World Academy of Science, Engineering and Technology

Authors: Krishna Mohan Bathula, Fatou Bintou Loucoubar, FNU Kaleemunnisa, Christelle Scharff, Mark Anthony De Castro

Abstract:

Voice recognition algorithms such as automatic speech recognition and text-to-speech systems with African languages can play an important role in bridging the digital divide of Artificial Intelligence in Africa, contributing to the establishment of a fully inclusive information society. This paper proposes a Deep Learning model that can classify the user responses as inputs for an interactive voice response system. A dataset with Wolof language words ‘yes’ and ‘no’ is collected as audio recordings. A two stage Data Augmentation approach is adopted for enhancing the dataset size required by the deep neural network. Data preprocessing and feature engineering with Mel-Frequency Cepstral Coefficients are implemented. Convolutional Neural Networks (CNNs) have proven to be very powerful in image classification and are promising for audio processing when sounds are transformed into spectra. For performing voice response classification, the recordings are transformed into sound frequency feature spectra and then applied image classification methodology using a deep CNN model. The inference model of this trained and reusable Wolof voice response recognition system can be integrated with many applications associated with both web and mobile platforms.

Keywords: automatic speech recognition, interactive voice response, voice response recognition, wolof word classification

Procedia PDF Downloads 87

Authors: Balaji Viswanathan, Tim Oates

Abstract:

A number of pilot projects have commenced across the world to use robots for senior care. We aim to measure the likeability of these robots among seniors and help robot designers focus on the features that matter. We built a robot likability score with over 30 parameters and used this to interview 50 seniors in various locations in the United States. This paper presents the results of this field research.

Keywords: HRI, assistive robotics, social robotics, HCI, aging

Procedia PDF Downloads 66

Authors: Magdalena Wilk-Kozubek, Jakub Pawłów, Maciej Czajkowski, Maria Zdończyk, Katarzyna Ślepokura, Joanna Cybińska

Abstract:

Thermochromic materials belong to the family of intelligent materials that change their color in response to temperature changes; this ability is called thermochromism. Thermochromic behavior can be displayed by both isolated compounds and multicomponent mixtures. Fluoran leuco dye-based mixtures are well-known thermochromic systems used, for example, in heat-sensitive FAX paper. Weak acids often serve as color developers for such systems. As the temperature increases, the acids melt, and the mixtures become colored. The objective of this research is to determine the influence of acids showing a liquid crystalline nematic phase on the development of the fluoran dye. For this purpose, fluoran-based mixtures with 4-n-alkylbenzoic acids were prepared. The mixtures are colored at room temperature, but they become colorless upon the melting of the acids. The melting of acids is associated not only with a change in the color of the mixtures but also with a change in their emission color. Phase transitions were investigated by temperature-dependent powder X-ray diffraction and differential scanning calorimetry; nematic phases were visualized by polarized optical microscopy, and color and emission changes were studied by UV-Vis diffuse reflectance and photoluminescence spectroscopies, respectively. When 4-n-alkylbenzoic acids are used as color developers, the fluoran-based mixtures become colorless after the melting of the acids. This is because the melting of acids is accompanied by the transition from the crystalline phase to the nematic phase, in which the molecular arrangement of the acids does not allow the fluoran dye to be developed.

Keywords: color developer, leuco dye, liquid crystal, thermochromism

Procedia PDF Downloads 74

Authors: Arpit Rai

Abstract:

In this work, we propose PatchMix, a semi-supervised method for pre-training visual representations. PatchMix mixes patches of two images and then solves an auxiliary task of predicting the label of each patch in the mixed image. Our experiments on the CIFAR-10, 100 and the SVHN dataset show that the representations learned by this method encodes useful information for transfer to new tasks and outperform the baseline Residual Network encoders by on CIFAR 10 by 12% on ResNet 101 and 2% on ResNet-56, by 4% on CIFAR-100 on ResNet101 and by 6% on SVHN dataset on the ResNet-101 baseline model.

Keywords: self-supervised learning, representation learning, computer vision, generalization

Procedia PDF Downloads 60

Authors: Mohamed Gaith, Zaid Haddadin, Abdulah Wahbe, Mahmoud Hamam, Mahmoud Qunees, Mohammad Al Khatib, Mohammad Bsaileh, Abd Al-Aziz Jaber, Ahmad Aqra’a

Abstract:

The crack is one of the most common types of failure in pipelines that convey fluid, and early detection of the crack may assist to avoid the piping system from experiencing catastrophic damage, which would otherwise be fatal. The influence of flow velocity and the presence of a crack on the performance of a tapered simply supported pipe containing moving fluid is explored using the finite element approach in this study. ANSYS software is used to simulate the pipe as Bernoulli's beam theory. In this paper, the fluctuation of natural frequencies and matching mode shapes for various scenarios owing to changes in fluid speed and the presence of damage is discussed in detail.

Keywords: damage detection, finite element, tapered pipe, vibration characteristics

Procedia PDF Downloads 136

Authors: Tsega Asresa Mengistu, Getahun Tigistu

Abstract:

Computer vision is an artificial intelligence subfield that allows computers and systems to retrieve meaning from digital images. It is applied in various fields of study self-driving cars, video surveillance, agriculture, Quality control, Health care, construction, military, and everyday life. Aromatic and medicinal plants are botanical raw materials used in cosmetics, medicines, health foods, and other natural health products for therapeutic and Aromatic culinary purposes. Herbal industries depend on these special plants. These plants and their products not only serve as a valuable source of income for farmers and entrepreneurs, and going to export not only industrial raw materials but also valuable foreign exchange. There is a lack of technologies for the classification and identification of Aromatic and medicinal plants in Ethiopia. The manual identification system of plants is a tedious, time-consuming, labor, and lengthy process. For farmers, industry personnel, academics, and pharmacists, it is still difficult to identify parts and usage of plants before ingredient extraction. In order to solve this problem, the researcher uses a deep learning approach for the efficient identification of aromatic and medicinal plants by using a convolutional neural network. The objective of the proposed study is to identify the aromatic and medicinal plant Parts and usages using computer vision technology. Therefore, this research initiated a model for the automatic classification of aromatic and medicinal plants by exploring computer vision technology. Morphological characteristics are still the most important tools for the identification of plants. Leaves are the most widely used parts of plants besides the root, flower and fruit, latex, and barks. The study was conducted on aromatic and medicinal plants available in the Ethiopian Institute of Agricultural Research center. An experimental research design is proposed for this study. This is conducted in Convolutional neural networks and Transfer learning. The Researcher employs sigmoid Activation as the last layer and Rectifier liner unit in the hidden layers. Finally, the researcher got a classification accuracy of 66.4 in convolutional neural networks and 67.3 in mobile networks, and 64 in the Visual Geometry Group.

Keywords: aromatic and medicinal plants, computer vision, deep convolutional neural network

Procedia PDF Downloads 386

Authors: M. Akhondizadeh, Mohsen Khajoei, Mojtaba Khajoei

Abstract:

The purpose of the present work was the design of a towing machine which was decided to be manufactured by Arman Gohar-e-Sirjan company in the Gol-e-Gohar iron ore complex in Iran. The load analysis has been conducted to determine the static and dynamic loads at the critical conditions. The inertial forces due to the velocity increment and road bump have been considered in load evaluation. The form of loading of the present machine is hauling and/or conveying the mineral machines on the mini ramp. Several stages of these forms of loading, from the initial touch of the tow and carried machine to the final position, have been assessed to determine the critical state. The stress analysis has been performed by the ANSYS software. Several geometries for the main load-carrying elements have been analyzed to have the optimum design by the minimum weight of the structure. Finally, a structure with a total weight of 38 tons has been designed with a static load-carrying capacity of 80 tons by considering the 40 tons additional capacity for dynamic effects. The stress analysis for 120 tons load gives the minimum safety factor of 1.18.

Keywords: mechanical design, stress analysis, tow structure, dynamic load, static load

Procedia PDF Downloads 80

Authors: Keda Li, Hong Hu

Abstract:

Auxetic materials, as an emerging artificial designed metamaterial has attracted growing attention due to their promising negative Poisson’s ratio behaviors and tunable properties. The conventional auxetic lattice structures for which the deformation process is governed by a bending-dominated mechanism have faced the limitation of poor mechanical performance for many potential engineering applications. Recently, both load-bearing and energy absorption capabilities have become a crucial consideration in auxetic structure design. This study reports the finite element analysis of a class of hexagonal double-arrowhead auxetic structures with enhanced stiffness and energy absorption performance. The structure design was developed by extending the traditional double-arrowhead honeycomb to a hexagon frame, the stretching-dominated deformation mechanism was determined according to Maxwell’s stability criterion. The finite element (FE) models of 2D lattice structures established with stainless steel material were analyzed in ABAQUS/Standard for predicting in-plane structural deformation mechanism, failure process, and compressive elastic properties. Based on the computational simulation, the parametric analysis was studied to investigate the effect of the structural parameters on Poisson’s ratio and mechanical properties. The geometrical optimization was then implemented to achieve the optimal Poisson’s ratio for the maximum specific energy absorption. In addition, the optimized 2D lattice structure was correspondingly converted into a 3D geometry configuration by using the orthogonally splicing method. The numerical results of 2D and 3D structures under compressive quasi-static loading conditions were compared separately with the traditional double-arrowhead re-entrant honeycomb in terms of specific Young's moduli, Poisson's ratios, and specified energy absorption. As a result, the energy absorption capability and stiffness are significantly reinforced with a wide range of Poisson’s ratio compared to traditional double-arrowhead re-entrant honeycomb. The auxetic behaviors, energy absorption capability, and yield strength of the proposed structure are adjustable with different combinations of joint angle, struts thickness, and the length-width ratio of the representative unit cell. The numerical prediction in this study suggests the proposed concept of hexagonal double-arrowhead structure could be a suitable candidate for the energy absorption applications with a constant request of load-bearing capacity. For future research, experimental analysis is required for the validation of the numerical simulation.

Keywords: auxetic, energy absorption capacity, finite element analysis, negative Poisson's ratio, re-entrant hexagonal honeycomb

Procedia PDF Downloads 64

Authors: Mohamed S. Selim, Nesreen A. Fatthallah, Shimaa A. Higazy, Zhifeng Hao, Xiang Chen

Abstract:

After the prohibition of tin-based fouling-prevention coatings in 2003, the researchers were directed toward eco-friendly coatings. Because of their nonstick, environmental, and economic benefits, foul-release nanocoatings have received a lot of attention. They use physical anti-adhesion terminology to deter any fouling attachment.Natural bioinspired surfaces have micro/nano-roughness and low surface free energy features, which may inspire the design of dynamic antifouling coatings. Graphene-based nanocomposite surfaces were designed to combat marine-fouling adhesion with ecological as well as eco-friendly effects rather than biocidal solutions. Polymer–graphenenanofiller hybrids are a novel class of composite materials in fouling-prevention applications. The controlled preparation of nanoscale orientation, arrangement, and direction along the composite building blocks would result in superior fouling prohibition. This work representsfoul-release nanocomposite top coats for marine coating applications with superhydrophobicity, surface inertness against fouling adherence, cost-effectiveness, and increased lifetime.

Keywords: foul-release nanocoatings, graphene-based nanocomposite, polymer, nanofillers

Procedia PDF Downloads 108

Authors: Ewelina Musielak, Agnieszka Feliczak-Guzik, Izabela Nowak

Abstract:

Based on the latest data, it is expected that the substances of therapeutic interest used will be as natural as possible. Therefore, active substances with the highest possible efficacy and low toxicity are sought. Among natural substances with therapeutic effects, those of plant origin stand out. Curcumin isolated from the Curcuma longa plant has proven to be particularly important from a medical point of view. Due to its ability to regulate many important transcription factors, cytokines, and protein kinases, curcumin has found use as an anti-inflammatory, antioxidant, antiproliferative, antiangiogenic, and anticancer agent. The unfavorable properties of curcumin, such as low solubility, poor bioavailability, and rapid degradation under neutral or alkaline pH conditions, limit its clinical application. These problems can be solved by combining curcumin with suitable carriers such as hierarchical zeolites. This is a new class of materials that exhibit several advantages. Hierarchical zeolites used as drug carriers enable delayed release of the active ingredient and promote drug transport to the desired tissues and organs. In addition, hierarchical zeolites play an important role in regulating micronutrient levels in the body and have been used successfully in cancer diagnosis and therapy. To apply curcumin to hierarchical zeolites synthesized from commercial FAU zeolite, solutions containing curcumin, carrier and acetone were prepared. The prepared mixtures were then stirred on a magnetic stirrer for 24 h at room temperature. The curcumin-filled hierarchical zeolites were drained into a glass funnel, where they were washed three times with acetone and distilled water, after which the obtained material was air-dried until completely dry. In addition, the effect of piperine addition to zeolite carrier containing a sufficient amount of curcumin was studied. The resulting products were weighed and the percentage of pure curcumin in the hierarchical zeolite was calculated. All the synthesized materials were characterized by several techniques: elemental analysis, transmission electron microscopy (TEM), Fourier transform infrared spectroscopy, Fourier transform infrared (FT-IR), N2 adsorption, and X-ray diffraction (XRD) and thermogravimetric analysis (TGA). The aim of the presented study was to improve the biological activity of curcumin by applying it to hierarchical zeolites based on FAU zeolite. The results showed that the loading efficiency of curcumin into hierarchical zeolites based on commercial FAU-type zeolite is enhanced by modifying the zeolite carrier itself. The hierarchical zeolites proved to be very good and efficient carriers of plant-derived active ingredients such as curcumin.

Keywords: carriers of active substances, curcumin, hierarchical zeolites, incorporation

Procedia PDF Downloads 71

Authors: Fatimah Al-Hayazi, Ehteram. A. Noor, Aisha H. Moubaraki

Abstract:

The inhibitive effect of Securigera securidaca seed extract (SSE) on mild steel corrosion in 1 M HCl solution has been studied by weight loss and electrochemical techniques at four different temperatures. All techniques studied provided data that the studied extract does well at all temperatures, and its inhibitory action increases with increasing its concentration. SEM images indicate thin-film formation on mild steel when corroded in solutions containing 1 g L-1 of inhibitor either at low or high temperatures. The polarization studies showed that SSE acts as an anodic inhibitor. Both polarization and impedance techniques show an acceleration behaviour for SSE at concentrations ≤ 0.1 g L-1 at all temperatures. At concentrations ≥ 0.1 g L-1, the efficiency of SSE is dramatically increased with increasing concentration, and its value does not change appreciably with increasing temperature. It was found that all adsorption data obeyed Temkin adsorption isotherm. Kinetic activation and thermodynamic adsorption parameters are evaluated and discussed. The results revealed an endothermic corrosion process with an associative activation mechanism, while a comprehensive adsorption mechanism for SSE on mild steel surfaces is suggested, in which both physical and chemical adsorption are involved in the adsorption process. A good correlation between inhibitor constituents and their inhibitory action was obtained.

Keywords: corrosion, inhibition of steel, hydrochloric acid, thermodynamic study

Procedia PDF Downloads 71

Authors: Ian Omung'a

Abstract:

Breast cancer remains to be one of the deadliest cancers for women worldwide, with the risk of developing tumors being as high as 50 percent in Sub-Saharan African countries like Kenya. With as many as 42 percent of these cases set to be diagnosed late when cancer has metastasized and or the prognosis has become terminal, Full Field Digital [FFD] Mammography remains an effective screening technique that leads to early detection where in most cases, successful interventions can be made to control or eliminate the tumors altogether. FFD Mammograms have been proven to multiply more effective when used together with Computer-Aided Detection and Diagnosis [CADe] systems, relying on algorithmic implementations of Deep Learning techniques in Computer Vision to carry out deep pattern recognition that is comparable to the level of a human radiologist and decipher whether specific areas of interest in the mammogram scan image portray abnormalities if any and whether these abnormalities are indicative of a benign or malignant tumor. Within this paper, we review emergent Deep Learning techniques that will prove relevant to the development of State-of-The-Art FFD Mammogram CADe systems. These techniques will span self-supervised learning for context-encoded occlusion, self-supervised learning for pre-processing and labeling automation, as well as the creation of a standardized large-scale mammography dataset as a benchmark for CADe systems' evaluation. Finally, comparisons are drawn between existing practices that pre-date these techniques and how the development of CADe systems that incorporate them will be different.

Keywords: breast cancer diagnosis, computer aided detection and diagnosis, deep learning, whole mammogram classfication, ultrasound classification, computer vision

Procedia PDF Downloads 70

Authors: Gholba Niranjan Dilip, Anil Kumar

Abstract:

Conduct of accurate reliable mapping of oil palm plantations and census of individual palm trees is a huge challenge. This study addresses this challenge and developed an optimized solution implemented deep learning techniques on remote sensing data. The oil palm is a very important tropical crop. To improve its productivity and land management, it is imperative to have accurate census over large areas. Since, manual census is costly and prone to approximations, a methodology for automated census using panchromatic images from Cartosat-2, SkySat and World View-3 satellites is demonstrated. It is selected two different study sites in Indonesia. The customized set of training data and ground-truth data are created for this study from Cartosat-2 images. The pre-trained model of Single Shot MultiBox Detector (SSD) Lite MobileNet V2 Convolutional Neural Network (CNN) from the TensorFlow Object Detection API is subjected to transfer learning on this customized dataset. The SSD model is able to generate the bounding boxes for each oil palm and also do the counting of palms with good accuracy on the panchromatic images. The detection yielded an F-Score of 83.16 % on seven different images. The detections are buffered and dissolved to generate polygons demarcating the boundaries of the oil palm plantations. This provided the area under the plantations and also gave maps of their location, thereby completing the automated census, with a fairly high accuracy (≈100%). The trained CNN was found competent enough to detect oil palm crowns from images obtained from multiple satellite sensors and of varying temporal vintage. It helped to estimate the increase in oil palm plantations from 2014 to 2021 in the study area. The study proved that high-resolution panchromatic satellite image can successfully be used to undertake census of oil palm plantations using CNNs.

Keywords: object detection, oil palm tree census, panchromatic images, single shot multibox detector

Procedia PDF Downloads 138

Authors: A. Feliczak Guzik, I. Nowak

Abstract:

Zeolites, classified as microporous materials, are a large group of crystalline aluminosilicate materials commonly used in the chemical industry. These materials are characterized by large specific surface area, high adsorption capacity, hydrothermal and thermal stability. However, the micropores present in them impose strong mass transfer limitations, resulting in low catalytic performance. Consequently, mesoporous (hierarchical) zeolites have attracted considerable attention from researchers. These materials possess additional porosity in the mesopore size region (2-50 nm according to IUPAC). Mesoporous zeolites, based on commercial MFI-type zeolites modified with silver, were synthesized as follows: 0.5 g of zeolite was dispersed in a mixture containing CTABr (template), water, ethanol, and ammonia under ultrasound for 30 min at 65°C. The silicon source, which was tetraethyl orthosilicate, was then added and stirred for 4 h. After this time, silver(I) nitrate was added. In a further step, the whole mixture was filtered and washed with water: ethanol mixture. The template was removed by calcination at 550°C for 5h. All the materials obtained were characterized by the following techniques: X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy (SEM), nitrogen adsorption/desorption isotherms, FTIR spectroscopy. X-ray diffraction and low-temperature nitrogen adsorption/desorption isotherms revealed additional secondary porosity. Moreover, the structure of the commercial zeolite was preserved during most of the material syntheses. The aforementioned materials were used in the epoxidation reaction of cyclohexene using conventional heating and microwave radiation heating. The composition of the reaction mixture was analyzed every 1 h by gas chromatography. As a result, about 60% conversion of cyclohexene and high selectivity to the desired reaction products i.e., 1,2-epoxy cyclohexane and 1,2-cyclohexane diol, were obtained.

Keywords: catalytic application, characterization, epoxidation, hierarchical zeolites, synthesis

Procedia PDF Downloads 66

Authors: Jun Seung Woo, Shin Dong Ho

Abstract:

The Image Captioning models with Attention technology have developed significantly compared to previous models, but it is still unsatisfactory in recognizing images. We perform an extensive search over seven interesting Convolutional Neural Networks(CNN) architectures to analyze the behavior of different models for image captioning. We compared seven different CNN Architectures, according to batch size, using on public benchmarks: MS-COCO datasets. In our experimental results, DenseNet and InceptionV3 got about 14% loss and about 160sec training time per epoch. It was the most satisfactory result among the seven CNN architectures after training 50 epochs on GPU.

Keywords: deep learning, image captioning, CNN architectures, densenet, inceptionV3

Procedia PDF Downloads 99

Authors: Minh Nguyen

Abstract:

Today, QR or Quick Response Codes are prevalent, and mobile/smart devices can efficiently read and understand them. Therefore, we can see their appearance in many areas, such as storing web pages/websites, business phone numbers, redirecting to an app download, business location, social media. The popularity of the QR Code is mainly because of its many advantages, such as it can hold a good amount of information, is small, easy to scan and read by a general RGB camera, and it can still work with some damages on its surface. However, there are still some issues. For instance, some areas needed to be kept untouched for its successful decode (e.g., the “Finder Patterns,” the “Quiet Zone,” etc.), the capability of built-in auto-correction is not robust enough, and it is not flexible enough for many application such as Augment Reality (AR). We proposed a new Colour Quick Response Code that has several advantages over the original ones: (1) there is no untouchable area, (2) it allows up to 40% of the entire code area to be damaged, (3) it is more beneficial for Augmented Reality applications, and (4) it is back-compatible and readable by available QR Code scanners such as Pyzbar. From our experience, our Colour Quick Response Code is significantly more flexible on damage compared to the original QR Code. Our code is believed to be suitable in situations where standard 2D Barcodes fail to work, such as curved and shiny surfaces, for instance, medical blood test sample tubes and syringes.

Keywords: QR code, computer vision, image processing, 2D barcode

Procedia PDF Downloads 93

Authors: Ci Lin, Tet Yeap, Iluju Kiringa

Abstract:

This paper introduces the subspace rotation algorithm (SRA) to train the Restricted Hopfield Network (RHN) as an auto-associative memory. Subspace rotation algorithm is a gradient-free subspace tracking approach based on the singular value decomposition (SVD). In comparison with Backpropagation Through Time (BPTT) on training RHN, it is observed that SRA could always converge to the optimal solution and BPTT could not achieve the same performance when the model becomes complex, and the number of patterns is large. The AUTS case study showed that the RHN model trained by SRA could achieve a better structure of attraction basin with larger radius(in general) than the Hopfield Network(HNN) model trained by Hebbian learning rule. Through learning 10000 patterns from MNIST dataset with RHN models with different number of hidden nodes, it is observed that an several components could be adjusted to achieve a balance between recovery accuracy and noise resistance.

Keywords: hopfield neural network, restricted hopfield network, subspace rotation algorithm, hebbian learning rule

Procedia PDF Downloads 90

Authors: Tian Zhang

Abstract:

Unsupervised object segmentation aims at segmenting objects in sequential images and obtaining the mask of each object without any manual intervention. Unsupervised segmentation remains a challenging task due to the lack of prior knowledge about these objects. Previous methods often require manually specifying the action of each object, which is often difficult to obtain. Instead, this paper does not need action information of objects and automatically learns the actions and relations among objects from the structured environment. To obtain the object segmentation of sequential images, the relationships between objects and images are extracted to infer the action and interaction of objects based on the multi-head attention mechanism. Three types of objects’ relationships in the object segmentation task are proposed: the relationship between objects in the same frame, the relationship between objects in two frames, and the relationship between objects and historical information. Based on these relationships, the proposed model (1) is effective in multiple objects segmentation tasks, (2) just needs images as input, and (3) produces better segmentation results as more relationships are considered. The experimental results on multiple datasets show that this paper’s method achieves state-of-art performance. The quantitative and qualitative analyses of the result are conducted. The proposed method could be easily extended to other similar applications.

Keywords: unsupervised object segmentation, attention mechanism, contrastive learning, structured environment

Procedia PDF Downloads 85

Authors: Muhammad Zubair Khan, Yugyung Lee

Abstract:

Deep learning has recently achieved enormous response in semantic image segmentation. The previously developed U-Net inspired architectures operate with continuous stride and pooling operations, leading to spatial data loss. Also, the methods lack establishing long-term pixels connection to preserve context knowledge and reduce spatial loss in prediction. This article developed encoder-decoder architecture with bi-directional LSTM embedded in long skip-connections and densely connected convolution blocks. The network non-linearly combines the feature maps across encoder-decoder paths for finding dependency and correlation between image pixels. Additionally, the densely connected convolutional blocks are kept in the final encoding layer to reuse features and prevent redundant data sharing. The method applied batch-normalization for reducing internal covariate shift in data distributions. The empirical evidence shows a promising response to our method compared with other semantic segmentation techniques.

Keywords: deep learning, semantic segmentation, image analysis, pixels connection, convolution neural network

Procedia PDF Downloads 76

Authors: Xiaoxiao Li, Shuangcheng Jia, Qian Li

Abstract:

Keypoint detection has always been a challenge in the field of image recognition. This paper proposes a novelty keypoint detection method which is called Multi-Scale Feature Fusion Convolutional Network with Attention (MFFCNA). We verified that the multi-scale features with the attention mechanism module have better feature expression capability. The feature fusion between different scales makes the information that the network model can express more abundant, and the network is easier to converge. On our self-made street sign corner dataset, we validate the MFFCNA model with an accuracy of 97.8% and a recall of 81%, which are 5 and 8 percentage points higher than the HRNet network, respectively. On the COCO dataset, the AP is 71.9%, and the AR is 75.3%, which are 3 points and 2 points higher than HRNet, respectively. Extensive experiments show that our method has a remarkable improvement in the keypoint recognition tasks, and the recognition effect is better than the existing methods. Moreover, our method can be applied not only to keypoint detection but also to image classification and semantic segmentation with good generality.

Keywords: keypoint detection, feature fusion, attention, semantic segmentation

Procedia PDF Downloads 95

Authors: Pengfei Meng, Shuangcheng Jia, Qian Li

Abstract:

We proposed a real-time high-precision semantic segmentation network based on a multi-resolution feature fusion module, the auxiliary feature extracting module, upsampling module, and atrous spatial pyramid pooling (ASPP) module. We designed a feature fusion structure, which is integrated with sufficient features of different resolutions. We also studied the effect of side-branch structure on the network and made discoveries. Based on the discoveries about the side-branch of the network structure, we used a side-branch auxiliary feature extraction layer in the network to improve the effectiveness of the network. We also designed upsampling module, which has better results than the original upsampling module. In addition, we also re-considered the locations and number of atrous spatial pyramid pooling (ASPP) modules and modified the network structure according to the experimental results to further improve the effectiveness of the network. The network presented in this paper takes the backbone network of Bisenetv2 as a basic network, based on which we constructed a network structure on which we made improvements. We named this network deep multiple-resolution bilateral networks for real-time, referred to as DMBR-Net. After experimental testing, our proposed DMBR-Net network achieved 81.2% mIoU at 119FPS on the Cityscapes validation dataset, 80.7% mIoU at 109FPS on the CamVid test dataset, 29.9% mIoU at 78FPS on the COCOStuff test dataset. Compared with all lightweight real-time semantic segmentation networks, our network achieves the highest accuracy at an appropriate speed.

Keywords: multi-resolution feature fusion, atrous convolutional, bilateral networks, pyramid pooling

Procedia PDF Downloads 108

Authors: Feixiang Zhao, Shuangcheng Jia, Qian Li

Abstract:

High-precision measurement of the target’s position and size is one of the hotspots in the field of vision inspection. This paper proposes a three-dimensional object positioning and measurement method using a monocular camera and GPS, namely the Distance Area Number-New Minimize Projection Error (DAN-NMPE). Our algorithm contains two parts: DAN and NMPE; specifically, DAN is a picture sequence algorithm, NMPE is a relatively positive optimization algorithm, which greatly improves the measurement accuracy of the target’s position and size. Comprehensive experiments validate the effectiveness of our proposed method on a self-made traffic sign dataset. The results show that with the laser point cloud as the ground truth, the size and position errors of the traffic sign measured by this method are ± 5% and 0.48 ± 0.3m, respectively. In addition, we also compared it with the current mainstream method, which uses a monocular camera to locate and measure traffic signs. DAN-NMPE attains significant improvements compared to existing state-of-the-art methods, which improves the measurement accuracy of size and position by 50% and 15.8%, respectively.

Keywords: monocular camera, GPS, positioning, measurement

Procedia PDF Downloads 110

Authors: Shaik Ayesha Fathima, Shaik Noor Jahan, Duvvada Rajeswara Rao

Abstract:

Earth's environment and its evolution can be seen through satellite images in near real-time. Through satellite imagery, remote sensing data provide crucial information that can be used for a variety of applications, including image fusion, change detection, land cover classification, agriculture, mining, disaster mitigation, and monitoring climate change. The objective of this project is to propose a method for classifying satellite images according to multiple predefined land cover classes. The proposed approach involves collecting data in image format. The data is then pre-processed using data pre-processing techniques. The processed data is fed into the proposed algorithm and the obtained result is analyzed. Some of the algorithms used in satellite imagery classification are U-Net, Random Forest, Deep Labv3, CNN, ANN, Resnet etc. In this project, we are using the DeepLabv3 (Atrous convolution) algorithm for land cover classification. The dataset used is the deep globe land cover classification dataset. DeepLabv3 is a semantic segmentation system that uses atrous convolution to capture multi-scale context by adopting multiple atrous rates in cascade or in parallel to determine the scale of segments.

Keywords: area calculation, atrous convolution, deep globe land cover classification, deepLabv3, land cover classification, resnet 50

Procedia PDF Downloads 117

Authors: Nithin Krisshna Gunasekaran, Prathima Prabhu Tumkur, Nicole Nazario Bayon, Krishnan Prabhakaran, Joseph C. Hall, Govindarajan T. Ramesh

Abstract:

Cerium oxide and turmeric have antioxidant properties, which have gained interest among researchers to study their applications in the field of biomedicine, such asanti-inflammatory, anticancer, and antimicrobial applications. In this study, the turmeric extract was prepared and mixed with cerium nitrate hexahydrate, stirred continuously to obtain a homogeneous solution and then heated on a hot plate to get the supernatant evaporated, then calcinated at 600°C to obtain the cerium oxide nanoparticles. Characterization of synthesized cerium oxide nanoparticles through Scanning Electron Microscopy determined the particle size to be in the range of 70 nm to 250 nm. Energy Dispersive X-Ray Spectroscopy determined the elemental composition of cerium and oxygen. Individual particles were identified through the characterization of cerium oxide nanoparticles using Field Emission Scanning Electron Microscopy, in which the particles were determined to be spherical and in the size of around 70 nm. The presence of cerium oxide was assured by analyzing the spectrum obtained through the characterization of cerium oxide nanoparticles by Fourier Transform Infrared Spectroscopy. The crystal structure of cerium oxide nanoparticles was determined to be face-centered cubic by analyzing the peaks obtained through theX-Ray Diffraction method. The crystal size of cerium oxide nanoparticles was determined to be around 13 nm by using the Debye Scherer equation. This study confirmed the synthesis of cerium oxide nanoparticles using turmeric extract.

Keywords: antioxidant, characterization, cerium oxide, synthesis, turmeric

Procedia PDF Downloads 131

Authors: Nicole Nazario Bayon, Prathima Prabhu Tumkur, Nithin Krisshna Gunasekaran, Krishnan Prabhakaran, Joseph C. Hall, Govindarajan T. Ramesh

Abstract:

Titanium nitride (TiN) nanoparticles have sparked interest over the past decade due to their characteristics such as thermal stability, extreme hardness, low production cost, and similar optical properties to gold. In this study, TiN nanoparticles were synthesized via a thermal benzene route to obtain a black powder of nanoparticles. The final product was drop cast onto conductive carbon tape and sputter coated with gold/palladium at a thickness of 4 nm for characterization by field emission scanning electron microscopy (FE-SEM) with energy dispersive X-Ray spectroscopy (EDX) that revealed they were spherical. ImageJ software determined the average size of the TiN nanoparticles was 79 nm in diameter. EDX revealed the elements present in the sample and showed no impurities. Further characterization by X-ray diffraction (XRD) revealed characteristic peaks of cubic phase titanium nitride, and crystallite size was calculated to be 14 nm using the Debye-Scherrer method. Dynamic light scattering (DLS) analysis revealed the size and size distribution of the TiN nanoparticles, with average size being 154 nm. Zeta potential concluded the surface of the TiN nanoparticles is negatively charged. Biocompatibility studies using MTT(3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) assay showed TiN nanoparticles are not cytotoxic at low concentrations (2, 5, 10, 25, 50, 75 mcg/well), and cell viability began to decrease at a concentration of 100 mcg/well.

Keywords: biocompatibility, characterization, cytotoxicity, nanoparticles, synthesis, titanium nitride

Procedia PDF Downloads 144

Authors: Gertrude I. Hewapathirana, Loi A. Nguyen, Mohammed M. Mostafa

Abstract:

Artificial intelligence (AI) technologies and tools are swiftly integrating into many functions of all organizations as a competitive drive to enhance innovations, productivity, efficiency, faster and precise decision making to keep up with rapid changes in the global business arena. Despite increasing research on AI technologies in production, manufacturing, and information management, AI in human resource disciplines is still lagging. Though a few research studies on HR informatics, recruitment, and HRM in general, how to integrate AI in other HR functional disciplines (e.g., compensation, training, mentoring and coaching, employee motivation) is rarely researched. Many inconsistencies of research hinder developing up-to-date knowledge on AI in HR disciplines. Therefore, exploring eight research questions, using bibliometric network analysis combined with a meta-analysis of published research literature. The authors attempt to generate knowledge on the role of AI in improving the efficiency of HR functional disciplines. To advance the knowledge for the benefit of researchers, academics, policymakers, and practitioners, the study highlights the types of AI innovations and outcomes, trends, gaps, themes and topics, fast-moving disciplines, key players, and future directions.AI in HR informatics in high tech firms is the dominant theme in many research publications. While there is increasing attention from researchers and practitioners, there are many gaps between the promise, potential, and real AI applications in HR disciplines. A higher knowledge gap raised many unanswered questions regarding legal, ethical, and morale aspects of AI in HR disciplines as well as the potential contributions of AI in HR disciplines that may guide future research directions. Though the study provides the most current knowledge, it is limited to peer-reviewed empirical, theoretical, and conceptual research publications stored in the WoS database. The implications for theory, practice, and future research are discussed.

Keywords: artificial intelligence, human resources, bibliometric analysis, research directions

Procedia PDF Downloads 67

Authors: Himanshu Sharma, Karthik Kumar, Harish Kumar

Abstract:

The major problem in crop cultivation is the occurrence of multiple crop diseases. During the growth stage, timely identification of crop diseases is paramount to ensure the high yield of crops, lower production costs, and minimize pesticide usage. In most cases, crop diseases produce observable characteristics and symptoms. The Surveyors usually diagnose crop diseases when they walk through the fields. However, surveyor inspections tend to be biased and error-prone due to the nature of the monotonous task and the subjectivity of individuals. In addition, visual inspection of each leaf or plant is costly, time-consuming, and labour-intensive. Furthermore, the plant pathologists and experts who can often identify the disease within the plant according to their symptoms in early stages are not readily available in remote regions. Therefore, this study specifically addressed early detection of leaf scald, red rot, and eyespot types of diseases within sugarcane plants. The study proposes a computer vision-based approach using a convolutional neural network (CNN) for automatic identification of crop diseases. To facilitate this, firstly, images of sugarcane diseases were taken from google without modifying the scene, background, or controlling the illumination to build the training dataset. Then, the testing dataset was developed based on the real-time collected images from the sugarcane field from India. Then, the image dataset is pre-processed for feature extraction and selection. Finally, the CNN-based Visual Geometry Group (VGG) model was deployed on the training and testing dataset to classify the images into diseased and healthy sugarcane plants and measure the model's performance using various parameters, i.e., accuracy, sensitivity, specificity, and F1-score. The promising result of the proposed model lays the groundwork for the automatic early detection of sugarcane disease. The proposed research directly sustains an increase in crop yield.

Keywords: automatic classification, computer vision, convolutional neural network, image processing, sugarcane disease, visual geometry group

Procedia PDF Downloads 92

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 79

Authors: James Rigor Camacho, Wansu Lim

Abstract:

Over the last few years, the development of new wearable and processing technologies has accelerated in order to harness physiological data such as electroencephalograms (EEGs) for EEG-based applications. EEG has been demonstrated to be a source of emotion recognition signals with the highest classification accuracy among physiological signals. However, when emotion recognition systems are used for real-time classification, the training unit is frequently left to run offline or in the cloud rather than working locally on the edge. That strategy has hampered research, and the full potential of using an edge AI device has yet to be realized. Edge AI devices are computers with high performance that can process complex algorithms. It is capable of collecting, processing, and storing data on its own. It can also analyze and apply complicated algorithms like localization, detection, and recognition on a real-time application, making it a powerful embedded device. The NVIDIA Jetson series, specifically the Jetson Nano device, was used in the implementation. The cEEGrid, which is integrated to the open-source brain computer-interface platform (OpenBCI), is used to collect EEG signals. An EEG-based real-time emotion recognition system on Edge AI is proposed in this paper. To perform graphical spectrogram categorization of EEG signals and to predict emotional states based on input data properties, machine learning-based classifiers were used. Until the emotional state was identified, the EEG signals were analyzed using the K-Nearest Neighbor (KNN) technique, which is a supervised learning system. In EEG signal processing, after each EEG signal has been received in real-time and translated from time to frequency domain, the Fast Fourier Transform (FFT) technique is utilized to observe the frequency bands in each EEG signal. To appropriately show the variance of each EEG frequency band, power density, standard deviation, and mean are calculated and employed. The next stage is to identify the features that have been chosen to predict emotion in EEG data using the K-Nearest Neighbors (KNN) technique. Arousal and valence datasets are used to train the parameters defined by the KNN technique.Because classification and recognition of specific classes, as well as emotion prediction, are conducted both online and locally on the edge, the KNN technique increased the performance of the emotion recognition system on the NVIDIA Jetson Nano. Finally, this implementation aims to bridge the research gap on cost-effective and efficient real-time emotion recognition using a resource constrained hardware device, like the NVIDIA Jetson Nano. On the cutting edge of AI, EEG-based emotion identification can be employed in applications that can rapidly expand the research and implementation industry's use.

Keywords: edge AI device, EEG, emotion recognition system, supervised learning algorithm, sensors

Procedia PDF Downloads 80

Authors: Ashish Soni, Suman Kalyan Pal

Abstract:

Transition metal dichalcogenides (TMDCs) are an emerging paradigm for the generation of advanced materials which are capable of utilizing in future device applications. In recent years TMDCs have attracted researchers for their unique band structure in monolayers. Large-area monolayers could become the most appropriate candidate for flexible and thin optoelectronic devices. For this purpose, it is crucial to understand the generation and transport of charge carriers in low dimensions. A deep understanding of photo-generated hot charges and trapped charges is essential to improve the performance of optoelectronic devices. Carrier trapping by the defect states that are introduced during the growth process of the monolayer could influence the dynamical behaviour of charge carriers. Herein, we investigated some aspects of the ultrafast evolution of the initially generated hot carriers and trapped charges in large-area monolayer WS₂ by measuring transient absorption at energies above and below the band gap energy. Our excitation density and energy-dependent measurements reveal the trapping of the initially generated charge carrier. Our results could be beneficial for the development of TMDC-based optoelectronic devices.

Keywords: transient absorption, optoelectronics, 2D materials, TMDCs, exciton

Procedia PDF Downloads 44