using an external database for avoiding a bias problem. Statistical evaluation was performed through the analysis of the lowest RMSE values computed among the predictions of all the gas geothermometers. The new gas geothermometers developed in this work have been successfully used for predicting subsurface temperatures in high-temperature geothermal systems of Mexico (e.g., Los Azufres, Mich., Los Humeros, Pue., and Cerro Prieto, B.C.) as well as in a blind geothermal system (known as Acoculco, Puebla). The last results of the gas geothermometers (inferred from gas-phase compositions of soil-gas bubble emissions) compare well with the temperature measured in two wells of the blind geothermal system of Acoculco, Puebla (México). Details of this new development are outlined in the present research work. Acknowledgements: The authors acknowledge the funding received from CeMIE-Geo P09 project (SENER-CONACyT).

Keywords: artificial intelligence, gas geochemistry, geochemometrics, geothermal energy

Procedia PDF Downloads 307

Authors: Debabrata Adhikari, Mikhail Matveev, Louise Brown, Andy Long, Jan Kočí

Abstract:

A good understanding of mesoscale permeability of complex architectures in fibrous porous preforms is of particular interest in order to achieve efficient and cost-effective resin impregnation of liquid composite molding (LCM). Fabrics used in structural reinforcements are typically woven or stitched. However, 3D fabric reinforcement is of particular interest because of the versatility in the weaving pattern with the binder yarn and in-plain yarn arrangements to manufacture thick composite parts, overcome the limitation in delamination, improve toughness etc. To predict the permeability based on the available pore spaces between the inter yarn spaces, unit cell-based computational fluid dynamics models have been using the Stokes Darcy model. Typically, the preform consists of an arrangement of yarns with spacing in the order of mm, wherein each yarn consists of thousands of filaments with spacing in the order of μm. The fluid flow during infusion exchanges the mass between the intra and inter yarn channels, meaning there is no dead-end of flow between the mesopore in the inter yarn space and the micropore in the yarn. Several studies have employed the Brinkman equation to take into account the flow through dual-scale porosity reinforcement to estimate their permeability. Furthermore, to reduce the computational effort of dual scale flow, scale separation criteria based on the ratio between yarn permeability to the yarn spacing was also proposed to quantify the dual scale and negligible micro-scale flow regime for the prediction of mesoscale permeability. In the present work, the key parameter to identify the influence of intra yarn permeability on the mesoscale permeability has been investigated with the systematic study of weft and warp yarn spacing on the plane weave as well as the position of binder yarn and number of in-plane yarn layers on 3D weave fabric. The permeability tensor has been estimated using an OpenFOAM-based model for the various weave pattern with idealized geometry of yarn implemented using open-source software TexGen. Additionally, scale separation criterion has been established based on the various configuration of yarn permeability for the 3D fabric with both the isotropic and anisotropic yarn from Gebart’s model. It was observed that the variation of mesoscale permeability Kxx within 30% when the isotropic porous yarn is considered for a 3D fabric with binder yarn. Furthermore, the permeability model developed in this study will be used for multi-objective optimizations of the preform mesoscale geometry in terms of yarn spacing, binder pattern, and a number of layers with an aim to obtain improved permeability and reduced void content during the LCM process.

Keywords: permeability, 3D fabric, dual-scale flow, liquid composite molding

Procedia PDF Downloads 71

Authors: Jeremy Jenrette, Z. Y. C. Liu, Pranav Chimote, Edward Fox, Trevor Hastie, Francesco Ferretti

Abstract:

Suitable shark conservation depends on well-informed population assessments. Direct methods such as scientific surveys and fisheries monitoring are adequate for defining population statuses, but species-specific indices of abundance and distribution coming from these sources are rare for most shark species. We can rapidly fill these information gaps by boosting media-based remote monitoring efforts with machine learning and automation. We created a database of shark images by sourcing 24,546 images covering 219 species of sharks from the web application spark pulse and the social network Instagram. We used object detection to extract shark features and inflate this database to 53,345 images. We packaged object-detection and image classification models into a Shark Detector bundle. We developed the Shark Detector to recognize and classify sharks from videos and images using transfer learning and convolutional neural networks (CNNs). We applied these models to common data-generation approaches of sharks: boosting training datasets, processing baited remote camera footage and online videos, and data-mining Instagram. We examined the accuracy of each model and tested genus and species prediction correctness as a result of training data quantity. The Shark Detector located sharks in baited remote footage and YouTube videos with an average accuracy of 89\%, and classified located subjects to the species level with 69\% accuracy (n =\ eight species). The Shark Detector sorted heterogeneous datasets of images sourced from Instagram with 91\% accuracy and classified species with 70\% accuracy (n =\ 17 species). Data-mining Instagram can inflate training datasets and increase the Shark Detector’s accuracy as well as facilitate archiving of historical and novel shark observations. Base accuracy of genus prediction was 68\% across 25 genera. The average base accuracy of species prediction within each genus class was 85\%. The Shark Detector can classify 45 species. All data-generation methods were processed without manual interaction. As media-based remote monitoring strives to dominate methods for observing sharks in nature, we developed an open-source Shark Detector to facilitate common identification applications. Prediction accuracy of the software pipeline increases as more images are added to the training dataset. We provide public access to the software on our GitHub page.

Keywords: classification, data mining, Instagram, remote monitoring, sharks

Procedia PDF Downloads 89

Authors: Shubham Gogri, Lucia Florescu

Abstract:

Medical diagnosis and radiotherapy treatment planning using Computed Tomography (CT) rely on the quantitative accuracy and quality of the CT images. At the same time, requirements for CT imaging include reducing the radiation dose exposure to patients and minimizing scanning time. A solution to this is the sparse-view CT technique, based on a reduced number of projection views. This, however, introduces a new problem— the incomplete projection data results in lower quality of the reconstructed images. To tackle this issue, deep learning methods have been applied to enhance the quality of the sparse-view CT images. A first approach involved employing Mir-Net, a dedicated deep neural network designed for image enhancement. This showed promise, utilizing an intricate architecture comprising encoder and decoder networks, along with the incorporation of the Charbonnier Loss. However, this approach was computationally demanding. Subsequently, a specialized Generative Adversarial Network (GAN) architecture, rooted in the Pix2Pix framework, was implemented. This GAN framework involves a U-Net-based Generator and a Discriminator based on Convolutional Neural Networks. To bolster the GAN's performance, both Charbonnier and Wasserstein loss functions were introduced, collectively focusing on capturing minute details while ensuring training stability. The integration of the perceptual loss, calculated based on feature vectors extracted from the VGG16 network pretrained on the ImageNet dataset, further enhanced the network's ability to synthesize relevant images. A series of comprehensive experiments with clinical CT data were conducted, exploring various GAN loss functions, including Wasserstein, Charbonnier, and perceptual loss. The outcomes demonstrated significant image quality improvements, confirmed through pertinent metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) between the corrected images and the ground truth. Furthermore, learning curves and qualitative comparisons added evidence of the enhanced image quality and the network's increased stability, while preserving pixel value intensity. The experiments underscored the potential of deep learning frameworks in enhancing the visual interpretation of CT scans, achieving outcomes with SSIM values close to one and PSNR values reaching up to 76.

Keywords: generative adversarial networks, sparse view computed tomography, CT image correction, Mir-Net

Procedia PDF Downloads 114

Authors: Efrain Rodriguez, Cristhian Riano, Alberto Alvares

Abstract:

In recent years, robots manipulators with parallel architectures are used in additive manufacturing processes – 3D printing. These robots have advantages such as speed and lightness that make them suitable to help with the efficiency and productivity of these processes. Consequently, the interest for the development of parallel robots for additive manufacturing applications has increased. This article deals with the conceptual design and dimensional synthesis of the linear delta robot for additive manufacturing. Firstly, a methodology based on structured processes for the development of products through the phases of informational design, conceptual design and detailed design is adopted: a) In the informational design phase the Mudge diagram and the QFD matrix are used to aid a set of technical requirements, to define the form, functions and features of the robot. b) In the conceptual design phase, the functional modeling of the system through of an IDEF0 diagram is performed, and the solution principles for the requirements are formulated using a morphological matrix. This phase includes the description of the mechanical, electro-electronic and computational subsystems that constitute the general architecture of the robot. c) In the detailed design phase, a digital model of the robot is drawn on CAD software. A list of commercial and manufactured parts is detailed. Tolerances and adjustments are defined for some parts of the robot structure. The necessary manufacturing processes and tools are also listed, including: milling, turning and 3D printing. Secondly, a dimensional synthesis method applied on design of the linear delta robot is presented. One of the most important key factors in the design of a parallel robot is the useful workspace, which strongly depends on the joint space, the dimensions of the mechanism bodies and the possible interferences between these bodies. The objective function is based on the verification of the kinematic model for a prescribed cylindrical workspace, considering geometric constraints that possibly lead to singularities of the mechanism. The aim is to determine the minimum dimensional parameters of the mechanism bodies for the proposed workspace. A method based on genetic algorithms was used to solve this problem. The method uses a cloud of points with the cylindrical shape of the workspace and checks the kinematic model for each of the points within the cloud. The evolution of the population (point cloud) provides the optimal parameters for the design of the delta robot. The development process of the linear delta robot with optimal dimensions for additive manufacture is presented. The dimensional synthesis enabled to design the mechanism of the delta robot in function of the prescribed workspace. Finally, the implementation of the robotic platform developed based on a linear delta robot in an additive manufacturing application using the Fused Deposition Modeling (FDM) technique is presented.

Keywords: additive manufacturing, delta parallel robot, dimensional synthesis, genetic algorithms

Procedia PDF Downloads 164

Authors: Bander Alzahrani, Mohammed Alreshoodi

Abstract:

Information-centric networking (ICN) using architectures such as the Publish-Subscribe Internet Technology (PURSUIT) has been proposed as a new networking model that aims at replacing the current used end-centric networking model of the Internet. This emerged model focuses on what is being exchanged rather than which network entities are exchanging information, which gives the control plane functions such as routing and host location the ability to be specified according to the content items. The forwarding plane of the PURSUIT ICN architecture uses a simple and light mechanism based on Bloom filter technologies to forward the packets. Although this forwarding scheme solve many problems of the today’s Internet such as the growth of the routing table and the scalability issues, it is vulnerable to brute force attacks which are starting point to distributed- denial-of-service (DDoS) attacks. In this work, we design and analyze a novel source-routing and information delivery technique that keeps the simplicity of using Bloom filter-based forwarding while being able to deter different attacks such as denial of service attacks at the ingress of the network. To achieve this, special forwarding nodes called Edge-FW are directly attached to end user nodes and used to perform a security test for malicious injected random packets at the ingress of the path to prevent any possible attack brute force attacks at early stage. In this technique, a core entity of the PURSUIT ICN architecture called topology manager, that is responsible for finding shortest path and creating a forwarding identifiers (FId), uses a cryptographically secure hash function to create a 64-bit hash, h, over the formed FId for authentication purpose to be included in the packet. Our proposal restricts the attacker from injecting packets carrying random FIds with a high amount of filling factor ρ, by optimizing and reducing the maximum allowed filling factor ρm in the network. We optimize the FId to the minimum possible filling factor where ρ ≤ ρm, while it supports longer delivery trees, so the network scalability is not affected by the chosen ρm. With this scheme, the filling factor of any legitimate FId never exceeds the ρm while the filling factor of illegitimate FIds cannot exceed the chosen small value of ρm. Therefore, injecting a packet containing an FId with a large value of filling factor, to achieve higher attack probability, is not possible anymore. The preliminary analysis of this proposal indicates that with the designed scheme, the forwarding function can detect and prevent malicious activities such DDoS attacks at early stage and with very high probability.

Keywords: forwarding identifier, filling factor, information centric network, topology manager

Procedia PDF Downloads 130

Authors: Yuta Nakano, Kozo Kajiwara, Atsushi Hori, Takeshi Fujita

Abstract:

In association with the wave of electric vehicles (EV), low energy consumption systems have become more and more important. One of the key technologies to realize low energy consumption is a dynamic vision sensor (DVS), or we can call it an event sensor, neuromorphic vision sensor and so on. This sensor has several features, such as high temporal resolution, which can achieve 1 Mframe/s, and a high dynamic range (120 DB). However, the point that can contribute to low energy consumption the most is its sparsity; to be more specific, this sensor only captures the pixels that have intensity change. In other words, there is no signal in the area that does not have any intensity change. That is to say, this sensor is more energy efficient than conventional sensors such as RGB cameras because we can remove redundant data. On the other side of the advantages, it is difficult to handle the data because the data format is completely different from RGB image; for example, acquired signals are asynchronous and sparse, and each signal is composed of x-y coordinate, polarity (two values: +1 or -1) and time stamp, it does not include intensity such as RGB values. Therefore, as we cannot use existing algorithms straightforwardly, we have to design a new processing algorithm to cope with DVS data. In order to solve difficulties caused by data format differences, most of the prior arts make a frame data and feed it to deep learning such as Convolutional Neural Networks (CNN) for object detection and recognition purposes. However, even though we can feed the data, it is still difficult to achieve good performance due to a lack of intensity information. Although polarity is often used as intensity instead of RGB pixel value, it is apparent that polarity information is not rich enough. Considering this context, we proposed to use the timestamp information as a data representation that is fed to deep learning. Concretely, at first, we also make frame data divided by a certain time period, then give intensity value in response to the timestamp in each frame; for example, a high value is given on a recent signal. We expected that this data representation could capture the features, especially of moving objects, because timestamp represents the movement direction and speed. By using this proposal method, we made our own dataset by DVS fixed on a parked car to develop an application for a surveillance system that can detect persons around the car. We think DVS is one of the ideal sensors for surveillance purposes because this sensor can run for a long time with low energy consumption in a NOT dynamic situation. For comparison purposes, we reproduced state of the art method as a benchmark, which makes frames the same as us and feeds polarity information to CNN. Then, we measured the object detection performances of the benchmark and ours on the same dataset. As a result, our method achieved a maximum of 7 points greater than the benchmark in the F1 score.

Keywords: event camera, dynamic vision sensor, deep learning, data representation, object recognition, low energy consumption

Procedia PDF Downloads 62

Authors: Dana Pugach, Michal Tamir

Abstract:

Most criminal cases end with a plea agreement, an issue whose many aspects have been discussed extensively in legal literature. One important feature, however, has gained little notice, and that is crime victims’ place in plea agreements following the federal Crime Victims Rights Act of 2004. This law has provided victims some meaningful and potentially revolutionary rights, including the right to be heard in the proceeding and a right to appeal against a decision made while ignoring the victim’s rights. While victims’ rights literature has always emphasized the importance of such right, references to this provision in the general literature about plea agreements are sparse, if existing at all. Furthermore, there are a few cases only mentioning this right. This article purports to bridge between these two bodies of legal thinking – the vast literature concerning plea agreements and victims’ rights research– by using behavioral economics. The article will, firstly, trace the possible structural reasons for the failure of this right to be materialized. Relevant incentives of all actors involved will be identified as well as their inherent consequential processes that lead to the victims’ rights malfunction. Secondly, the article will use nudge theory in order to suggest solutions that will enhance incentives for the repeat players in the system (prosecution, judges, defense attorneys) and lead to the strengthening of weaker group’s interests – the crime victims. Behavioral psychology literature recognizes that the framework in which an individual confronts a decision can significantly influence his decision. Richard Thaler and Cass Sunstein developed the idea of ‘choice architecture’ - ‘the context in which people make decisions’ - which can be manipulated to make particular decisions more likely. Choice architectures can be changed by adjusting ‘nudges,’ influential factors that help shape human behavior, without negating their free choice. The nudges require decision makers to make choices instead of providing a familiar default option. In accordance with this theory, we suggest a rule, whereby a judge should inquire the victim’s view prior to accepting the plea. This suggestion leaves the judge’s discretion intact; while at the same time nudges her not to go directly to the default decision, i.e. automatically accepting the plea. Creating nudges that force actors to make choices is particularly significant when an actor intends to deviate from routine behaviors but experiences significant time constraints, as in the case of judges and plea bargains. The article finally recognizes some far reaching possible results of the suggestion. These include meaningful changes to the earlier stages of criminal process even before reaching court, in line with the current criticism of the plea agreements machinery.

Keywords: plea agreements, victims' rights, nudge theory, criminal justice

Procedia PDF Downloads 300

Authors: Rafael do Amaral Teodoro, Leandro Augusto da Silva

Abstract:

Originated from graphite, graphene is a two-dimensional (2D) material that promises to revolutionize technology in many different areas, such as energy, telecommunications, civil construction, aviation, textile, and medicine. This is possible because its structure, formed by carbon bonds, provides desirable optical, thermal, and mechanical characteristics that are interesting to multiple areas of the market. Thus, several research and development centers are studying different manufacturing methods and material applications of graphene, which are often compromised by the scarcity of more agile and accurate methodologies to characterize the material – that is to determine its composition, shape, size, and the number of layers and crystals. To engage in this search, this study proposes a computational methodology that applies deep learning to identify graphene oxide crystals in order to characterize samples by crystal sizes. To achieve this, a fully convolutional neural network called U-net has been trained to segment SEM graphene oxide images. The segmentation generated by the U-net is fine-tuned with a standard deviation technique by classes, which allows crystals to be distinguished with different labels through an object delimitation algorithm. As a next step, the characteristics of the position, area, perimeter, and lateral measures of each detected crystal are extracted from the images. This information generates a database with the dimensions of the crystals that compose the samples. Finally, graphs are automatically created showing the frequency distributions by area size and perimeter of the crystals. This methodological process resulted in a high capacity of segmentation of graphene oxide crystals, presenting accuracy and F-score equal to 95% and 94%, respectively, over the test set. Such performance demonstrates a high generalization capacity of the method in crystal segmentation, since its performance considers significant changes in image extraction quality. The measurement of non-overlapping crystals presented an average error of 6% for the different measurement metrics, thus suggesting that the model provides a high-performance measurement for non-overlapping segmentations. For overlapping crystals, however, a limitation of the model was identified. To overcome this limitation, it is important to ensure that the samples to be analyzed are properly prepared. This will minimize crystal overlap in the SEM image acquisition and guarantee a lower error in the measurements without greater efforts for data handling. All in all, the method developed is a time optimizer with a high measurement value, considering that it is capable of measuring hundreds of graphene oxide crystals in seconds, saving weeks of manual work.

Keywords: characterization, graphene oxide, nanomaterials, U-net, deep learning

Procedia PDF Downloads 134

Authors: J. J. Rickard, A. Belli, P. Goldberg Oppenheimer

Abstract:

Medical diagnostics, environmental monitoring, homeland security and forensics increasingly demand specific and field-deployable analytical technologies for quick point-of-care diagnostics. Although technological advancements have made optical methods well-suited for miniaturization, a highly-sensitive detection technique for minute sample volumes is required. Raman spectroscopy is a well-known analytical tool, but has very weak signals and hence is unsuitable for trace level analysis. Enhancement via localized optical fields (surface plasmons resonances) on nanoscale metallic materials generates huge signals in surface-enhanced Raman scattering (SERS), enabling single molecule detection. This enhancement can be tuned by manipulation of the surface roughness and architecture at the sub-micron level. Nevertheless, the development and application of SERS has been inhibited by the irreproducibility and complexity of fabrication routes. The ability to generate straightforward, cost-effective, multiplex-able and addressable SERS substrates with high enhancements is of profound interest for SERS-based sensing devices. While most SERS substrates are manufactured by conventional lithographic methods, the development of a cost-effective approach to create nanostructured surfaces is a much sought-after goal in the SERS community. Here, a method is established to create controlled, self-organized, hierarchical nanostructures using electrohydrodynamic (HEHD) instabilities. The created structures are readily fine-tuned, which is an important requirement for optimizing SERS to obtain the highest enhancements. HEHD pattern formation enables the fabrication of multiscale 3D structured arrays as SERS-active platforms. Importantly, each of the HEHD-patterned individual structural units yield a considerable SERS enhancement. This enables each single unit to function as an isolated sensor. Each of the formed structures can be effectively tuned and tailored to provide high SERS enhancement, while arising from different HEHD morphologies. The HEHD fabrication of sub-micrometer architectures is straightforward and robust, providing an elegant route for high-throughput biological and chemical sensing. The superior detection properties and the ability to fabricate SERS substrates on the miniaturized scale, will facilitate the development of advanced and novel opto-fluidic devices, such as portable detection systems, and will offer numerous applications in biomedical diagnostics, forensics, ecological warfare and homeland security.

Keywords: hierarchical electrohydrodynamic patterning, medical diagnostics, point-of care devices, SERS

Procedia PDF Downloads 317

Authors: Gaurangi Saxena, Ravindra Saxena

Abstract:

Lately, cloud computing is used to enhance the ability to attain corporate goals more effectively and efficiently at lower cost. This new computing paradigm “The Cloud Computing” has emerged as a powerful tool for optimum utilization of resources and gaining competitiveness through cost reduction and achieving business goals with greater flexibility. Realizing the importance of this new technique, most of the well known companies in computer industry like Microsoft, IBM, Google and Apple are spending millions of dollars in researching cloud computing and investigating the possibility of producing interface hardware for cloud computing systems. It is believed that by using the right middleware, a cloud computing system can execute all the programs a normal computer could run. Potentially, everything from most simple generic word processing software to highly specialized and customized programs designed for specific company could work successfully on a cloud computing system. A Cloud is a pool of virtualized computer resources. Clouds are not limited to grid environments, but also support “interactive user-facing applications” such as web applications and three-tier architectures. Cloud Computing is not a fundamentally new paradigm. It draws on existing technologies and approaches, such as utility Computing, Software-as-a-service, distributed computing, and centralized data centers. Some companies rent physical space to store servers and databases because they don’t have it available on site. Cloud computing gives these companies the option of storing data on someone else’s hardware, removing the need for physical space on the front end. Prominent service providers like Amazon, Google, SUN, IBM, Oracle, Salesforce etc. are extending computing infrastructures and platforms as a core for providing top-level services for computation, storage, database and applications. Application services could be email, office applications, finance, video, audio and data processing. By using cloud computing system a company can improve its customer relationship management. A CRM cloud computing system may be highly useful in delivering a sales team a blend of unique functionalities to improve agent/customer interactions. This paper attempts to first define the cloud computing as a tool for running business activities more effectively and efficiently at a lower cost; and then it distinguishes cloud computing with grid computing. Based on exhaustive literature review, authors discuss application of cloud computing in different disciplines of management especially in the field of marketing with special reference to use of cloud computing in CRM. Study concludes that CRM cloud computing platform helps a company track any data, such as orders, discounts, references, competitors and many more. By using CRM cloud computing, companies can improve its customer interactions and by serving them more efficiently that too at a lower cost can help gaining competitive advantage.

Keywords: cloud computing, competitive advantage, customer relationship management, grid computing

Procedia PDF Downloads 278

Authors: Yasmine Salehy, Yann Leroy, Francois Cluzel, Hong-Minh Hoang, Laurence Fournaison, Anthony Delahaye, Bernard Yannou

Abstract:

Taking into consideration all the life cycle of a product is now an important step in the eco-design of a product or a technology. Life cycle assessment (LCA) is a standard tool to evaluate the environmental impacts of a system or a process. Despite the improvement in refrigerant regulation through protocols, the environmental damage of refrigeration systems remains important and needs to be improved. In this paper, the environmental impacts of refrigeration systems in a typical supermarket are compared using the LCA methodology under different conditions. The system is used to provide cold at two levels of temperature: medium and low temperature during a life period of 15 years. The most commonly used architectures of supermarket cold production systems are investigated: centralized direct expansion systems and indirect systems using a secondary loop to transport the cold. The variation of power needed during seasonal changes and during the daily opening/closure periods of the supermarket are considered. R134a as the primary refrigerant fluid and two types of secondary fluids are considered. The composition of each system and the leakage rate of the refrigerant through its life cycle are taken from the literature and industrial data. Twelve scenarios are examined. They are based on the variation of three parameters, 1. location: France (Paris), Spain (Toledo) and Sweden (Stockholm), 2. different sources of electric consumption: photovoltaic panels and low voltage electric network and 3. architecture: direct and indirect refrigeration systems. OpenLCA, SimaPro softwares, and different impact assessment methods were compared; CML method is used to evaluate the midpoint environmental indicators. This study highlights the significant contribution of electric consumption in environmental damages compared to the impacts of refrigerant leakage. The secondary loop allows lowering the refrigerant amount in the primary loop which results in a decrease in the climate change indicators compared to the centralized direct systems. However, an exhaustive cost evaluation (CAPEX and OPEX) of both systems shows more important costs related to the indirect systems. A significant difference between the countries has been noticed, mostly due to the difference in electric production. In Spain, using photovoltaic panels helps to reduce efficiently the environmental impacts and the related costs. This scenario is the best alternative compared to the other scenarios. Sweden is a country with less environmental impacts. For both France and Sweden, the use of photovoltaic panels does not bring a significant difference, due to a less sunlight exposition than in Spain. Alternative solutions exist to reduce the impact of refrigerating systems, and a brief introduction is presented.

Keywords: eco-design, industrial engineering, LCA, refrigeration system

Procedia PDF Downloads 141

Authors: Nguyen J. M., Lucas G., Ruan S., Digonnet H., Antonioli D.

Abstract:

Introduction: Deep Learning (DL) is a very powerful tool for analyzing image data. But for tabular data, it cannot compete with machine learning methods like XGBoost. The research question becomes: can tabular data be transformed into images that can be analyzed by simple CNNs (Convolutional Neuron Networks)? Will DL be the absolute tool for data classification? All current solutions consist in repositioning the variables in a 2x2 matrix using their correlation proximity. In doing so, it obtains an image whose pixels are the variables. We implement a technology, DeepNIC, that offers the possibility of obtaining an image for each variable, which can be analyzed by simple CNNs. Material and method: The 'ROP' (Regression OPtimized) model is a binary and atypical decision tree whose nodes are managed by a new artificial neuron, the Neurop. By positioning an artificial neuron in each node of the decision trees, it is possible to make an adjustment on a theoretically infinite number of variables at each node. From this new decision tree whose nodes are artificial neurons, we created the concept of a 'Random Forest of Perfect Trees' (RFPT), which disobeys Breiman's concepts by assembling very large numbers of small trees with no classification errors. From the results of the RFPT, we developed a family of 10 statistical information criteria, Nguyen Information Criterion (NICs), which evaluates in 3 dimensions the predictive quality of a variable: Performance, Complexity and Multiplicity of solution. A NIC is a probability that can be transformed into a grey level. The value of a NIC depends essentially on 2 super parameters used in Neurops. By varying these 2 super parameters, we obtain a 2x2 matrix of probabilities for each NIC. We can combine these 10 NICs with the functions AND, OR, and XOR. The total number of combinations is greater than 100,000. In total, we obtain for each variable an image of at least 1166x1167 pixels. The intensity of the pixels is proportional to the probability of the associated NIC. The color depends on the associated NIC. This image actually contains considerable information about the ability of the variable to make the prediction of Y, depending on the presence or absence of other variables. A basic CNNs model was trained for supervised classification. Results: The first results are impressive. Using the GSE22513 public data (Omic data set of markers of Taxane Sensitivity in Breast Cancer), DEEPNic outperformed other statistical methods, including XGBoost. We still need to generalize the comparison on several databases. Conclusion: The ability to transform any tabular variable into an image offers the possibility of merging image and tabular information in the same format. This opens up great perspectives in the analysis of metadata.

Keywords: tabular data, CNNs, NICs, DeepNICs, random forest of perfect trees, classification

Procedia PDF Downloads 72

Authors: Barbara Urasinska-Wojcik, Neil Chilton, Peter Todd, Christopher Elsworthy, Gregory J. Gibbons

Abstract:

The recent increase in the application of Additive Manufacturing (AM) of products has resulted in new demands on capability. The ability to integrate both form and function within printed objects is the next frontier in the 3D printing area. To move beyond prototyping into low volume production, we demonstrate a UK-designed and built AM hybrid system that combines polymer based structural deposition with digital deposition of electrically conductive elements. This hybrid manufacturing system is based on a multi-planar build approach to improve on many of the limitations associated with AM, such as poor surface finish, low geometric tolerance, and poor robustness. Specifically, the approach involves a multi-planar Material Extrusion (ME) process in which separated build stations with up to 5 axes of motion replace traditional horizontally-sliced layer modeling. The construction of multi-material architectures also involved using multiple print systems in order to combine both ME and digital deposition of conductive material. To demonstrate multi-material 3D printing, three thermoplastics, acrylonitrile butadiene styrene (ABS), polyamide 6,6/6 copolymers (CoPA) and polyamide 12 (PA) were used to print specimens, on top of which our high viscosity Ag-particulate ink was printed in a non-contact process, during which drop characteristics such as shape, velocity, and volume were assessed using a drop watching system. Spectroscopic analysis of these 3D printed materials in the IR region helped to determine the optimum in-situ curing system for implementation into the AM system to achieve improved adhesion and surface refinement. Thermal Analyses were performed to determine the printed materials glass transition temperature (Tg), stability and degradation behavior to find the optimum annealing conditions post printing. Electrical analysis of printed conductive tracks on polymer surfaces during mechanical testing (static tensile and 3-point bending and dynamic fatigue) was performed to assess the robustness of the electrical circuits. The tracks on CoPA, ABS, and PA exhibited low electrical resistance, and in case of PA resistance values of tracks remained unchanged across hundreds of repeated tensile cycles up to 0.5% strain amplitude. Our developed AM printer has the ability to fabricate fully functional objects in one build, including complex electronics. It enables product designers and manufacturers to produce functional saleable electronic products from a small format modular platform. It will make 3D printing better, faster and stronger.

Keywords: additive manufacturing, conductive tracks, hybrid 3D printer, integrated manufacture

Procedia PDF Downloads 140

Authors: Manaranjan Pradhan, Shailaja Grover, U. Dinesh Kumar

Abstract:

Farming community in India, as well as other parts of the world, is one of the highly stressed communities due to reasons such as increasing input costs (cost of seeds, fertilizers, pesticide), droughts, reduced revenue leading to farmer suicides. Lack of integrated farm advisory system in India adds to the farmers problems. Farmers need right information during the early stages of crop’s lifecycle to prevent damage and loss in revenue. In this paper, we use deep learning techniques to develop an early warning system for detection of crop diseases using images taken by farmers using their smart phone. The research work leads to building a smart assistant using analytics and big data which could help the farmers with early diagnosis of the crop diseases and corrective actions. The classical approach for crop disease management has been to identify diseases at crop level. Recently, ImageNet Classification using the convolutional neural network (CNN) has been successfully used to identify diseases at individual plant level. Our model uses convolution filters, max pooling, dense layers and dropouts (to avoid overfitting). The models are built for binary classification (healthy or not healthy) and multi class classification (identifying which disease). Transfer learning is used to modify the weights of parameters learnt through ImageNet dataset and apply them on crop diseases, which reduces number of epochs to learn. One shot learning is used to learn from very few images, while data augmentation techniques are used to improve accuracy with images taken from farms by using techniques such as rotation, zoom, shift and blurred images. Models built using combination of these techniques are more robust for deploying in the real world. Our model is validated using tomato crop. In India, tomato is affected by 10 different diseases. Our model achieves an accuracy of more than 95% in correctly classifying the diseases. The main contribution of our research is to create a personal assistant for farmers for managing plant disease, although the model was validated using tomato crop, it can be easily extended to other crops. The advancement of technology in computing and availability of large data has made possible the success of deep learning applications in computer vision, natural language processing, image recognition, etc. With these robust models and huge smartphone penetration, feasibility of implementation of these models is high resulting in timely advise to the farmers and thus increasing the farmers' income and reducing the input costs.

Keywords: analytics in agriculture, CNN, crop disease detection, data augmentation, image recognition, one shot learning, transfer learning

Procedia PDF Downloads 95

Authors: Burak Oksuzoglu, Savas Yildirim, Ferhat Kutlu

Abstract:

Opinion mining (OM) is one of the natural language processing (NLP) problems to determine the polarity of opinions, mostly represented on a positive-neutral-negative axis. The data for OM is usually collected from various social media platforms. In an era where social media has considerable control over companies’ futures, it’s worth understanding social media and taking actions accordingly. OM comes to the fore here as the scale of the discussion about companies increases, and it becomes unfeasible to gauge opinion on individual levels. Thus, the companies opt to automize this process by applying machine learning (ML) approaches to their data. For the last two decades, OM or sentiment analysis (SA) has been mainly performed by applying ML classification algorithms such as support vector machines (SVM) and Naïve Bayes to a bag of n-gram representations of textual data. With the advent of deep learning and its apparent success in NLP, traditional methods have become obsolete. Transfer learning paradigm that has been commonly used in computer vision (CV) problems started to shape NLP approaches and language models (LM) lately. This gave a sudden rise to the usage of the pretrained language model (PTM), which contains language representations that are obtained by training it on the large datasets using self-supervised learning objectives. The PTMs are further fine-tuned by a specialized downstream task dataset to produce efficient models for various NLP tasks such as OM, NER (Named-Entity Recognition), Question Answering (QA), and so forth. In this study, the traditional and modern NLP approaches have been evaluated for OM by using a sizable corpus belonging to a large private company containing about 76,000 comments in Turkish: SVM with a bag of n-grams, and two chosen pre-trained models, multilingual universal sentence encoder (MUSE) and bidirectional encoder representations from transformers (BERT). The MUSE model is a multilingual model that supports 16 languages, including Turkish, and it is based on convolutional neural networks. The BERT is a monolingual model in our case and transformers-based neural networks. It uses a masked language model and next sentence prediction tasks that allow the bidirectional training of the transformers. During the training phase of the architecture, pre-processing operations such as morphological parsing, stemming, and spelling correction was not used since the experiments showed that their contribution to the model performance was found insignificant even though Turkish is a highly agglutinative and inflective language. The results show that usage of deep learning methods with pre-trained models and fine-tuning achieve about 11% improvement over SVM for OM. The BERT model achieved around 94% prediction accuracy while the MUSE model achieved around 88% and SVM did around 83%. The MUSE multilingual model shows better results than SVM, but it still performs worse than the monolingual BERT model.

Keywords: BERT, MUSE, opinion mining, pretrained language model, SVM, Turkish

Procedia PDF Downloads 112

Authors: Nutcha Taneepanichskul, Helen C. Hailes, Mark Miodownik

Abstract:

Plastic waste has emerged as a critical global environmental challenge, primarily driven by the prevalent use of conventional plastics derived from petrochemical refining and manufacturing processes in modern packaging. While these plastics serve vital functions, their persistence in the environment post-disposal poses significant threats to ecosystems. Addressing this issue necessitates approaches, one of which involves the development of biodegradable plastics designed to degrade under controlled conditions, such as industrial composting facilities. It is imperative to note that compostable plastics are engineered for degradation within specific environments and are not suited for uncontrolled settings, including natural landscapes and aquatic ecosystems. The full benefits of compostable packaging are realized when subjected to industrial composting, preventing environmental contamination and waste stream pollution. Therefore, effective sorting technologies are essential to enhance composting rates for these materials and diminish the risk of contaminating recycling streams. In this study, it leverage hyperspectral imaging technology (HSI) coupled with advanced machine learning algorithms to accurately identify various types of plastics, encompassing conventional variants like Polyethylene terephthalate (PET), Polypropylene (PP), Low density polyethylene (LDPE), High density polyethylene (HDPE) and biodegradable alternatives such as Polybutylene adipate terephthalate (PBAT), Polylactic acid (PLA), and Polyhydroxyalkanoates (PHA). The dataset is partitioned into three subsets: a training dataset comprising uncontaminated conventional and biodegradable plastics, a validation dataset encompassing contaminated plastics of both types, and a testing dataset featuring real-world packaging items in both pristine and contaminated states. Five distinct machine learning algorithms, namely Partial Least Squares Discriminant Analysis (PLS-DA), Support Vector Machine (SVM), Convolutional Neural Network (CNN), Logistic Regression, and Decision Tree Algorithm, were developed and evaluated for their classification performance. Remarkably, the Logistic Regression and CNN model exhibited the most promising outcomes, achieving a perfect accuracy rate of 100% for the training and validation datasets. Notably, the testing dataset yielded an accuracy exceeding 80%. The successful implementation of this sorting technology within recycling and composting facilities holds the potential to significantly elevate recycling and composting rates. As a result, the envisioned circular economy for plastics can be established, thereby offering a viable solution to mitigate plastic pollution.

Keywords: biodegradable plastics, sorting technology, hyperspectral imaging technology, machine learning algorithms

Procedia PDF Downloads 39

Authors: Robert Rorie, Lisa Fern

Abstract:

The integration of Unmanned Aircraft Systems (UAS) into the National Airspace System (NAS) poses a variety of technical challenges to UAS developers and aviation regulators. In response to growing demand for access to civil airspace in the United States, the Federal Aviation Administration (FAA) has produced a roadmap identifying key areas requiring further research and development. One such technical challenge is the development of a ‘detect and avoid’ system (DAA; previously referred to as ‘sense and avoid’) to replace the ‘see and avoid’ requirement in manned aviation. The purpose of the DAA system is to support the pilot, situated at a ground control station (GCS) rather than in the cockpit of the aircraft, in maintaining ‘well clear’ of nearby aircraft through the use of GCS displays and alerts. In addition to its primary function of aiding the pilot in maintaining well clear, the DAA system must also safely interoperate with existing NAS systems and operations, such as the airspace management procedures of air traffic controllers (ATC) and collision avoidance (CA) systems currently in use by manned aircraft, namely the Traffic alert and Collision Avoidance System (TCAS) II. It is anticipated that many UAS architectures will integrate both a DAA system and a TCAS II. It is therefore necessary to explicitly study the integration of DAA and TCAS II alerting structures and maneuver guidance formats to ensure that pilots understand the appropriate type and urgency of their response to the various alerts. This paper presents a concept of interoperability for the two systems. The concept was developed with the goal of avoiding any negative impact on the performance level of TCAS II (understanding that TCAS II must largely be left as-is) while retaining a DAA system that still effectively enables pilots to maintain well clear, and, as a result, successfully reduces the frequency of collision hazards. The interoperability concept described in the paper focuses primarily on facilitating the transition from a late-stage DAA encounter (where a loss of well clear is imminent) to a TCAS II corrective Resolution Advisory (RA), which requires pilot compliance with the directive RA guidance (e.g., climb, descend) within five seconds of its issuance. The interoperability concept was presented to 10 participants (6 active UAS pilots and 4 active commercial pilots) in a medium-fidelity, human-in-the-loop simulation designed to stress different aspects of the DAA and TCAS II systems. Pilot response times, compliance rates and subjective assessments were recorded. Results indicated that pilots exhibited comprehension of, and appropriate prioritization within, the DAA-TCAS II combined alert structure. Pilots demonstrated a high rate of compliance with TCAS II RAs and were also seen to respond to corrective RAs within the five second requirement established for manned aircraft. The DAA system presented under test was also shown to be effective in supporting pilots’ ability to maintain well clear in the overwhelming majority of cases in which pilots had sufficient time to respond. The paper ends with a discussion of next steps for research on integrating UAS into civil airspace.

Keywords: detect and avoid, interoperability, traffic alert and collision avoidance system (TCAS II), unmanned aircraft systems

Procedia PDF Downloads 242

Authors: Lakshmi N. Reddi, Akanksha Varma Sagi

Abstract:

Studies on biomimetics are largely developed, deriving inspiration from natural processes in our objective world to develop novel technologies. Recent studies are diverse in nature, making their categorization quite challenging. Based on an exhaustive survey, we developed categorizations based on either the essential elements of nature - air, water, land, fire, and space, or on form/shape, functionality, and process. Such diverse studies as aircraft wings inspired by bird wings, a self-cleaning coating inspired by a lotus petal, wetsuits inspired by beaver fur, and search algorithms inspired by arboreal ant path networks lend themselves to these categorizations. Our categorizations of biomimetic studies allowed us to define a different dimension of biomimetics. This new dimension is not restricted to inspiration from the objective world. It is based on the premise that the biological processes observed in the objective world find their reflections in our human bodies in a variety of ways. For example, the lungs provide the most efficient example for liquid-gas phase exchange, the heart exemplifies a very efficient pumping and circulatory system, and the kidneys epitomize the most effective cleaning system. The main focus of this paper is to bring out the magnificence of the cerebro-spinal system (CSS) insofar as it relates to our current computer architecture. In particular, the paper uses four key measures to analyze the differences between CSS and human- engineered computational systems. These are adaptability, sustainability, energy efficiency, and resilience. We found that the cerebrospinal system reveals some important challenges in the development and evolution of our current computer architectures. In particular, the myriad ways in which the CSS is integrated with other systems/processes (circulatory, respiration, etc) offer useful insights on how the human-engineered computational systems could be made more sustainable, energy-efficient, resilient, and adaptable. In our paper, we highlight the energy consumption differences between CSS and our current computational designs. Apart from the obvious differences in materials used between the two, the systemic nature of how CSS functions provides clues to enhance life-cycles of our current computational systems. The rapid formation and changes in the physiology of dendritic spines and their synaptic plasticity causing memory changes (ex., long-term potentiation and long-term depression) allowed us to formulate differences in the adaptability and resilience of CSS. In addition, the CSS is sustained by integrative functions of various organs, and its robustness comes from its interdependence with the circulatory system. The paper documents and analyzes quantifiable differences between the two in terms of the four measures. Our analyses point out the possibilities in the development of computational systems that are more adaptable, sustainable, energy efficient, and resilient. It concludes with the potential approaches for technological advancement through creation of more interconnected and interdependent systems to replicate the effective operation of cerebro-spinal system.

Keywords: cerebrospinal system, computer architecture, adaptability, sustainability, resilience, energy efficiency

Procedia PDF Downloads 54

Authors: Madushani Rodrigo, Banuka Athuraliya

Abstract:

In today's world of medical diagnosis and prediction, machine learning stands out as a strong tool, transforming old ways of caring for health. This study analyzes the use of machine learning in the specialized domain of sports medicine, with a focus on the timely and accurate detection of bone fractures in cricket athletes. Failure to identify bone fractures in real time can result in malunion or non-union conditions. To ensure proper treatment and enhance the bone healing process, accurately identifying fracture locations and types is necessary. When interpreting X-ray images, it relies on the expertise and experience of medical professionals in the identification process. Sometimes, radiographic images are of low quality, leading to potential issues. Therefore, it is necessary to have a proper approach to accurately localize and classify fractures in real time. The research has revealed that the optimal approach needs to address the stated problem and employ appropriate radiographic image processing techniques and object detection algorithms. These algorithms should effectively localize and accurately classify all types of fractures with high precision and in a timely manner. In order to overcome the challenges of misidentifying fractures, a distinct model for fracture localization and classification has been implemented. The research also incorporates radiographic image enhancement and preprocessing techniques to overcome the limitations posed by low-quality images. A classification ensemble model has been implemented using ResNet18 and VGG16. In parallel, a fracture segmentation model has been implemented using the enhanced U-Net architecture. Combining the results of these two implemented models, the FracXpert system can accurately localize exact fracture locations along with fracture types from the available 12 different types of fracture patterns, which include avulsion, comminuted, compressed, dislocation, greenstick, hairline, impacted, intraarticular, longitudinal, oblique, pathological, and spiral. This system will generate a confidence score level indicating the degree of confidence in the predicted result. Using ResNet18 and VGG16 architectures, the implemented fracture segmentation model, based on the U-Net architecture, achieved a high accuracy level of 99.94%, demonstrating its precision in identifying fracture locations. Simultaneously, the classification ensemble model achieved an accuracy of 81.0%, showcasing its ability to categorize various fracture patterns, which is instrumental in the fracture treatment process. In conclusion, FracXpert has become a promising ML application in sports medicine, demonstrating its potential to revolutionize fracture detection processes. By leveraging the power of ML algorithms, this study contributes to the advancement of diagnostic capabilities in cricket athlete healthcare, ensuring timely and accurate identification of bone fractures for the best treatment outcomes.

Keywords: multiclass classification, object detection, ResNet18, U-Net, VGG16

Procedia PDF Downloads 27

Authors: Reuben Newell, Edward Archer, Alistair McIlhagger, Calvin Ralph

Abstract:

Modern industry has developed a need for innovative 3D composite materials due to their attractive material properties. Composite materials are composed of a fibre reinforcement encased in a polymer matrix. The fibre reinforcement consists of warp, weft and binder yarns or tows woven together into a preform. The mechanical performance of composite material is largely controlled by the properties of the preform. As a result, the bulk of recent textile research has been focused on the design of high-strength preform architectures. Studies looking at optimisation of the weaving process have largely been neglected. It has been reported that yarns experience varying levels of damage during weaving, resulting in filament breakage and ultimately compromised composite mechanical performance. The weaving parameters involved in causing this yarn damage are not fully understood. Recent studies indicate that poor yarn tension control may be an influencing factor. As tension is increased, the yarn-to-yarn and yarn-to-weaving-equipment interactions are heightened, maximising damage. The correlation between yarn tension variation and weaving damage severity has never been adequately researched or quantified. A novel study is needed which accesses the influence of tension variation on the mechanical properties of woven yarns. This study has looked to quantify the variation of yarn tension throughout weaving and sought to link the impact of tension to weaving damage. Multiple yarns were randomly selected, and their tension was measured across the creel and shedding stages of weaving, using a hand-held tension meter. Sections of the same yarn were subsequently cut from the loom machine and tensile tested. A comparison study was made between the tensile strength of pristine and tensioned yarns to determine the induced weaving damage. Yarns from bobbins at the rear of the creel were under the least amount of tension (0.5-2.0N) compared to yarns positioned at the front of the creel (1.5-3.5N). This increase in tension has been linked to the sharp turn in the yarn path between bobbins at the front of the creel and creel I-board. Creel yarns under the lower tension suffered a 3% loss of tensile strength, compared to 7% for the greater tensioned yarns. During shedding, the tension on the yarns was higher than in the creel. The upper shed yarns were exposed to a decreased tension (3.0-4.5N) compared to the lower shed yarns (4.0-5.5N). Shed yarns under the lower tension suffered a 10% loss of tensile strength, compared to 14% for the greater tensioned yarns. Interestingly, the most severely damaged yarn was exposed to both the largest creel and shedding tensions. This study confirms for the first time that yarns under a greater level of tension suffer an increased amount of weaving damage. Significant variation of yarn tension has been identified across the creel and shedding stages of weaving. This leads to a variance of mechanical properties across the woven preform and ultimately the final composite part. The outcome from this study highlights the need for optimised yarn tension control during preform manufacture to minimize yarn-induced weaving damage.

Keywords: optimisation of preform manufacture, tensile testing of damaged tows, variation of yarn weaving tension, weaving damage

Procedia PDF Downloads 202