Search results for: deep residual networks

Authors: Selvavinayagan Babiharan

Abstract:

This research aims to develop a deep learning platform to enhance mathematics education for secondary students in Sri Lanka. The platform will be designed to incorporate interactive and user-friendly features to engage students in active learning and promote their mathematical skills. The proposed platform will be developed using TensorFlow and Keras, two widely used deep learning frameworks. The system will be trained on a large dataset of math problems, which will be collected from Sri Lankan school curricula. The results of this research will contribute to the improvement of mathematics education in Sri Lanka and provide a valuable tool for teachers to enhance the learning experience of their students.

Keywords: information technology, education, machine learning, mathematics

Procedia PDF Downloads 58

Authors: Mahdi Pirmoradian, Olayinka Adigun, Christos Politis

Abstract:

Cooperative spectrum sensing is a crucial challenge in cognitive radio networks. Cooperative sensing can increase the reliability of spectrum hole detection, optimize sensing time and reduce delay in cooperative networks. In this paper, an efficient central capacity optimization algorithm is proposed to minimize cooperative sensing time in a homogenous sensor network using OR decision rule subject to the detection and false alarm probabilities constraints. The evaluation results reveal significant improvement in the sensing time and normalized capacity of the cognitive sensors.

Keywords: cooperative networks, normalized capacity, sensing time

Procedia PDF Downloads 605

Authors: Boonsiri Dandumrongsin, Halis Simsek, Chaiwat Rongsayamanont

Abstract:

Dissolved organic nitrogen (DON) is known as one of the persistence nitrogenous pollutant being originated from secondary treated effluent of municipal sewage treatment plant. However, effect of key system operating condition on the fate and behavior of residual DON in the treated effluent is still not known. This study aims to investigate effect of organic loading rate (OLR) on the residual level of DON in the biofilm reactor effluent. Synthetic municipal wastewater was fed into moving bed biofilm reactors at OLR of 1.6x10-3 and 3.2x10-3 kg SCOD/m3-d. The results showed higher organic removal efficiency was found in the reactor operating at higher OLR. However, DON was observed at higher value in the effluent of the higher OLR reactor than that of the lower OLR reactor evidencing a clear influence of OLR on the residual DON level in the treated effluent of the biofilm reactors. It is possible that the lower DON being observed in the reactor at lower OLR is likely to be a result of providing the microbe with the additional period for utilizing the refractory DON molecules during operation at lower organic loading. All the experiments were repeated using raw wastewaters and similar trend was obtained.

Keywords: dissolved organic nitrogen, hydraulic retention time, moving bed biofilm reactor, soluble microbial products

Procedia PDF Downloads 257

Authors: Ole-Christian Galbo Engstrøm, Erik Schou Dreier, Birthe Møller Jespersen, Kim Steenstrup Pedersen

Abstract:

Grain quality analysis is a multi-parameterized problem that includes a variety of qualitative and quantitative parameters such as grain type classification, damage type classification, and nutrient regression. Currently, these parameters require human inspection, a multitude of instruments employing a variety of sensor technologies, and predictive model types or destructive and slow chemical analysis. This paper investigates the feasibility of applying near-infrared hyperspectral imaging (NIR-HSI) to grain quality analysis. For this study two datasets of NIR hyperspectral images in the wavelength range of 900 nm - 1700 nm have been used. Both datasets contain images of sparsely and densely packed grain kernels. The first dataset contains ~87,000 image crops of bulk wheat samples from 63 harvests where protein value has been determined by the FOSS Infratec NOVA which is the golden industry standard for protein content estimation in bulk samples of cereal grain. The second dataset consists of ~28,000 image crops of bulk grain kernels from seven different wheat varieties and a single rye variety. In the first dataset, protein regression analysis is the problem to solve while variety classification analysis is the problem to solve in the second dataset. Deep convolutional neural networks (CNNs) have the potential to utilize spatio-spectral correlations within a hyperspectral image to simultaneously estimate the qualitative and quantitative parameters. CNNs can autonomously derive meaningful representations of the input data reducing the need for advanced preprocessing techniques required for classical chemometric model types such as artificial neural networks (ANNs) and partial least-squares regression (PLS-R). A comparison between different CNN architectures utilizing 2D and 3D convolution is conducted. These results are compared to the performance of ANNs and PLS-R. Additionally, a variety of preprocessing techniques from image analysis and chemometrics are tested. These include centering, scaling, standard normal variate (SNV), Savitzky-Golay (SG) filtering, and detrending. The results indicate that the combination of NIR-HSI and CNNs has the potential to be the foundation for an automatic system unifying qualitative and quantitative grain quality analysis within a single sensor technology and predictive model type.

Keywords: deep learning, grain analysis, hyperspectral imaging, preprocessing techniques

Procedia PDF Downloads 75

Authors: I. D. Laryushin, V. A. Kostin, A. A. Silaev, N. V. Vvedenskii

Abstract:

Generation of broadband intense terahertz (THz) radiation attracts reasonable interest due to various applications, such as the THz time-domain spectroscopy, the probing and control of various ultrafast processes, the THz imaging with subwavelength resolution, and many others. One of the most promising methods for generating powerful and broadband terahertz pulses is based on focusing two-color femtosecond ionizing laser pulses in gases, including ambient air. For this method, the amplitudes of terahertz pulses are determined by the free-electron current density remaining in a formed plasma after the passage of the laser pulse. The excitation of this residual current density can be treated as multi-wave mixing: Аn effective generation of terahertz radiation is possible only when the frequency ratio of one-color components in the two-color pulse is close to irreducible rational fraction a/b with small odd sum a + b. This work focuses on the optimal parameters (polarizations and intensities) of laser components for the strongest THz generation. The optimal values of parameters are found numerically and analytically with the use of semiclassical approach for calculating the residual current density. For frequency ratios close to a/(a ± 1) with natural a, the strongest THz generation is shown to take place when the both laser components have circular polarizations and equal intensities. For this optimal case, an analytical formula for the residual current density was derived. For the frequency ratios such as 2/5, the two-color ionizing pulses with circularly polarized components practically do not excite the residual current density. However, the optimal parameters correspond generally to specific elliptical (not linear) polarizations of the components and intensity ratios close to unity.

Keywords: broadband terahertz radiation, ionization, laser plasma, ultrashort two-color pulses

Procedia PDF Downloads 186

Authors: Degale Desta, Cheng Jian

Abstract:

The most important factor in identifying a person is their face. Even identical twins have their own distinct faces. As a result, identification and face recognition are needed to tell one person from another. A face recognition system is a verification tool used to establish a person's identity using biometrics. Nowadays, face recognition is a common technique used in a variety of applications, including home security systems, criminal identification, and phone unlock systems. This system is more secure because it only requires a facial image instead of other dependencies like a key or card. Face detection and face identification are the two phases that typically make up a human recognition system.The idea behind designing and creating a face recognition system using deep learning with Azure ML Python's OpenCV is explained in this paper. Face recognition is a task that can be accomplished using deep learning, and given the accuracy of this method, it appears to be a suitable approach. To show how accurate the suggested face recognition system is, experimental results are given in 98.46% accuracy using Fast-RCNN Performance of algorithms under different training conditions.

Keywords: deep learning, face recognition, identification, fast-RCNN

Procedia PDF Downloads 110

Authors: Jacob Green, Cecilia Cabrera, Maximilian Jakobs, Andrea Dimitracopoulos, Mark van der Wilk, Ryan Greenhalgh

Abstract:

Predicting drug properties is key in drug discovery to enable de-risking of assets before expensive clinical trials and to find highly active compounds faster. Interest from the machine learning community has led to the release of a variety of benchmark datasets and proposed methods. However, it remains unclear for practitioners which method or approach is most suitable, as different papers benchmark on different datasets and methods, leading to varying conclusions that are not easily compared. Our large-scale empirical study links together numerous earlier works on different datasets and methods, thus offering a comprehensive overview of the existing property classes, datasets, and their interactions with different methods. We emphasise the importance of uncertainty quantification and the time and, therefore, cost of applying these methods in the drug development decision-making cycle. To the best of the author's knowledge, it has been observed that the optimal approach varies depending on the dataset and that engineered features with classical machine learning methods often outperform deep learning. Specifically, QSAR datasets are typically best analysed with classical methods such as Gaussian Processes, while ADMET datasets are sometimes better described by Trees or deep learning methods such as Graph Neural Networks or language models. Our work highlights that practitioners do not yet have a straightforward, black-box procedure to rely on and sets a precedent for creating practitioner-relevant benchmarks. Deep learning approaches must be proven on these benchmarks to become the practical method of choice in drug property prediction.

Keywords: activity (QSAR), ADMET, classical methods, drug property prediction, empirical study, machine learning

Procedia PDF Downloads 52

Authors: Donghee Noh, Seonhyeong Kim, Junhwan Choi, Heegon Kim, Sooho Jung, Keunho Park

Abstract:

In this paper, we describe an image post-processing method to enhance the performance of the parcel segmentation method using deep learning-based aerial images conducted in previous studies. The study results were evaluated using a confusion matrix, IoU, Precision, Recall, and F1-Score. In the case of the confusion matrix, it was observed that the false positive value, which is the result of misclassification, was greatly reduced as a result of image post-processing. The average IoU was 0.9688 in the image post-processing, which is higher than the deep learning result of 0.8362, and the F1-Score was also 0.9822 in the image post-processing, which was higher than the deep learning result of 0.8850. As a result of the experiment, it was found that the proposed technique positively complements the deep learning results in segmenting the parcel of interest.

Keywords: aerial image, image process, machine vision, open field smart farm, segmentation

Procedia PDF Downloads 56

Authors: Aisulu Tileukulova, Zhexebay Dauren

Abstract:

Radar plays an important role because it is widely used in civil and military fields. Target detection is one of the most important radar applications. The accuracy of detecting inconspicuous aerial objects in radar facilities is lower against the background of noise. Convolutional neural networks can be used to improve the recognition of this type of aerial object. The purpose of this work is to develop an algorithm for recognizing aerial objects using convolutional neural networks, as well as training a neural network. In this paper, the structure of a convolutional neural network (CNN) consists of different types of layers: 8 convolutional layers and 3 layers of a fully connected perceptron. ReLU is used as an activation function in convolutional layers, while the last layer uses softmax. It is necessary to form a data set for training a neural network in order to detect a target. We built a Confusion Matrix of the CNN model to measure the effectiveness of our model. The results showed that the accuracy when testing the model was 95.7%. Classification of echo signals using CNN shows high accuracy and significantly speeds up the process of predicting the target.

Keywords: radar, neural network, convolutional neural network, echo signals

Procedia PDF Downloads 320

Authors: Dongjoo Kim

Abstract:

Effective control of sloshing in a liquid container is an important issue to be resolved in many applications. In this study, numerical simulations are performed to design the velocity profile of rectangular container and investigate the effectiveness of input shaping for sloshing control. Trapezoidal profiles of container velocity are chosen to be reference commands and they are convolved with a series of impulses to generate shaped ones that induce minimal residual oscillations. The performances of several input shapers are compared from the viewpoint of transient peak and residual oscillations of sloshing. Results show that sloshing can be effectively controlled by input shaping (Supported by the NRF programs, NRF-2015R1D1A1A01059675, of Korean government).

Keywords: input shaping, rectangular container, sloshing, trapezoidal profile

Procedia PDF Downloads 236

Authors: Sorin Valcan, Mihail Gaianu

Abstract:

Labeling is a very costly and time consuming process which aims to generate datasets for training neural networks in several functionalities and projects. For driver monitoring system projects, the need for labeled images has a significant impact on the budget and distribution of effort. This paper presents the modifications done to an algorithm used for the generation of ground truth data for 2D eyes location on infrared images with drivers in order to improve the quality of the data and performance of the trained neural networks. The algorithm restrictions become tougher, which makes it more accurate but also less constant. The resulting dataset becomes smaller and shall not be altered by any kind of manual label adjustment before being used in the neural networks training process. These changes resulted in a much better performance of the trained neural networks.

Keywords: labeling automation, infrared camera, driver monitoring, eye detection, convolutional neural networks

Procedia PDF Downloads 89

Authors: Edgardo V. Gubatanga Jr., Mark Joshua Salvacion

Abstract:

Most aerial LiDAR systems have accompanying aerial cameras in order to capture not only the terrain of the surveyed area but also its true-color appearance. However, the presence of atmospheric clouds, poor lighting conditions, and aerial camera problems during an aerial survey may cause absence of aerial photographs. These leave areas having terrain information but lacking aerial photographs. Intensity images can be derived from LiDAR data but they are only grayscale images. A deep learning model is developed to create a complex function in a form of a deep neural network relating the pixel values of LiDAR-derived intensity images and true-color images. This complex function can then be used to predict the true-color images of a certain area using intensity images from LiDAR data. The predicted true-color images do not necessarily need to be accurate compared to the real world. They are only intended to look realistic so that they can be used as base maps.

Keywords: aerial LiDAR, colorization, deep learning, intensity images

Procedia PDF Downloads 138

Authors: Harshadkumar Rajgor, Jeff Butterworth

Abstract:

Background: Barretts oesophagus increases the risk of developing oesophageal adenocarcinoma. Over the last 40 years, there has been a 6 fold increase in the incidence of oesophageal adenocarcinoma in the western world and the incidence rates are increasing at a greater rate than cancers of the colon, breast and lung. Endoscopic mucosal resection (EMR) is a relatively new technique being used by 2 centres in the greater midlands cancer network. EMR can be used for curative or staging purposes, for high-grade dysplasia’s and T1 tumours of the oesophagus. EMR is also suitable for those who are deemed high risk for oesophagectomy. EMR has a recurrence rate of 21% according to the Wiesbaden data. Method: A retrospective study of prospectively collected data was carried out involving 24 patients who had EMR for curative or staging purposes. Complications of residual or recurrent disease following EMR that required further treatment were investigated. Results: In 54% of cases residual or recurrent disease was suspected. 96% of patients were given clear and concise information regarding their diagnosis of high-grade dysplasia or T1 tumours. All 24 patients consulted the same specialist healthcare team. Conclusion: EMR is a safe and effective treatment for patients who have high-grade dysplasia and T1NO tumours. In 54% of cases residual or recurrent disease was suspected. Initially, only single resections were undertaken. Multiple resections are now being carried out to reduce the risk of recurrence. Complications from EMR remain low in this series and consisted of a single episode of post procedural bleeding.

Keywords: endoscopic mucosal resection, oesophageal dysplasia, T1 tumours, cancer network

Procedia PDF Downloads 299

Authors: Nabil Shawkat, Jamil Saquer

Abstract:

Social media has accelerated our ability to engage with others and eliminated many communication barriers. On the other hand, the widespread use of social media resulted in an increase in online hate speech. This has drastic impacts on vulnerable individuals and societies. Therefore, it is critical to detect hate speech to prevent innocent users and vulnerable communities from becoming victims of hate speech. We investigate the performance of different deep learning and machine learning algorithms on three different datasets. Our results show that the BERT model gives the best performance among all the models by achieving an F1-score of 90.6% on one of the datasets and F1-scores of 89.7% and 88.2% on the other two datasets.

Keywords: hate speech, machine learning, deep learning, abusive words, social media, text classification

Procedia PDF Downloads 109

Authors: Kumar Divyendra

Abstract:

Underwater Sensor Networks (UWSNs) have wide applications in areas such as water quality monitoring, marine wildlife management etc. A typical UWSN system consists of a set of sensors deployed randomly underwater which communicate with each other using acoustic links. RF communication doesn't work underwater, and GPS too isn't available underwater. Additionally Automated Underwater Vehicles (AUVs) are deployed to collect data from some special nodes called Cluster Heads (CHs). These CHs aggregate data from their neighboring nodes and forward them to the AUVs using optical links when an AUV is in range. This helps reduce the number of hops covered by data packets and helps conserve energy. We consider the three-dimensional model of the UWSN. Nodes are initially deployed randomly underwater. They attach themselves to the surface using a rod and can only move upwards or downwards using a pump and bladder mechanism. We use graph theory concepts to maximize the coverage volume while every node maintaining connectivity with at least one surface node. We treat the surface nodes as landmarks and each node finds out its hop distance from every surface node. We treat these hop-distances as coordinates and use them for AUV navigation. An AUV intending to move closer to a node with given coordinates moves hop by hop through nodes that are closest to it in terms of these coordinates. In absence of GPS, multiple different approaches like Inertial Navigation System (INS), Doppler Velocity Log (DVL), computer vision-based navigation, etc., have been proposed. These systems have their own drawbacks. INS accumulates error with time, vision techniques require prior information about the environment. We propose a method that makes use of the earth's magnetic field values for navigation and combines it with other methods that simultaneously increase the coverage volume under the UWSN. The AUVs are fitted with magnetometers that measure the magnetic intensity (I), horizontal inclination (H), and Declination (D). The International Geomagnetic Reference Field (IGRF) is a mathematical model of the earth's magnetic field, which provides the field values for the geographical coordinateson earth. Researchers have developed an inverse deep learning model that takes the magnetic field values and predicts the location coordinates. We make use of this model within our work. We combine this with with the hop-by-hop movement described earlier so that the AUVs move in such a sequence that the deep learning predictor gets trained as quickly and precisely as possible We run simulations in MATLAB to prove the effectiveness of our model with respect to other methods described in the literature.

Keywords: clustering, deep learning, network backbone, parallel computing

Procedia PDF Downloads 75

Authors: Dipti Yadav, Latha Rangan, Pinakeswar Mahanta

Abstract:

Recent fuel-development strategies to reduce oil dependency, mitigate greenhouse gas emissions, and utilize domestic resources have generated interest in the search for alternative sources of fuel supplies. Bioenergy production from lignocellulosic biomass has a great potential. Cellulose, hemicellulose and Lignin are main constituent of woods or agrowaste. In all the industries there are always left over or waste products mainly lignin, due to the heterogeneous nature of wood and pulp fibers and the heterogeneity that exists between individual fibers, no method is currently available for the quantitative isolation of native or residual lignin without the risk of structural changes during the isolation. The potential benefits from finding alternative uses of lignin are extensive, and with a double effect. Lignin can be used to replace fossil-based raw materials in a wide range of products, from plastics to individual chemical products, activated carbon, motor fuels and carbon fibers. Furthermore, if there is a market for lignin for such value-added products, the mills will also have an additional economic incentive to take measures for higher energy efficiency. In this study residual lignin were isolated from areca nut shells by acid hydrolysis and were analyzed and characterized by Fourier Transform Infrared (FTIR), LCMS and complexity of its structure investigated by NMR.

Keywords: Areca nut, Lignin, wood, bioenergy

Procedia PDF Downloads 454

Authors: Shafique Qadir Memon, Muhammad Safar Mirjat, Abdul Quadir Mughal, Nadeem Amjad

Abstract:

The aims of this study were to determine direct input energy and indirect energy in maize production, to evaluate the inputs energy consumption and outputs energy gained for maize production in Islamabad, Pakistan for spring 2013. Results showed that grain yield was maximum under deep tillage as compared to conventional and zero tillage. Total energy input/output were maximum in deep tillage as compared to conventional tillage while lowest in zero tillage, net energy gain were found maximum under deep tillage.

Keywords: tillage, energy, grain yield, net energy gain

Procedia PDF Downloads 439

Authors: Min Jin Kwon, Seung Won Kim, Eui Yeon Kim, Haeyoung Lee

Abstract:

Numerous anonymous chat apps that help people to connect with random strangers have been released in South Korea. However, they become a serious problem for young people since young people often use them for channels of prostitution or sexual violence. Although ISPs in South Korea are responsible for making inappropriate content inaccessible on their networks, they do not block traffic of random chat apps since 1) the use of random chat apps is entirely legal. 2) it is reported that they use HTTP proxy blocking so that non-HTTP traffic cannot be blocked. In this paper, we propose a service model that can block random chat apps at home routers. A service provider manages a blacklist that contains blocked apps’ information. Home routers that subscribe the service filter the traffic of the apps out using deep packet inspection. We have implemented a prototype of the proposed model, including a centralized server providing the blacklist, a Raspberry Pi-based home router that can filter traffic of the apps out, and an Android app used by the router’s administrator to locally customize the blacklist.

Keywords: deep packet inspection, internet filtering, juvenile protection, technical blocking

Procedia PDF Downloads 324

Authors: G. Blugan, B. Jiang, J. Thornberry, P. Sturzenegger, U. Gonzenbach, M. Misson, D. Cartlidge, R. Stenerud, J. Kuebler

Abstract:

Low-cost ceramic spheres were developed and manufactured from the engineering ceramic aluminium oxide. Hollow spheres of 50 mm diameter with a wall thickness of 0.5-1.0 mm were produced via an adapted slip casting technique. It was possible to produce the spheres with good repeatability and with no defects or failures in the spheres due to the manufacturing process. The spheres were developed specifically for use in buoyancy devices for deep-sea exploration conditions at depths of 3000 m below sea level. The spheres with a 1.0 mm wall thickness exhibit a buoyancy of over 54% while the spheres with a 0.5 mm wall thickness exhibit a buoyancy of over 73%. The mechanical performance of the spheres was confirmed by performing a hydraulic burst pressure test on individual spheres. With a safety factor of 3, all spheres with 1.0 mm wall thickness survived a hydraulic pressure of greater than 150 MPa which is equivalent to a depth of more than 5000 m below sea level. The spheres were then incorporated into a buoyancy module. These hollow aluminium oxide ceramic spheres offer an excellent possibility of deep-sea exploration to depths greater than the currently used technology.

Keywords: buoyancy, ceramic spheres, deep-sea, oil exploration

Procedia PDF Downloads 396

Authors: Yohannes Bisa Biramo

Abstract:

This study examined the meditational role of surface acting and deep acting on the relationship between calling orientation and work engagement of teachers in secondary schools of Wolaita Zone, Wolaita, Ethiopia. A predictive non-experimental correlational design was performed among 300 secondary school teachers. Stratified random sampling followed by a systematic random sampling technique was used as the basis for selecting samples from the target population. To analyze the data, Structural Equation Modeling (SEM) was used to test the association between the independent variables and the dependent variables. Furthermore, the goodness of fit of the study variables was tested using SEM to see and explain the path influence of the independent variable on the dependent variable. Confirmatory factor analysis (CFA) was conducted to test the validity of the scales in the study and to assess the measurement model fit indices. The analysis result revealed that calling was significantly and positively correlated with surface acting, deep acting and work engagement. Similarly, surface acting was significantly and positively correlated with deep acting and work engagement. And also, deep acting was significantly and positively correlated with work engagement. With respect to mediation analysis, the result revealed that surface acting mediated the relationship between calling and work engagement and also deep acting mediated the relationship between calling and work engagement. Besides, by using the model of the present study, the school leaders and practitioners can identify a core area to be considered in recruiting and letting teachers teach, in giving induction training for newly employed teachers and in performance appraisal.

Keywords: calling, surface acting, deep acting, work engagement, mediation, teachers

Procedia PDF Downloads 55

Authors: Naim Karasekreter, Ugur Fidan, Fatih Basciftci

Abstract:

Wireless sensor networks are networks in which more than one sensor node is organized among themselves. The working principle is based on the transfer of the sensed data over the other nodes in the network to the central station. Wireless sensor networks concentrate on routing algorithms, energy efficiency and clustering algorithms. In the clustering method, the nodes in the network are divided into clusters using different parameters and the most suitable cluster head is selected from among them. The data to be sent to the center is sent per cluster, and the cluster head is transmitted to the center. With this method, the network traffic is reduced and the energy efficiency of the nodes is increased. In this study, clustering algorithms were examined in terms of clustering performances and cluster head selection characteristics to try to identify weak and strong sides. This work is supported by the Project 17.Kariyer.123 of Afyon Kocatepe University BAP Commission.

Keywords: wireless sensor networks (WSN), clustering algorithm, cluster head, clustering

Procedia PDF Downloads 477

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 117

Authors: Tiago do Carmo Nogueira, Cássio Dener Noronha Vinhal, Gélson da Cruz Júnior, Matheus Rudolfo Diedrich Ullmann

Abstract:

Generating automatic image descriptions through natural language is a challenging task. Image captioning is a task that consistently describes an image by combining computer vision and natural language processing techniques. To accomplish this task, cutting-edge models use encoder-decoder structures. Thus, Convolutional Neural Networks (CNN) are used to extract the characteristics of the images, and Recurrent Neural Networks (RNN) generate the descriptive sentences of the images. However, cutting-edge approaches still suffer from problems of generating incorrect captions and accumulating errors in the decoders. To solve this problem, we propose a model based on the encoder-decoder structure, introducing a module that generates the weights according to the importance of the word to form the sentence, using the part-of-speech (PoS). Thus, the results demonstrate that our model surpasses state-of-the-art models.

Keywords: gated recurrent units, caption generation, convolutional neural network, part-of-speech

Procedia PDF Downloads 75

Authors: Khalak Bin Khair, Saqib Jahir, Mohammed Ibrahim, Fahad Bin, Debajyoti Karmaker

Abstract:

Object detection is a computer technology that deals with searching through digital images and videos for occurrences of semantic elements of a particular class. It is associated with image processing and computer vision. On top of object detection, we detect camouflage objects within an image using Deep Learning techniques. Deep learning may be a subset of machine learning that's essentially a three-layer neural network Over 6500 images that possess camouflage properties are gathered from various internet sources and divided into 4 categories to compare the result. Those images are labeled and then trained and tested using vgg16 architecture on the jupyter notebook using the TensorFlow platform. The architecture is further customized using Transfer Learning. Methods for transferring information from one or more of these source tasks to increase learning in a related target task are created through transfer learning. The purpose of this transfer of learning methodologies is to aid in the evolution of machine learning to the point where it is as efficient as human learning.

Keywords: deep learning, transfer learning, TensorFlow, camouflage, object detection, architecture, accuracy, model, VGG16

Procedia PDF Downloads 112

Authors: Ajay Kumar Choubey, Geeta Agnihotri, C. Sasikumar, Rashmi Dwivedi

Abstract:

Deep-drawing operations are performed widely in industrial applications. It is very important for efficiency to achieve parts with no or minimum defects. Deep drawn parts are used in high performance, high strength and high reliability applications where tension, stress, load and human safety are critical considerations. Wrinkling is a kind of defect caused by stresses in the flange part of the blank during metal forming operations. To avoid wrinkling appropriate blank-holder pressure/force or drawbead can be applied. Now-a-day computer simulation plays a vital role in the field of manufacturing process. So computer simulation of manufacturing has much advantage over previous conventional process i.e. mass production, good quality of product, fast working etc. In this study, a two dimensional elasto-plastic Finite Element (F.E.) model for Mild Steel material blank has been developed to study the behavior of the flange wrinkling and deep drawing parameters under different Blank-Holder Pressure (B.H.P.). For this, commercially available Finite Element software ANSYS 14 has been used in this study. Simulation results are critically studied and salient conclusions have been drawn.

Keywords: ANSYS, deep drawing, BHP, finite element simulation, wrinkling

Procedia PDF Downloads 435

Authors: Ina Nasto, Denada Sota, Pudrila Haskoçelaj, Mariola Ismailaj, Hajdar Kicaj

Abstract:

Most of the fishing areas of the Mediterranean Sea are considered to be overfished, consequently fishing has decreased or is static. Considering the continuous increase in demand for fish, the option of aquaculture production has had a growing development in recent decades. The environmental impact of aquaculture in the marine ecosystem has been a subject of study for several years in the Mediterranean. In the case of the Albanian waters, and in particular the Gulf of Vlora, have had a progressive growing of aquaculture activity in the last twenty years. Given the convenient and secluded location for tourist activities, the bay of Ragusa was considered as the most suitable area to install the aquaculture cage system for the breeding of sea bass and sea bream. The impact of aquaculture in on the soft bottom benthic communities has been assessed at the biggest commercial fish farm (Alb-Adriatico Sh.P.K) established in coastal waters of Ragusa bay 30–50 m deep, in the southern part of the Gulf of Vlora. In order to determine if there is a possible impact on the aquaculture cage in benthic communities, a comparative analysis was undertaken between transects and samples with differences in distances between them and with a gradient of distance from the fish cages. A total of 275 taxa were identified (1 Foraminifera, 1 Porifera, 3 Cnidaria, 2 Platyhelminthes, 2 Nemertea, 1 Bryozoa, 171 Mollusca, 39 Annelida, 35 Crustacea, 14 Echinodermata, 1 Hemichordata, and 5 Tunicata). The anaysis showed three main habitats in the area: biocoenosis of terrigenous mud, residual areas with Possidonia oceanica and also residual assemblages of algal coralligenous. Four benthic biotic indexes were calculated (Shannon H ’, BENTIX, Simpson's Diversity and Peilou’s J’) also benthic indicators as total abundance, number of taxa and species frequency to evaluate possible ecological impact of fish cages in Ragusa bay.

Keywords: Bentix index, Benthic community, invertebrates, aquaculture, Raguza bay

Procedia PDF Downloads 75

Authors: Seyed Alireza Mirmohammadi

Abstract:

Tehran Metro vendors are one of the working minorities in the capital, which is an essential cross-cultural case study. Today, with difficult economic conditions, subway vendors are increasing. Tehran metro vendors are in daily contact with many people in different metro stations. Due to the ban on their activities in this place and sometimes the humiliating look of some people, they experience special conditions compared to other people in the community. One of the most critical sources of shaping people's mentality toward their social status and identity in the media and, in the meantime, virtual social networks, due to various communication facilities such as Dualism and the possibility of high activity of users have a special place. Statistics have shown that virtual social networks have become an indispensable source of communication, information, and entertainment today. In this study, 15 semi-structured interviews were conducted with 15 metro vendors in Tehran about their membership in various virtual social networks and their mental perception of using them. The research results indicate that the obtained mentality of metro peddlers towards themselves is negative in virtual social networks, and they do not receive a good image of themselves in these networks.

Keywords: metro, tehran, intercultural communication, metro vendors, self image

Procedia PDF Downloads 84

Authors: Hesheng Wang, Haoyu Wang, Chungang Zhuang

Abstract:

Estimating the 6D pose of objects is a core step for robot bin-picking tasks. The problem is that various objects are usually randomly stacked with heavy occlusion in real applications. In this work, we propose a method to regress 6D poses by predicting three points for each object in the 3D point cloud through deep learning. To solve the ambiguity of symmetric pose, we propose a labeling method to help the network converge better. Based on the predicted pose, an iterative method is employed for pose optimization. In real-world experiments, our method outperforms the classical approach in both precision and recall.

Keywords: pose estimation, deep learning, point cloud, bin-picking, 3D computer vision

Procedia PDF Downloads 140

Authors: Soofiyan Atar, Adil Shaikh, Sahil Rajpurkar, Pragnesh Bhalala, Aniket Desai, Irfan Siddavatam

Abstract:

Deep reinforcement learning (deep RL) algorithms leverage the symbolic power of complex controllers by automating it by mapping sensory inputs to low-level actions. Deep RL eliminates the complex robot dynamics with minimal engineering. Deep RL provides high-risk involvement by directly implementing it in real-world scenarios and also high sensitivity towards hyperparameters. Tuning of hyperparameters on a pneumatic quadruped robot becomes very expensive through trial-and-error learning. This paper presents an automated learning control for a pneumatic quadruped robot using sample efficient deep Q learning, enabling minimal tuning and very few trials to learn the neural network. Long training hours may degrade the pneumatic cylinder due to jerk actions originated through stochastic weights. We applied this method to the pneumatic quadruped robot, which resulted in a hopping gait. In our process, we eliminated the use of a simulator and acquired a stable gait. This approach evolves so that the resultant gait matures more sturdy towards any stochastic changes in the environment. We further show that our algorithm performed very well as compared to programmed gait using robot dynamics.

Keywords: model-based reinforcement learning, gait stability, supervised learning, pneumatic quadruped

Procedia PDF Downloads 287

Authors: Avinash Malladhi

Abstract:

Forensic accounting is a specialized field that involves the application of accounting principles, investigative skills, and legal knowledge to detect and prevent fraud. With the rise of big data and technological advancements, artificial intelligence (AI) and machine learning (ML) algorithms have emerged as powerful tools for forensic accountants to enhance their fraud detection capabilities. In this paper, we review and analyze various AI/ML algorithms that are commonly used in forensic accounting, including supervised and unsupervised learning, deep learning, natural language processing Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Support Vector Machines (SVMs), Decision Trees, and Random Forests. We discuss their underlying principles, strengths, and limitations and provide empirical evidence from existing research studies demonstrating their effectiveness in detecting financial fraud. We also highlight potential ethical considerations and challenges associated with using AI/ML in forensic accounting. Furthermore, we highlight the benefits of these technologies in improving fraud detection and prevention in forensic accounting.

Keywords: AI, machine learning, forensic accounting & fraud detection, anti money laundering, Benford's law, fraud triangle theory

Procedia PDF Downloads 64