Search results for: deep capsule network
6180 Deep Neural Networks for Restoration of Sky Images Affected by Static and Anisotropic Aberrations
Authors: Constanza A. Barriga, Rafael Bernardi, Amokrane Berdja, Christian D. Guzman
Abstract:
Most image restoration methods in astronomy rely upon probabilistic tools that infer the best solution for a deconvolution problem. They achieve good performances when the point spread function (PSF) is spatially invariable in the image plane. However, this latter condition is not always satisfied with real optical systems. PSF angular variations cannot be evaluated directly from the observations, neither be corrected at a pixel resolution. We have developed a method for the restoration of images affected by static and anisotropic aberrations using deep neural networks that can be directly applied to sky images. The network is trained using simulated sky images corresponding to the T-80 telescope optical system, an 80 cm survey imager at Cerro Tololo (Chile), which are synthesized using a Zernike polynomial representation of the optical system. Once trained, the network can be used directly on sky images, outputting a corrected version of the image, which has a constant and known PSF across its field-of-view. The method was tested with the T-80 telescope, achieving better results than with PSF deconvolution techniques. We present the method and results on this telescope.Keywords: aberrations, deep neural networks, image restoration, variable point spread function, wide field images
Procedia PDF Downloads 1286179 A Comprehensive Study of Camouflaged Object Detection Using Deep Learning
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 1316178 Integrating Wound Location Data with Deep Learning for Improved Wound Classification
Authors: Mouli Banga, Chaya Ravindra
Abstract:
Wound classification is a crucial step in wound diagnosis. An effective classifier can aid wound specialists in identifying wound types with reduced financial and time investments, facilitating the determination of optimal treatment procedures. This study presents a deep neural network-based based classifier that leverages wound images and their corresponding locations to categorize wounds into various classes, such as diabetic, pressure, surgical, and venous ulcers. By incorporating a developed body map, the process of tagging wound locations is significantly enhanced, providing healthcare specialists with a more efficient tool for wound analysis. We conducted a comparative analysis between two prominent convolutional neural network models, ResNet50 and MobileNetV2, utilizing a dataset of 730 images. Our findings reveal that the RestNet50 outperforms MovileNetV2, achieving an accuracy of approximately 90%, compared to MobileNetV2’s 83%. This disparity highlights the superior capability of ResNet50 in the context of this dataset. The results underscore the potential of integrating deep learning with spatial data to improve the precision and efficiency of wound diagnosis, ultimately contributing to better patient outcomes and reducing healthcare costs.Keywords: wound classification, MobileNetV2, ResNet50, multimodel
Procedia PDF Downloads 126177 GA3C for Anomalous Radiation Source Detection
Authors: Chia-Yi Liu, Bo-Bin Xiao, Wen-Bin Lin, Hsiang-Ning Wu, Liang-Hsun Huang
Abstract:
In order to reduce the risk of radiation damage that personnel may suffer during operations in the radiation environment, the use of automated guided vehicles to assist or replace on-site personnel in the radiation environment has become a key technology and has become an important trend. In this paper, we demonstrate our proof of concept for autonomous self-learning radiation source searcher in an unknown environment without a map. The research uses GPU version of Asynchronous Advantage Actor-Critic network (GA3C) of deep reinforcement learning to search for radiation sources. The searcher network, based on GA3C architecture, has self-directed learned and improved how search the anomalous radiation source by training 1 million episodes under three simulation environments. In each episode of training, the radiation source position, the radiation source intensity, starting position, are all set randomly in one simulation environment. The input for searcher network is the fused data from a 2D laser scanner and a RGB-D camera as well as the value of the radiation detector. The output actions are the linear and angular velocities. The searcher network is trained in a simulation environment to accelerate the learning process. The well-performance searcher network is deployed to the real unmanned vehicle, Dashgo E2, which mounts LIDAR of YDLIDAR G4, RGB-D camera of Intel D455, and radiation detector made by Institute of Nuclear Energy Research. In the field experiment, the unmanned vehicle is enable to search out the radiation source of the 18.5MBq Na-22 by itself and avoid obstacles simultaneously without human interference.Keywords: deep reinforcement learning, GA3C, source searching, source detection
Procedia PDF Downloads 1036176 Relative Toxicity of Apparent Pesticides against Safflower Capsule Fly, Acanthiophilus helianthi Rossi (Diptera: Tephritidae) under Laboratory Conditions
Authors: Karim Saeidi
Abstract:
Safflower capsule fly, Acanthiophilus helianthi Rossi (Diptera: Tephritidae), is a key pest of safflower in Iran. The toxicity of Methidathion, Malathion, Deltamethrin, and Lufenuron to adult males and females of Acanthiophilus helianthi was studied under laboratory conditions. Malathion was the most toxic among the tested compounds followed by Methidathion, Lufenuron, and Deltamethrin to Acanthiophilus helianthi at 24 h post treatment, the respective LC50 values were 0.40 ppm, 0.68 ppm, 10.99 ppm, and 11.75 ppm for males and 0.46 ppm, 0.97 ppm, 13.45 ppm, and 16.32 ppm for females. At 48 h post treatment, Malathion was the most toxic followed by Methidathion, Deltamethrin, and Lufenuron to Acanthiophilus helianthi, LC50 values were 0.08 ppm, 0.54 ppm, 1.80 ppm, and 1.96 ppm for males and 0.34 ppm, 0.64 ppm, 1.88 ppm, and 2.37 ppm for females. At 72 h post treatment, Malathion was the most toxic followed by Methidathion, Lufenuron, and Deltamethrin to Acanthiophilus helianthi LC50 values were 0.04 ppm, 0.33 ppm, 0.44 ppm, and 0.71 ppm for males and 0.09 ppm, 0.36 ppm, 0.75 ppm, and 0.82 ppm for females. It is observed that LC50 values for treated adult females increased more than in the treated adult males at 24 h, 48 h, and 72 h post treatment. It means that the adult males were more susceptible to the tested insecticides than the adult females.Keywords: safflower, Methidathion, Deltamethrin, Lufenuron, Malathion, Tephritidae, safflower capsule fly, Acanthiophilus helianthi
Procedia PDF Downloads 3266175 Physics Informed Deep Residual Networks Based Type-A Aortic Dissection Prediction
Abstract:
Purpose: Acute Type A aortic dissection is a well-known cause of extremely high mortality rate. A highly accurate and cost-effective non-invasive predictor is critically needed so that the patient can be treated at earlier stage. Although various CFD approaches have been tried to establish some prediction frameworks, they are sensitive to uncertainty in both image segmentation and boundary conditions. Tedious pre-processing and demanding calibration procedures requirement further compound the issue, thus hampering their clinical applicability. Using the latest physics informed deep learning methods to establish an accurate and cost-effective predictor framework are amongst the main goals for a better Type A aortic dissection treatment. Methods: Via training a novel physics-informed deep residual network, with non-invasive 4D MRI displacement vectors as inputs, the trained model can cost-effectively calculate all these biomarkers: aortic blood pressure, WSS, and OSI, which are used to predict potential type A aortic dissection to avoid the high mortality events down the road. Results: The proposed deep learning method has been successfully trained and tested with both synthetic 3D aneurysm dataset and a clinical dataset in the aortic dissection context using Google colab environment. In both cases, the model has generated aortic blood pressure, WSS, and OSI results matching the expected patient’s health status. Conclusion: The proposed novel physics-informed deep residual network shows great potential to create a cost-effective, non-invasive predictor framework. Additional physics-based de-noising algorithm will be added to make the model more robust to clinical data noises. Further studies will be conducted in collaboration with big institutions such as Cleveland Clinic with more clinical samples to further improve the model’s clinical applicability.Keywords: type-a aortic dissection, deep residual networks, blood flow modeling, data-driven modeling, non-invasive diagnostics, deep learning, artificial intelligence.
Procedia PDF Downloads 816174 Advances of Image Processing in Precision Agriculture: Using Deep Learning Convolution Neural Network for Soil Nutrient Classification
Authors: Halimatu S. Abdullahi, Ray E. Sheriff, Fatima Mahieddine
Abstract:
Agriculture is essential to the continuous existence of human life as they directly depend on it for the production of food. The exponential rise in population calls for a rapid increase in food with the application of technology to reduce the laborious work and maximize production. Technology can aid/improve agriculture in several ways through pre-planning and post-harvest by the use of computer vision technology through image processing to determine the soil nutrient composition, right amount, right time, right place application of farm input resources like fertilizers, herbicides, water, weed detection, early detection of pest and diseases etc. This is precision agriculture which is thought to be solution required to achieve our goals. There has been significant improvement in the area of image processing and data processing which has being a major challenge. A database of images is collected through remote sensing, analyzed and a model is developed to determine the right treatment plans for different crop types and different regions. Features of images from vegetations need to be extracted, classified, segmented and finally fed into the model. Different techniques have been applied to the processes from the use of neural network, support vector machine, fuzzy logic approach and recently, the most effective approach generating excellent results using the deep learning approach of convolution neural network for image classifications. Deep Convolution neural network is used to determine soil nutrients required in a plantation for maximum production. The experimental results on the developed model yielded results with an average accuracy of 99.58%.Keywords: convolution, feature extraction, image analysis, validation, precision agriculture
Procedia PDF Downloads 3066173 An Empirical Study on Switching Activation Functions in Shallow and Deep Neural Networks
Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha
Abstract:
Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs –Sigmoid, ReLU, and Tanh–have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment with multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLUReLU) combination. Our results show that using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).Keywords: activation function, universal approximation function, neural networks, convergence
Procedia PDF Downloads 1466172 Defect Detection for Nanofibrous Images with Deep Learning-Based Approaches
Authors: Gaokai Liu
Abstract:
Automatic defect detection for nanomaterial images is widely required in industrial scenarios. Deep learning approaches are considered as the most effective solutions for the great majority of image-based tasks. In this paper, an edge guidance network for defect segmentation is proposed. First, the encoder path with multiple convolution and downsampling operations is applied to the acquisition of shared features. Then two decoder paths both are connected to the last convolution layer of the encoder and supervised by the edge and segmentation labels, respectively, to guide the whole training process. Meanwhile, the edge and encoder outputs from the same stage are concatenated to the segmentation corresponding part to further tune the segmentation result. Finally, the effectiveness of the proposed method is verified via the experiments on open nanofibrous datasets.Keywords: deep learning, defect detection, image segmentation, nanomaterials
Procedia PDF Downloads 1396171 Using Deep Learning for the Detection of Faulty RJ45 Connectors on a Radio Base Station
Authors: Djamel Fawzi Hadj Sadok, Marrone Silvério Melo Dantas Pedro Henrique Dreyer, Gabriel Fonseca Reis de Souza, Daniel Bezerra, Ricardo Souza, Silvia Lins, Judith Kelner
Abstract:
A radio base station (RBS), part of the radio access network, is a particular type of equipment that supports the connection between a wide range of cellular user devices and an operator network access infrastructure. Nowadays, most of the RBS maintenance is carried out manually, resulting in a time consuming and costly task. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. This paper proposes and compares two deep learning solutions to identify attached RJ45 connectors on network ports. We named connector detection, the solution based on object detection, and connector classification, the one based on object classification. With the connector detection, we get an accuracy of 0:934, mean average precision 0:903. Connector classification, get a maximum accuracy of 0:981 and an AUC of 0:989. Although connector detection was outperformed in this study, this should not be viewed as an overall result as connector detection is more flexible for scenarios where there is no precise information about the environment and the possible devices. At the same time, the connector classification requires that information to be well-defined.Keywords: radio base station, maintenance, classification, detection, deep learning, automation
Procedia PDF Downloads 1926170 Obstacle Avoidance Using Image-Based Visual Servoing Based on Deep Reinforcement Learning
Authors: Tong He, Long Chen, Irag Mantegh, Wen-Fang Xie
Abstract:
This paper proposes an image-based obstacle avoidance and tracking target identification strategy in GPS-degraded or GPS-denied environment for an Unmanned Aerial Vehicle (UAV). The traditional force algorithm for obstacle avoidance could produce local minima area, in which UAV cannot get away obstacle effectively. In order to eliminate it, an artificial potential approach based on harmonic potential is proposed to guide the UAV to avoid the obstacle by using the vision system. And image-based visual servoing scheme (IBVS) has been adopted to implement the proposed obstacle avoidance approach. In IBVS, the pixel accuracy is a key factor to realize the obstacle avoidance. In this paper, the deep reinforcement learning framework has been applied by reducing pixel errors through constant interaction between the environment and the agent. In addition, the combination of OpenTLD and Tensorflow based on neural network is used to identify the type of tracking target. Numerical simulation in Matlab and ROS GAZEBO show the satisfactory result in target identification and obstacle avoidance.Keywords: image-based visual servoing, obstacle avoidance, tracking target identification, deep reinforcement learning, artificial potential approach, neural network
Procedia PDF Downloads 1346169 Efficacy of Cool's and Rhythmic Stabilization Exercises on Scapular up Ward Rotation and Ut/Sa Ratio in Patients with Shoulder Impingement Syndrome
Authors: Mohammed Moustafa, Khaled Ayad, Waleed Reda
Abstract:
Shoulder impingement syndrome is the most common disorder of the shoulder, resulting in functional loss and disability. Objective: This study was designed to compare between the effects of scapular muscle training versus rhythmic stabilization exercises in treatment of shoulder impingement syndrome. Methods: Thirty patients participated in this study; they were assigned randomly into two experimental groups. The first experimental group (A) consisted of 15 patients with a mean age (21.87±2.72) years; they received graduated rhythmic stabilization exercises and stretching of the posterior capsule. The second experimental group (B) consisted of 15 patients with a mean age (22.27±2.94) years; they received scapular muscle training exercises in addition to stretching of the posterior capsule. Treatment was given three times per week, every other day, for four consecutive weeks. Patients have been evaluated pretreatment and post treatment for shoulder pain severity and functional disability. Results: Both groups showed highly statistical significant reduction in pain severity and functional disability measured post-treatment when compared with their corresponding values in pretreatment assessment. Conclusion: Both of rhythmic stabilization exercises and scapular muscle training are effective interventions to reduce shoulder pain severity and functional disability.Keywords: impingement syndrome, scapular exercises, rhythmic stabilization exercises, posterior capsule stretch
Procedia PDF Downloads 2326168 COVID-19 Analysis with Deep Learning Model Using Chest X-Rays Images
Authors: Uma Maheshwari V., Rajanikanth Aluvalu, Kumar Gautam
Abstract:
The COVID-19 disease is a highly contagious viral infection with major worldwide health implications. The global economy suffers as a result of COVID. The spread of this pandemic disease can be slowed if positive patients are found early. COVID-19 disease prediction is beneficial for identifying patients' health problems that are at risk for COVID. Deep learning and machine learning algorithms for COVID prediction using X-rays have the potential to be extremely useful in solving the scarcity of doctors and clinicians in remote places. In this paper, a convolutional neural network (CNN) with deep layers is presented for recognizing COVID-19 patients using real-world datasets. We gathered around 6000 X-ray scan images from various sources and split them into two categories: normal and COVID-impacted. Our model examines chest X-ray images to recognize such patients. Because X-rays are commonly available and affordable, our findings show that X-ray analysis is effective in COVID diagnosis. The predictions performed well, with an average accuracy of 99% on training photographs and 88% on X-ray test images.Keywords: deep CNN, COVID–19 analysis, feature extraction, feature map, accuracy
Procedia PDF Downloads 716167 Shedding Light on the Black Box: Explaining Deep Neural Network Prediction of Clinical Outcome
Authors: Yijun Shao, Yan Cheng, Rashmee U. Shah, Charlene R. Weir, Bruce E. Bray, Qing Zeng-Treitler
Abstract:
Deep neural network (DNN) models are being explored in the clinical domain, following the recent success in other domains such as image recognition. For clinical adoption, outcome prediction models require explanation, but due to the multiple non-linear inner transformations, DNN models are viewed by many as a black box. In this study, we developed a deep neural network model for predicting 1-year mortality of patients who underwent major cardio vascular procedures (MCVPs), using temporal image representation of past medical history as input. The dataset was obtained from the electronic medical data warehouse administered by Veteran Affairs Information and Computing Infrastructure (VINCI). We identified 21,355 veterans who had their first MCVP in 2014. Features for prediction included demographics, diagnoses, procedures, medication orders, hospitalizations, and frailty measures extracted from clinical notes. Temporal variables were created based on the patient history data in the 2-year window prior to the index MCVP. A temporal image was created based on these variables for each individual patient. To generate the explanation for the DNN model, we defined a new concept called impact score, based on the presence/value of clinical conditions’ impact on the predicted outcome. Like (log) odds ratio reported by the logistic regression (LR) model, impact scores are continuous variables intended to shed light on the black box model. For comparison, a logistic regression model was fitted on the same dataset. In our cohort, about 6.8% of patients died within one year. The prediction of the DNN model achieved an area under the curve (AUC) of 78.5% while the LR model achieved an AUC of 74.6%. A strong but not perfect correlation was found between the aggregated impact scores and the log odds ratios (Spearman’s rho = 0.74), which helped validate our explanation.Keywords: deep neural network, temporal data, prediction, frailty, logistic regression model
Procedia PDF Downloads 1436166 Data Augmentation for Early-Stage Lung Nodules Using Deep Image Prior and Pix2pix
Authors: Qasim Munye, Juned Islam, Haseeb Qureshi, Syed Jung
Abstract:
Lung nodules are commonly identified in computed tomography (CT) scans by experienced radiologists at a relatively late stage. Early diagnosis can greatly increase survival. We propose using a pix2pix conditional generative adversarial network to generate realistic images simulating early-stage lung nodule growth. We have applied deep images prior to 2341 slices from 895 computed tomography (CT) scans from the Lung Image Database Consortium (LIDC) dataset to generate pseudo-healthy medical images. From these images, 819 were chosen to train a pix2pix network. We observed that for most of the images, the pix2pix network was able to generate images where the nodule increased in size and intensity across epochs. To evaluate the images, 400 generated images were chosen at random and shown to a medical student beside their corresponding original image. Of these 400 generated images, 384 were defined as satisfactory - meaning they resembled a nodule and were visually similar to the corresponding image. We believe that this generated dataset could be used as training data for neural networks to detect lung nodules at an early stage or to improve the accuracy of such networks. This is particularly significant as datasets containing the growth of early-stage nodules are scarce. This project shows that the combination of deep image prior and generative models could potentially open the door to creating larger datasets than currently possible and has the potential to increase the accuracy of medical classification tasks.Keywords: medical technology, artificial intelligence, radiology, lung cancer
Procedia PDF Downloads 596165 Deep Learning in Chest Computed Tomography to Differentiate COVID-19 from Influenza
Authors: Hongmei Wang, Ziyun Xiang, Ying liu, Li Yu, Dongsheng Yue
Abstract:
Intro: The COVID-19 (Corona Virus Disease 2019) has greatly changed the global economic, political and financial ecology. The mutation of the coronavirus in the UK in December 2020 has brought new panic to the world. Deep learning was performed on Chest Computed tomography (CT) of COVID-19 and Influenza and describes their characteristics. The predominant features of COVID-19 pneumonia was ground-glass opacification, followed by consolidation. Lesion density: most lesions appear as ground-glass shadows, and some lesions coexist with solid lesions. Lesion distribution: the focus is mainly on the dorsal side of the periphery of the lung, with the lower lobe of the lungs as the focus, and it is often close to the pleura. Other features it has are grid-like shadows in ground glass lesions, thickening signs of diseased vessels, air bronchi signs and halo signs. The severe disease involves whole bilateral lungs, showing white lung signs, air bronchograms can be seen, and there can be a small amount of pleural effusion in the bilateral chest cavity. At the same time, this year's flu season could be near its peak after surging throughout the United States for months. Chest CT for Influenza infection is characterized by focal ground glass shadows in the lungs, with or without patchy consolidation, and bronchiole air bronchograms are visible in the concentration. There are patchy ground-glass shadows, consolidation, air bronchus signs, mosaic lung perfusion, etc. The lesions are mostly fused, which is prominent near the hilar and two lungs. Grid-like shadows and small patchy ground-glass shadows are visible. Deep neural networks have great potential in image analysis and diagnosis that traditional machine learning algorithms do not. Method: Aiming at the two major infectious diseases COVID-19 and influenza, which are currently circulating in the world, the chest CT of patients with two infectious diseases is classified and diagnosed using deep learning algorithms. The residual network is proposed to solve the problem of network degradation when there are too many hidden layers in a deep neural network (DNN). The proposed deep residual system (ResNet) is a milestone in the history of the Convolutional neural network (CNN) images, which solves the problem of difficult training of deep CNN models. Many visual tasks can get excellent results through fine-tuning ResNet. The pre-trained convolutional neural network ResNet is introduced as a feature extractor, eliminating the need to design complex models and time-consuming training. Fastai is based on Pytorch, packaging best practices for in-depth learning strategies, and finding the best way to handle diagnoses issues. Based on the one-cycle approach of the Fastai algorithm, the classification diagnosis of lung CT for two infectious diseases is realized, and a higher recognition rate is obtained. Results: A deep learning model was developed to efficiently identify the differences between COVID-19 and influenza using chest CT.Keywords: COVID-19, Fastai, influenza, transfer network
Procedia PDF Downloads 1336164 Use Cloud-Based Watson Deep Learning Platform to Train Models Faster and More Accurate
Authors: Susan Diamond
Abstract:
Machine Learning workloads have traditionally been run in high-performance computing (HPC) environments, where users log in to dedicated machines and utilize the attached GPUs to run training jobs on huge datasets. Training of large neural network models is very resource intensive, and even after exploiting parallelism and accelerators such as GPUs, a single training job can still take days. Consequently, the cost of hardware is a barrier to entry. Even when upfront cost is not a concern, the lead time to set up such an HPC environment takes months from acquiring hardware to set up the hardware with the right set of firmware, software installed and configured. Furthermore, scalability is hard to achieve in a rigid traditional lab environment. Therefore, it is slow to react to the dynamic change in the artificial intelligent industry. Watson Deep Learning as a service, a cloud-based deep learning platform that mitigates the long lead time and high upfront investment in hardware. It enables robust and scalable sharing of resources among the teams in an organization. It is designed for on-demand cloud environments. Providing a similar user experience in a multi-tenant cloud environment comes with its own unique challenges regarding fault tolerance, performance, and security. Watson Deep Learning as a service tackles these challenges and present a deep learning stack for the cloud environments in a secure, scalable and fault-tolerant manner. It supports a wide range of deep-learning frameworks such as Tensorflow, PyTorch, Caffe, Torch, Theano, and MXNet etc. These frameworks reduce the effort and skillset required to design, train, and use deep learning models. Deep Learning as a service is used at IBM by AI researchers in areas including machine translation, computer vision, and healthcare.Keywords: deep learning, machine learning, cognitive computing, model training
Procedia PDF Downloads 2006163 Black-Box-Base Generic Perturbation Generation Method under Salient Graphs
Authors: Dingyang Hu, Dan Liu
Abstract:
DNN (Deep Neural Network) deep learning models are widely used in classification, prediction, and other task scenarios. To address the difficulties of generic adversarial perturbation generation for deep learning models under black-box conditions, a generic adversarial ingestion generation method based on a saliency map (CJsp) is proposed to obtain salient image regions by counting the factors that influence the input features of an image on the output results. This method can be understood as a saliency map attack algorithm to obtain false classification results by reducing the weights of salient feature points. Experiments also demonstrate that this method can obtain a high success rate of migration attacks and is a batch adversarial sample generation method.Keywords: adversarial sample, gradient, probability, black box
Procedia PDF Downloads 876162 Gaits Stability Analysis for a Pneumatic Quadruped Robot Using Reinforcement Learning
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 3056161 Malaria Parasite Detection Using Deep Learning Methods
Authors: Kaustubh Chakradeo, Michael Delves, Sofya Titarenko
Abstract:
Malaria is a serious disease which affects hundreds of millions of people around the world, each year. If not treated in time, it can be fatal. Despite recent developments in malaria diagnostics, the microscopy method to detect malaria remains the most common. Unfortunately, the accuracy of microscopic diagnostics is dependent on the skill of the microscopist and limits the throughput of malaria diagnosis. With the development of Artificial Intelligence tools and Deep Learning techniques in particular, it is possible to lower the cost, while achieving an overall higher accuracy. In this paper, we present a VGG-based model and compare it with previously developed models for identifying infected cells. Our model surpasses most previously developed models in a range of the accuracy metrics. The model has an advantage of being constructed from a relatively small number of layers. This reduces the computer resources and computational time. Moreover, we test our model on two types of datasets and argue that the currently developed deep-learning-based methods cannot efficiently distinguish between infected and contaminated cells. A more precise study of suspicious regions is required.Keywords: convolution neural network, deep learning, malaria, thin blood smears
Procedia PDF Downloads 1226160 Deep Learning to Improve the 5G NR Uplink Control Channel
Authors: Ahmed Krobba, Meriem Touzene, Mohamed Debeyche
Abstract:
The wireless communications system (5G) will provide more diverse applications and higher quality services for users compared to the long-term evolution 4G (LTE). 5G uses a higher carrier frequency, which suffers from information loss in 5G coverage. Most 5G users often cannot obtain high-quality communications due to transmission channel noise and channel complexity. Physical Uplink Control Channel (PUCCH-NR: Physical Uplink Control Channel New Radio) plays a crucial role in 5G NR telecommunication technology, which is mainly used to transmit link control information uplink (UCI: Uplink Control Information. This study based of evaluating the performance of channel physical uplink control PUCCH-NR under low Signal-to-Noise Ratios with various antenna numbers reception. We propose the artificial intelligence approach based on deep neural networks (Deep Learning) to estimate the PUCCH-NR channel in comparison with this approach with different conventional methods such as least-square (LS) and minimum-mean-square-error (MMSE). To evaluate the channel performance we use the block error rate (BLER) as an evaluation criterion of the communication system. The results show that the deep neural networks method gives best performance compared with MMSE and LSKeywords: 5G network, uplink (Uplink), PUCCH channel, NR-PUCCH channel, deep learning
Procedia PDF Downloads 616159 Incorporating Lexical-Semantic Knowledge into Convolutional Neural Network Framework for Pediatric Disease Diagnosis
Authors: Xiaocong Liu, Huazhen Wang, Ting He, Xiaozheng Li, Weihan Zhang, Jian Chen
Abstract:
The utilization of electronic medical record (EMR) data to establish the disease diagnosis model has become an important research content of biomedical informatics. Deep learning can automatically extract features from the massive data, which brings about breakthroughs in the study of EMR data. The challenge is that deep learning lacks semantic knowledge, which leads to impracticability in medical science. This research proposes a method of incorporating lexical-semantic knowledge from abundant entities into a convolutional neural network (CNN) framework for pediatric disease diagnosis. Firstly, medical terms are vectorized into Lexical Semantic Vectors (LSV), which are concatenated with the embedded word vectors of word2vec to enrich the feature representation. Secondly, the semantic distribution of medical terms serves as Semantic Decision Guide (SDG) for the optimization of deep learning models. The study evaluate the performance of LSV-SDG-CNN model on four kinds of Chinese EMR datasets. Additionally, CNN, LSV-CNN, and SDG-CNN are designed as baseline models for comparison. The experimental results show that LSV-SDG-CNN model outperforms baseline models on four kinds of Chinese EMR datasets. The best configuration of the model yielded an F1 score of 86.20%. The results clearly demonstrate that CNN has been effectively guided and optimized by lexical-semantic knowledge, and LSV-SDG-CNN model improves the disease classification accuracy with a clear margin.Keywords: convolutional neural network, electronic medical record, feature representation, lexical semantics, semantic decision
Procedia PDF Downloads 1206158 Network Conditioning and Transfer Learning for Peripheral Nerve Segmentation in Ultrasound Images
Authors: Harold Mauricio Díaz-Vargas, Cristian Alfonso Jimenez-Castaño, David Augusto Cárdenas-Peña, Guillermo Alberto Ortiz-Gómez, Alvaro Angel Orozco-Gutierrez
Abstract:
Precise identification of the nerves is a crucial task performed by anesthesiologists for an effective Peripheral Nerve Blocking (PNB). Now, anesthesiologists use ultrasound imaging equipment to guide the PNB and detect nervous structures. However, visual identification of the nerves from ultrasound images is difficult, even for trained specialists, due to artifacts and low contrast. The recent advances in deep learning make neural networks a potential tool for accurate nerve segmentation systems, so addressing the above issues from raw data. The most widely spread U-Net network yields pixel-by-pixel segmentation by encoding the input image and decoding the attained feature vector into a semantic image. This work proposes a conditioning approach and encoder pre-training to enhance the nerve segmentation of traditional U-Nets. Conditioning is achieved by the one-hot encoding of the kind of target nerve a the network input, while the pre-training considers five well-known deep networks for image classification. The proposed approach is tested in a collection of 619 US images, where the best C-UNet architecture yields an 81% Dice coefficient, outperforming the 74% of the best traditional U-Net. Results prove that pre-trained models with the conditional approach outperform their equivalent baseline by supporting learning new features and enriching the discriminant capability of the tested networks.Keywords: nerve segmentation, U-Net, deep learning, ultrasound imaging, peripheral nerve blocking
Procedia PDF Downloads 936157 Detecting Memory-Related Gene Modules in sc/snRNA-seq Data by Deep-Learning
Authors: Yong Chen
Abstract:
To understand the detailed molecular mechanisms of memory formation in engram cells is one of the most fundamental questions in neuroscience. Recent single-cell RNA-seq (scRNA-seq) and single-nucleus RNA-seq (snRNA-seq) techniques have allowed us to explore the sparsely activated engram ensembles, enabling access to the molecular mechanisms that underlie experience-dependent memory formation and consolidation. However, the absence of specific and powerful computational methods to detect memory-related genes (modules) and their regulatory relationships in the sc/snRNA-seq datasets has strictly limited the analysis of underlying mechanisms and memory coding principles in mammalian brains. Here, we present a deep-learning method named SCENTBOX, to detect memory-related gene modules and causal regulatory relationships among themfromsc/snRNA-seq datasets. SCENTBOX first constructs codifferential expression gene network (CEGN) from case versus control sc/snRNA-seq datasets. It then detects the highly correlated modules of differential expression genes (DEGs) in CEGN. The deep network embedding and attention-based convolutional neural network strategies are employed to precisely detect regulatory relationships among DEG genes in a module. We applied them on scRNA-seq datasets of TRAP; Ai14 mouse neurons with fear memory and detected not only known memory-related genes, but also the modules and potential causal regulations. Our results provided novel regulations within an interesting module, including Arc, Bdnf, Creb, Dusp1, Rgs4, and Btg2. Overall, our methods provide a general computational tool for processing sc/snRNA-seq data from case versus control studie and a systematic investigation of fear-memory-related gene modules.Keywords: sc/snRNA-seq, memory formation, deep learning, gene module, causal inference
Procedia PDF Downloads 1136156 Automatic Number Plate Recognition System Based on Deep Learning
Authors: T. Damak, O. Kriaa, A. Baccar, M. A. Ben Ayed, N. Masmoudi
Abstract:
In the last few years, Automatic Number Plate Recognition (ANPR) systems have become widely used in the safety, the security, and the commercial aspects. Forethought, several methods and techniques are computing to achieve the better levels in terms of accuracy and real time execution. This paper proposed a computer vision algorithm of Number Plate Localization (NPL) and Characters Segmentation (CS). In addition, it proposed an improved method in Optical Character Recognition (OCR) based on Deep Learning (DL) techniques. In order to identify the number of detected plate after NPL and CS steps, the Convolutional Neural Network (CNN) algorithm is proposed. A DL model is developed using four convolution layers, two layers of Maxpooling, and six layers of fully connected. The model was trained by number image database on the Jetson TX2 NVIDIA target. The accuracy result has achieved 95.84%.Keywords: ANPR, CS, CNN, deep learning, NPL
Procedia PDF Downloads 2966155 Vehicle Detection and Tracking Using Deep Learning Techniques in Surveillance Image
Authors: Abe D. Desta
Abstract:
This study suggests a deep learning-based method for identifying and following moving objects in surveillance video. The proposed method uses a fast regional convolution neural network (F-RCNN) trained on a substantial dataset of vehicle images to first detect vehicles. A Kalman filter and a data association technique based on a Hungarian algorithm are then used to monitor the observed vehicles throughout time. However, in general, F-RCNN algorithms have been shown to be effective in achieving high detection accuracy and robustness in this research study. For example, in one study The study has shown that the vehicle detection and tracking, the system was able to achieve an accuracy of 97.4%. In this study, the F-RCNN algorithm was compared to other popular object detection algorithms and was found to outperform them in terms of both detection accuracy and speed. The presented system, which has application potential in actual surveillance systems, shows the usefulness of deep learning approaches in vehicle detection and tracking.Keywords: artificial intelligence, computer vision, deep learning, fast-regional convolutional neural networks, feature extraction, vehicle tracking
Procedia PDF Downloads 1066154 Improving Axial-Attention Network via Cross-Channel Weight Sharing
Authors: Nazmul Shahadat, Anthony S. Maida
Abstract:
In recent years, hypercomplex inspired neural networks improved deep CNN architectures due to their ability to share weights across input channels and thus improve cohesiveness of representations within the layers. The work described herein studies the effect of replacing existing layers in an Axial Attention ResNet with their quaternion variants that use cross-channel weight sharing to assess the effect on image classification. We expect the quaternion enhancements to produce improved feature maps with more interlinked representations. We experiment with the stem of the network, the bottleneck layer, and the fully connected backend by replacing them with quaternion versions. These modifications lead to novel architectures which yield improved accuracy performance on the ImageNet300k classification dataset. Our baseline networks for comparison were the original real-valued ResNet, the original quaternion-valued ResNet, and the Axial Attention ResNet. Since improvement was observed regardless of which part of the network was modified, there is a promise that this technique may be generally useful in improving classification accuracy for a large class of networks.Keywords: axial attention, representational networks, weight sharing, cross-channel correlations, quaternion-enhanced axial attention, deep networks
Procedia PDF Downloads 706153 Application of Deep Learning in Top Pair and Single Top Quark Production at the Large Hadron Collider
Authors: Ijaz Ahmed, Anwar Zada, Muhammad Waqas, M. U. Ashraf
Abstract:
We demonstrate the performance of a very efficient tagger applies on hadronically decaying top quark pairs as signal based on deep neural network algorithms and compares with the QCD multi-jet background events. A significant enhancement of performance in boosted top quark events is observed with our limited computing resources. We also compare modern machine learning approaches and perform a multivariate analysis of boosted top-pair as well as single top quark production through weak interaction at √s = 14 TeV proton-proton Collider. The most relevant known background processes are incorporated. Through the techniques of Boosted Decision Tree (BDT), likelihood and Multlayer Perceptron (MLP) the analysis is trained to observe the performance in comparison with the conventional cut based and count approachKeywords: top tagger, multivariate, deep learning, LHC, single top
Procedia PDF Downloads 1036152 Real-Time Pedestrian Detection Method Based on Improved YOLOv3
Authors: Jingting Luo, Yong Wang, Ying Wang
Abstract:
Pedestrian detection in image or video data is a very important and challenging task in security surveillance. The difficulty of this task is to locate and detect pedestrians of different scales in complex scenes accurately. To solve these problems, a deep neural network (RT-YOLOv3) is proposed to realize real-time pedestrian detection at different scales in security monitoring. RT-YOLOv3 improves the traditional YOLOv3 algorithm. Firstly, the deep residual network is added to extract vehicle features. Then six convolutional neural networks with different scales are designed and fused with the corresponding scale feature maps in the residual network to form the final feature pyramid to perform pedestrian detection tasks. This method can better characterize pedestrians. In order to further improve the accuracy and generalization ability of the model, a hybrid pedestrian data set training method is used to extract pedestrian data from the VOC data set and train with the INRIA pedestrian data set. Experiments show that the proposed RT-YOLOv3 method achieves 93.57% accuracy of mAP (mean average precision) and 46.52f/s (number of frames per second). In terms of accuracy, RT-YOLOv3 performs better than Fast R-CNN, Faster R-CNN, YOLO, SSD, YOLOv2, and YOLOv3. This method reduces the missed detection rate and false detection rate, improves the positioning accuracy, and meets the requirements of real-time detection of pedestrian objects.Keywords: pedestrian detection, feature detection, convolutional neural network, real-time detection, YOLOv3
Procedia PDF Downloads 1346151 The Influence of Mechanical and Physicochemical Characteristics of Perfume Microcapsules on Their Rupture Behaviour and How This Relates to Performance in Consumer Products
Authors: Andrew Gray, Zhibing Zhang
Abstract:
The ability for consumer products to deliver a sustained perfume response can be a key driver for a variety of applications. Many compounds in perfume oils are highly volatile, meaning they readily evaporate once the product is applied, and the longevity of the scent is poor. Perfume capsules have been introduced as a means of abating this evaporation once the product has been delivered. The impermeable capsules are aimed to be stable within the formulation, and remain intact during delivery to the desired substrate, only rupturing to release the core perfume oil through application of mechanical force applied by the consumer. This opens up the possibility of obtaining an olfactive response hours, weeks or even months after delivery, depending on the nature of the desired application. Tailoring the properties of the polymeric capsules to better address the needs of the application is not a trivial challenge and currently design of capsules is largely done by trial and error. The aim of this work is to have more predictive methods for capsule design depending on the consumer application. This means refining formulations such that they rupture at the right time for the specific consumer application, not too early, not too late. Finding the right balance between these extremes is essential if a benefit is sought with respect to neat addition of perfume to formulations. It is important to understand the forces that influence capsule rupture, first, by quantifying the magnitude of these different forces, and then by assessing bulk rupture in real-world applications to understand how capsules actually respond. Samples were provided by an industrial partner and the mechanical properties of individual capsules within the samples were characterized via a micromanipulation technique, developed by Professor Zhang at the University of Birmingham. The capsules were synthesized such as to change one particular physicochemical property at a time, such as core: wall material ratio, and the average size of capsules. Analysis of shell thickness via Transmission Electron Microscopy, size distribution via the use of a Mastersizer, as well as a variety of other techniques confirmed that only one particular physicochemical property was altered for each sample. The mechanical analysis was subsequently undertaken, showing the effect that changing certain capsule properties had on the response under compression. It was, however, important to link this fundamental mechanical response to capsule performance in real-world applications. As such, the capsule samples were introduced to a formulation and exposed to full scale stresses. GC-MS headspace analysis of the perfume oil released from broken capsules enabled quantification of what the relative strengths of capsules truly means for product performance. Correlations have been found between the mechanical strength of capsule samples and performance in terms of perfume release in consumer applications. Having a better understanding of the key parameters that drive performance benefits the design of future formulations by offering better guidelines on the parameters that can be adjusted without worrying about the performance effects, and singles out those parameters that are essential in finding the sweet spot for capsule performance.Keywords: consumer products, mechanical and physicochemical properties, perfume capsules, rupture behaviour
Procedia PDF Downloads 127