Search results for: deep feed forward neural network

Authors: René A. Sørensen, Søren Skovsen, Peter Christiansen, Henrik Karstoft

Abstract:

Autonomous agricultural machines act in stochastic surroundings and therefore, must be able to perceive the surroundings in real time. This perception can be achieved using image sensors combined with advanced machine learning, in particular Deep Learning. Deep convolutional neural networks excel in labeling and perceiving color images and since the cost of high-quality RGB-cameras is low, the hardware cost of good perception depends heavily on memory and computation power. This paper investigates the possibility of designing lightweight convolutional neural networks for semantic segmentation (pixel wise classification) with reduced hardware requirements, to allow for embedded usage in autonomous agricultural machines. Using compression techniques, a lightweight convolutional neural network is designed to perform real-time semantic segmentation on an embedded platform. The network is trained on two large datasets, ImageNet and Pascal Context, to recognize up to 400 individual classes. The 400 classes are remapped into agricultural superclasses (e.g. human, animal, sky, road, field, shelterbelt and obstacle) and the ability to provide accurate real-time perception of agricultural surroundings is studied. The network is applied to the case of autonomous grass mowing using the NVIDIA Tegra X1 embedded platform. Feeding case-specific images to the network results in a fully segmented map of the superclasses in the image. As the network is still being designed and optimized, only a qualitative analysis of the method is complete at the abstract submission deadline. Proceeding this deadline, the finalized design is quantitatively evaluated on 20 annotated grass mowing images. Lightweight convolutional neural networks for semantic segmentation can be implemented on an embedded platform and show competitive performance with regards to accuracy and speed. It is feasible to provide cost-efficient perceptive capabilities related to semantic segmentation for autonomous agricultural machines.

Keywords: autonomous agricultural machines, deep learning, safety, visual perception

Procedia PDF Downloads 396

Authors: Pitigalage Chamath Chandira Peiris

Abstract:

A widely researched domain in the field of image processing in recent times has been single image super-resolution, which tries to restore a high-resolution image from a single low-resolution image. Many more single image super-resolution efforts have been completed utilizing equally traditional and deep learning methodologies, as well as a variety of other methodologies. Deep learning-based super-resolution methods, in particular, have received significant interest. As of now, the most advanced image restoration approaches are based on convolutional neural networks; nevertheless, only a few efforts have been performed using Transformers, which have demonstrated excellent performance on high-level vision tasks. The effectiveness of CNN-based algorithms in image super-resolution has been impressive. However, these methods cannot completely capture the non-local features of the data. Enhancer is a simple yet powerful Transformer-based approach for enhancing the resolution of images. A method for single image super-resolution was developed in this study, which utilized an efficient and effective transformer design. This proposed architecture makes use of a locally enhanced window transformer block to alleviate the enormous computational load associated with non-overlapping window-based self-attention. Additionally, it incorporates depth-wise convolution in the feed-forward network to enhance its ability to capture local context. This study is assessed by comparing the results obtained for popular datasets to those obtained by other techniques in the domain.

Keywords: single image super resolution, computer vision, vision transformers, image restoration

Procedia PDF Downloads 105

Authors: David Pham, Yongfeng Zhang

Abstract:

Graphs are an important structure for data storage and computation. Recent years have seen the success of deep learning on graphs such as Graph Neural Networks (GNN) on various data mining and machine learning tasks. However, most of the deep learning models on graphs cannot easily explain their predictions and are thus often labelled as “black boxes.” For example, Graph Attention Network (GAT) is a frequently used GNN architecture, which adopts an attention mechanism to carefully select the neighborhood nodes for message passing and aggregation. However, it is difficult to explain why certain neighbors are selected while others are not and how the selected neighbors contribute to the final classification result. In this paper, we present a graph learning model called Explainable Graph Attention Network (XGAT), which integrates graph attention modeling and explainability. We use a single model to target both the accuracy and explainability of problem spaces and show that in the context of graph attention modeling, we can design a unified neighborhood selection strategy that selects appropriate neighbor nodes for both better accuracy and enhanced explainability. To justify this, we conduct extensive experiments to better understand the behavior of our model under different conditions and show an increase in both accuracy and explainability.

Keywords: explainable AI, graph attention network, graph neural network, node classification

Procedia PDF Downloads 199

Authors: Nan Deng, Zhenqiu Liu

Abstract:

Thousands of patients with metastatic tumors were diagnosed with cancers of unknown primary sites each year. The inability to identify the primary cancer site may lead to inappropriate treatment and unexpected prognosis. Nowadays, a large amount of genomics and transcriptomics cancer data has been generated by next-generation sequencing (NGS) technologies, and The Cancer Genome Atlas (TCGA) database has accrued thousands of human cancer tumors and healthy controls, which provides an abundance of resource to differentiate cancer types. Meanwhile, deep convolutional neural networks (CNNs) have shown high accuracy on classification among a large number of image object categories. Here, we utilize 25 cancer primary tumors and 3 normal tissues from TCGA and convert their RNA-Seq gene expression profiling to color images; train, validate and test a CNN classifier directly from these images. The performance result shows that our CNN classifier can archive >80% test accuracy on most of the tumors and normal tissues. Since the gene expression pattern of distant metastases is similar to their primary tumors, the CNN classifier may provide a potential computational strategy on identifying the unknown primary origin of metastatic cancer in order to plan appropriate treatment for patients.

Keywords: bioinformatics, cancer, convolutional neural network, deep leaning, gene expression pattern

Procedia PDF Downloads 299

Authors: Huy Hieu Pham, Houssam Salmane, Louahdi Khoudour, Alain Crouzil, Pablo Zegers, Sergio Velastin

Abstract:

We present a deep learning-based multitask framework for joint 3D human pose estimation and action recognition from color video sequences. Our approach proceeds along two stages. In the first, we run a real-time 2D pose detector to determine the precise pixel location of important key points of the body. A two-stream neural network is then designed and trained to map detected 2D keypoints into 3D poses. In the second, we deploy the Efficient Neural Architecture Search (ENAS) algorithm to find an optimal network architecture that is used for modeling the Spatio-temporal evolution of the estimated 3D poses via an image-based intermediate representation and performing action recognition. Experiments on Human3.6M, Microsoft Research Redmond (MSR) Action3D, and Stony Brook University (SBU) Kinect Interaction datasets verify the effectiveness of the proposed method on the targeted tasks. Moreover, we show that our method requires a low computational budget for training and inference.

Keywords: human action recognition, pose estimation, D-CNN, deep learning

Procedia PDF Downloads 146

Authors: Li Hui, Riyadh Hindi

Abstract:

In the United States, bridge construction often employs overhang brackets to support the deck overhang, the weight of fresh concrete, and loads from construction equipment. This approach, however, can lead to significant torsional moments on the exterior girders, potentially causing excessive girder rotation. Such rotations can result in various safety and maintenance issues, including thinning of the deck, reduced concrete cover, and cracking during service. Traditionally, these issues are addressed by installing temporary lateral bracing systems and conducting comprehensive torsional analysis through detailed finite element analysis for the construction of bridge deck overhang. However, this process is often intricate and time-intensive, with the spacing between temporary lateral bracing systems usually relying on the field engineers’ expertise. In this study, a deep neural network model is introduced to limit exterior girder rotation during bridge deck construction. The model predicts the optimal spacing between temporary bracing systems. To train this model, over 10,000 finite element models were generated in SAP2000, incorporating varying parameters such as girder dimensions, span length, and types and spacing of lateral bracing systems. The findings demonstrate that the deep neural network provides an effective and efficient alternative for limiting the exterior girder rotation for bridge deck construction. By reducing dependence on extensive finite element analyses, this approach stands out as a significant advancement in improving safety and maintenance effectiveness in the construction of bridge decks.

Keywords: bridge deck construction, exterior girder rotation, deep learning, finite element analysis

Procedia PDF Downloads 62

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 141

Authors: Evan Lowhorn, Rocio Alba-Flores

Abstract:

The purpose of this work is to develop a method for classifying hand signals and using the output in a drone control algorithm. To achieve this, methods based on Convolutional Neural Networks (CNN) were applied. CNN's are a subset of deep learning, which allows grid-like inputs to be processed and passed through a neural network to be trained for classification. This type of neural network allows for classification via imaging, which is less intrusive than previous methods using biosensors, such as EMG sensors. Classification CNN's operate purely from the pixel values in an image; therefore they can be used without additional exteroceptive sensors. A development bench was constructed using a desktop computer connected to a high-definition webcam mounted on a scissor arm. This allowed the camera to be pointed downwards at the desk to provide a constant solid background for the dataset and a clear detection area for the user. A MATLAB script was created to automate dataset image capture at the development bench and save the images to the desktop. This allowed the user to create their own dataset of 12,000 images within three hours. These images were evenly distributed among seven classes. The defined classes include forward, backward, left, right, idle, and land. The drone has a popular flip function which was also included as an additional class. To simplify control, the corresponding hand signals chosen were the numerical hand signs for one through five for movements, a fist for land, and the universal “ok” sign for the flip command. Transfer learning with PyTorch (Python) was performed using a pre-trained 18-layer residual learning network (ResNet-18) to retrain the network for custom classification. An algorithm was created to interpret the classification and send encoded messages to a Ryze Tello drone over its 2.4 GHz Wi-Fi connection. The drone’s movements were performed in half-meter distance increments at a constant speed. When combined with the drone control algorithm, the classification performed as desired with negligible latency when compared to the delay in the drone’s movement commands.

Keywords: classification, computer vision, convolutional neural networks, drone control

Procedia PDF Downloads 210

Authors: Gizem Eser Erdek

Abstract:

This study is research on predicting the remaining life of industrial cutting tools used in the industrial production process with deep learning methods. When the life of cutting tools decreases, they cause destruction to the raw material they are processing. This study it is aimed to predict the remaining life of the cutting tool based on the damage caused by the cutting tools to the raw material. For this, hole photos were collected from the hole-drilling machine for 8 months. Photos were labeled in 5 classes according to hole quality. In this way, the problem was transformed into a classification problem. Using the prepared data set, a model was created with convolutional neural networks, which is a deep learning method. In addition, VGGNet and ResNet architectures, which have been successful in the literature, have been tested on the data set. A hybrid model using convolutional neural networks and support vector machines is also used for comparison. When all models are compared, it has been determined that the model in which convolutional neural networks are used gives successful results of a %74 accuracy rate. In the preliminary studies, the data set was arranged to include only the best and worst classes, and the study gave ~93% accuracy when the binary classification model was applied. The results of this study showed that the remaining life of the cutting tools could be predicted by deep learning methods based on the damage to the raw material. Experiments have proven that deep learning methods can be used as an alternative for cutting tool life estimation.

Keywords: classification, convolutional neural network, deep learning, remaining life of industrial cutting tools, ResNet, support vector machine, VggNet

Procedia PDF Downloads 77

Authors: Ons Slimene, Aroua Taamallah, Maha Khemaja

Abstract:

In recent years, human activity recognition (HAR) has been a key area of research due to its diverse applications. It has garnered increasing attention in the field of computer vision. HAR plays an important role in people’s daily lives as it has the ability to learn advanced knowledge about human activities from data. In HAR, activities are usually represented by exploiting different types of sensors, such as embedded sensors or visual sensors. However, these sensors have limitations, such as local obstacles, image-related obstacles, sensor unreliability, and consumer concerns. Recently, several deep learning-based approaches have been proposed for HAR and these approaches are classified into two categories based on the type of data used: vision-based approaches and sensor-based approaches. This research paper highlights the importance of multimodal data fusion from skeleton data obtained from videos and data generated by embedded sensors using deep neural networks for achieving HAR. We propose a deep multimodal fusion network based on a twostream architecture. These two streams use the Convolutional Neural Network combined with the Bidirectional LSTM (CNN BILSTM) to process skeleton data and data generated by embedded sensors and the fusion at the feature level is considered. The proposed model was evaluated on a public OPPORTUNITY++ dataset and produced a accuracy of 96.77%.

Keywords: human activity recognition, action recognition, sensors, vision, human-centric sensing, deep learning, context-awareness

Procedia PDF Downloads 101

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 153

Authors: Abdolreza Memari

Abstract:

In this article in order to estimate the viscosity of crude oil,a numerical method has been used. We use this method to measure the crude oil's viscosity for 3 states: Saturated oil's viscosity, viscosity above the bubble point and viscosity under the saturation pressure. Then the crude oil's viscosity is estimated by using KHAN model and roller ball method. After that using these data that include efficient conditions in measuring viscosity, the estimated viscosity by the presented method, a radial based neural method, is taught. This network is a kind of two layered artificial neural network that its stimulation function of hidden layer is Gaussian function and teaching algorithms are used to teach them. After teaching radial based neural network, results of experimental method and artificial intelligence are compared all together. Teaching this network, we are able to estimate crude oil's viscosity without using KHAN model and experimental conditions and under any other condition with acceptable accuracy. Results show that radial neural network has high capability of estimating crude oil saving in time and cost is another advantage of this investigation.

Keywords: viscosity, Iranian crude oil, radial based, neural network, roller ball method, KHAN model

Procedia PDF Downloads 501

Authors: Yao-Hong Tsai

Abstract:

Due to the sensor technology, video surveillance has become the main way for security control in every big city in the world. Surveillance is usually used by governments for intelligence gathering, the prevention of crime, the protection of a process, person, group or object, or the investigation of crime. Many surveillance systems based on computer vision technology have been developed in recent years. Moving target tracking is the most common task for Unmanned Aerial Vehicle (UAV) to find and track objects of interest in mobile aerial surveillance for civilian applications. The paper is focused on vision-based collision avoidance for UAVs by recurrent neural networks. First, images from cameras on UAV were fused based on deep convolutional neural network. Then, a recurrent neural network was constructed to obtain high-level image features for object tracking and extracting low-level image features for noise reducing. The system distributed the calculation of the whole system to local and cloud platform to efficiently perform object detection, tracking and collision avoidance based on multiple UAVs. The experiments on several challenging datasets showed that the proposed algorithm outperforms the state-of-the-art methods.

Keywords: unmanned aerial vehicle, object tracking, deep learning, collision avoidance

Procedia PDF Downloads 160

Authors: Nihal Murali, Kunal Gupta, Surekha Bhanot

Abstract:

Training of artificial neural networks (ANNs) using reinforcement learning (RL) techniques is being widely discussed in the robot learning literature. The high model complexity of ANNs along with the model-free nature of RL algorithms provides a desirable combination for many robotics applications. There is a huge need for algorithms that generalize using raw sensory inputs, such as vision, without any hand-engineered features or domain heuristics. In this paper, the standard control problem of line following robot was used as a test-bed, and an ANN controller for the robot was trained on images from a live video feed using Q-learning. A virtual agent was first trained in simulation environment and then deployed onto a robot’s hardware. The robot successfully learns to traverse a wide range of curves and displays excellent generalization ability. Qualitative analysis of the evolution of policies, performance and weights of the network provide insights into the nature and convergence of the learning algorithm.

Keywords: artificial neural networks, q-learning, reinforcement learning, robot learning

Procedia PDF Downloads 372

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 215

Authors: Christopher Godwin Udomboso, Angela Unna Chukwu, Isaac Kwame Dontwi

Abstract:

In selecting a Statistical Neural Network model, the Network Information Criterion (NIC) has been observed to be sample biased, because it does not account for sample sizes. The selection of a model from a set of fitted candidate models requires objective data-driven criteria. In this paper, we derived and investigated the Adjusted Network Information Criterion (ANIC), based on Kullback’s symmetric divergence, which has been designed to be an asymptotically unbiased estimator of the expected Kullback-Leibler information of a fitted model. The analyses show that on a general note, the ANIC improves model selection in more sample sizes than does the NIC.

Keywords: statistical neural network, network information criterion, adjusted network, information criterion, transfer function

Procedia PDF Downloads 566

Authors: Arbnor Pajaziti, Hasan Cana

Abstract:

In this paper, the structural genetic algorithm is used to optimize the neural network to control the joint movements of robotic arm. The robotic arm has also been modeled in 3D and simulated in real-time in MATLAB. It is found that Neural Networks provide a simple and effective way to control the robot tasks. Computer simulation examples are given to illustrate the significance of this method. By combining Genetic Algorithm optimization method and Neural Networks for the given robotic arm with 5 D.O.F. the obtained the results shown that the base joint movements overshooting time without controller was about 0.5 seconds, while with Neural Network controller (optimized with Genetic Algorithm) was about 0.2 seconds, and the population size of 150 gave best results.

Keywords: robotic arm, neural network, genetic algorithm, optimization

Procedia PDF Downloads 523

Authors: Rodrigo S. Moreira, Nelson F. F. Ebecken

Abstract:

This paper proposes a maritime vessel tracker composed of an ensemble of WiSARD weightless neural network classifiers. A failure detector analyzes vessel movement with a Kalman filter and corrects the tracking, if necessary, using FFT matching. The use of the WiSARD neural network to track objects is uncommon. The additional contributions of the present study include a performance comparison with four state-of-art trackers, an experimental study of the features that improve maritime vessel tracking, the first use of an ensemble of classifiers to track maritime vessels and a new quantization algorithm that compares the values of pixel pairs.

Keywords: ram memory, WiSARD weightless neural network, object tracking, quantization

Procedia PDF Downloads 310

Authors: Faezeh Aghazadeh, Mohammad Javad Sharifi

Abstract:

The application of neural networks to model a full-scale industrial acetylene hydrogenation in olefin plant has been studied. The operating variables studied are the, input-temperature of the reactor, output-temperature of the reactor, hydrogen ratio of the reactor, [C₂H₂]input, and [C₂H₆]input. The studied operating variables were used as the input to the constructed neural network to predict the [C₂H₆]output at any time as the output or the target. The constructed neural network was found to be highly precise in predicting the quantity of [C₂H₆]output for the new input data, which are kept unaware of the trained neural network showing its applicability to determine the [C₂H₆]output for any operating conditions. The enhancement of [C₂H₆]output as compared with [C₂H₆]input was a consequence of low selective acetylene hydrogenation to ethylene.

Keywords: acetylene hydrogenation, Pd-Ag/Al₂O₃, artificial neural network, modeling, optimal design

Procedia PDF Downloads 276

Authors: Ayyaz Hussain, Tariq Sadad

Abstract:

Breast cancer (BC) is one of the widespread ailments among females globally. The early prognosis of BC can decrease the mortality rate. Exact findings of benign tumors can avoid unnecessary biopsies and further treatments of patients under investigation. However, due to variations in images, it is a tough job to isolate cancerous cases from normal and benign ones. The machine learning technique is widely employed in the classification of BC pattern and prognosis. In this research, a deep convolution neural network (DCNN) called AlexNet architecture is employed to get more discriminative features from breast tissues. To achieve higher accuracy, K-nearest neighbor (KNN) classifiers are employed as a substitute for the softmax layer in deep learning. The proposed model is tested on a widely used breast image database called MIAS dataset for experimental purposes and achieved 99% accuracy.

Keywords: breast cancer, DCNN, KNN, mammography

Procedia PDF Downloads 136

Authors: Hadjer Sadoune, Liza Lamini, Scherazade Krim, Amel Djouadi, Rachida Rihani

Abstract:

Deep neural networks are highly regarded for their accuracy in predicting intricate fermentation processes. Their ability to learn from a large amount of datasets through artificial intelligence makes them particularly effective models. The primary obstacle in improving the performance of these models is to carefully choose the suitable hyperparameters, including the neural network architecture (number of hidden layers and hidden units), activation function, optimizer, learning rate, and other relevant factors. This study predicts biogas production from real wastewater treatment plant data using a sophisticated approach: hybrid Bayesian optimization with a tree-structured Parzen estimator (BO-TPE) for an optimised deep neural network (DNN) model. The plant utilizes an Upflow Anaerobic Sludge Blanket (UASB) digester that treats industrial wastewater from soft drinks and breweries. The digester has a working volume of 1574 m3 and a total volume of 1914 m3. Its internal diameter and height were 19 and 7.14 m, respectively. The data preprocessing was conducted with meticulous attention to preserving data quality while avoiding data reduction. Three normalization techniques were applied to the pre-processed data (MinMaxScaler, RobustScaler and StandardScaler) and compared with the Non-Normalized data. The RobustScaler approach has strong predictive ability for estimating the volume of biogas produced. The highest predicted biogas volume was 2236.105 Nm³/d, with coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE) values of 0.712, 164.610, and 223.429, respectively.

Keywords: anaerobic digestion, biogas production, deep neural network, hybrid bo-tpe, hyperparameters tuning

Procedia PDF Downloads 38

Authors: Donya Ashtiani Haghighi, Amirali Baniasadi

Abstract:

Coronavirus (COVID-19) disease has spread abruptly all over the world since the end of 2019. Computed tomography (CT) scans and X-ray images are used to detect this disease. Different Deep Neural Network (DNN)-based diagnosis solutions have been developed, mainly based on Convolutional Neural Networks (CNNs), to accelerate the identification of COVID-19 cases. However, CNNs lose important information in intermediate layers and require large datasets. In this paper, Capsule Network (CapsNet) is used. Capsule Network performs better than CNNs for small datasets. Accuracy of 0.9885, f1-score of 0.9883, precision of 0.9859, recall of 0.9908, and Area Under the Curve (AUC) of 0.9948 are achieved on the Capsule-based framework with hyperparameter tuning. Moreover, different dropout rates are investigated to decrease overfitting. Accordingly, a dropout rate of 0.1 shows the best results. Finally, we remove one convolution layer and decrease the number of trainable parameters to 146,752, which is a promising result.

Keywords: capsule network, dropout, hyperparameter tuning, classification

Procedia PDF Downloads 77

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 126

Authors: Sajjad Baghernezhad

Abstract:

Meta-heuristic and hybrid algorithms have high adeer in modeling medical problems. In this study, a neural network was used to predict covid-19 among high-risk and low-risk patients. This study was conducted to collect the applied method and its target population consisting of 550 high-risk and low-risk patients from the Kerman University of medical sciences medical center to predict the coronavirus. In this study, the memetic algorithm, which is a combination of a genetic algorithm and a local search algorithm, has been used to update the weights of the neural network and develop the accuracy of the neural network. The initial study showed that the accuracy of the neural network was 88%. After updating the weights, the memetic algorithm increased by 93%. For the proposed model, sensitivity, specificity, positive predictivity value, value/accuracy to 97.4, 92.3, 95.8, 96.2, and 0.918, respectively; for the genetic algorithm model, 87.05, 9.20 7, 89.45, 97.30 and 0.967 and for logistic regression model were 87.40, 95.20, 93.79, 0.87 and 0.916. Based on the findings of this study, neural network models have a lower error rate in the diagnosis of patients based on individual variables and vital signs compared to the regression model. The findings of this study can help planners and health care providers in signing programs and early diagnosis of COVID-19 or Corona.

Keywords: COVID-19, decision support technique, neural network, genetic algorithm, memetic algorithm

Procedia PDF Downloads 66

Authors: Svitov David, Alyamkin Sergey

Abstract:

The usage of convolutional neural networks (CNNs) in conjunction with the margin-based softmax approach demonstrates the state-of-the-art performance for the face recognition problem. Recently, lightweight neural network models trained with the margin-based softmax have been introduced for the face identification task for edge devices. In this paper, we propose a distillation method for lightweight neural network architectures that outperforms other known methods for the face recognition task on LFW, AgeDB-30 and Megaface datasets. The idea of the proposed method is to use class centers from the teacher network for the student network. Then the student network is trained to get the same angles between the class centers and face embeddings predicted by the teacher network.

Keywords: ArcFace, distillation, face recognition, margin-based softmax

Procedia PDF Downloads 146

Authors: Chaitanya Chawla, Divya Panwar, Gurneesh Singh Anand, M. P. S Bhatia

Abstract:

This paper presents a deep-learning mechanism for classifying computer generated images and photographic images. The proposed method accounts for a convolutional layer capable of automatically learning correlation between neighbouring pixels. In the current form, Convolutional Neural Network (CNN) will learn features based on an image's content instead of the structural features of the image. The layer is particularly designed to subdue an image's content and robustly learn the sensor pattern noise features (usually inherited from image processing in a camera) as well as the statistical properties of images. The paper was assessed on latest natural and computer generated images, and it was concluded that it performs better than the current state of the art methods.

Keywords: image forensics, computer graphics, classification, deep learning, convolutional neural networks

Procedia PDF Downloads 336

Authors: Iliya Musa Adamu

Abstract:

This paper seeks to develop an expert system that would troubleshoot computer network and evaluate the network system performance so as to reduce the workload on technicians and increase the efficiency and effectiveness of solutions proffered to computer network problems. The platform of the system was developed using ASP.NET, whereas the codes are implemented in Visual Basic and integrated with SQL Server 2005. The knowledge base was represented using production rule, whereas the searching method that was used in developing the network troubleshooting expert system is the forward-chaining-rule-based-system. This software tool offers the advantage of providing an immediate solution to most computer network problems encountered by computer users.

Keywords: expert system, forward chaining rule based system, network, troubleshooting

Procedia PDF Downloads 647

Authors: Yagya Raj Pandeya, Joonwhoan Lee

Abstract:

The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.

Keywords: multimodal, 3D convolution, music-video feature extraction, generalized mean

Procedia PDF Downloads 215

Authors: Gautam Kumar Saharia, Sagardeep Talukdar, Riki Dutta, Sudipta Nandy

Abstract:

The physics informed neural network (PINN) method opens up an approach for numerically solving nonlinear partial differential equations leveraging fast calculating speed and high precession of modern computing systems. We construct the PINN based on strong universal approximation theorem and apply the initial-boundary value data and residual collocation points to weekly impose initial and boundary condition to the neural network and choose the optimization algorithms adaptive moment estimation (ADAM) and Limited-memory Broyden-Fletcher-Golfard-Shanno (L-BFGS) algorithm to optimize learnable parameter of the neural network. Next, we improve the PINN with a weighted loss function to obtain both the bright and dark soliton solutions of Fokas-Lenells equation (FLE). We find the proposed scheme of adjustable weight coefficients into PINN has a better convergence rate and generalizability than the basic PINN algorithm. We believe that the PINN approach to solve the partial differential equation appearing in nonlinear optics would be useful to study various optical phenomena.

Keywords: deep learning, optical Soliton, neural network, partial differential equation

Procedia PDF Downloads 126

Authors: Maryam Zarabian, Hector Guzman, Pedro Pereira-Almao, Abraham Fapojuwo

Abstract:

Dry reforming of methane (DRM) has sparked significant industrial and scientific interest not only as a viable alternative for addressing the environmental concerns of two main contributors of the greenhouse effect, i.e., carbon dioxide (CO₂) and methane (CH₄), but also produces syngas, i.e., a mixture of hydrogen (H₂) and carbon monoxide (CO) utilized by a wide range of downstream processes as a feedstock for other chemical productions. In this study, we develop an AI-enable syngas production model to tackle the problem of achieving an equivalent H₂/CO ratio [1:1] with respect to the most efficient conversion. Firstly, the unsupervised density-based spatial clustering of applications with noise (DBSAN) algorithm removes outlier data points from the original experimental dataset. Then, random forest (RF) and deep neural network (DNN) models employ the error-free dataset to predict the DRM results. DNN models inherently would not be able to obtain accurate predictions without a huge dataset. To cope with this limitation, we employ reusing pre-trained layers’ approaches such as transfer learning and greedy layer-wise pretraining. Compared to the other deep models (i.e., pure deep model and transferred deep model), the greedy layer-wise pre-trained deep model provides the most accurate prediction as well as similar accuracy to the RF model with R² values 1.00, 0.999, 0.999, 0.999, 0.999, and 0.999 for the total outlet flow, H₂/CO ratio, H₂ yield, CO yield, CH₄ conversion, and CO₂ conversion outputs, respectively.

Keywords: artificial intelligence, dry reforming of methane, artificial neural network, deep learning, machine learning, transfer learning, greedy layer-wise pretraining

Procedia PDF Downloads 86