Advancing Autoencoder Architectures for Enhanced Anomaly Detection in Multivariate Industrial Time Series

In the context of rapid digitization in industrial environments,how effective are advanced unsupervised learning models,particularly hybrid autoencoder models,at detecting anomalies in industrial control system(ICS)datasets?This study is crucial because it addresses the challenge of identifying rare and complex anomalous patterns in the vast amounts of time series data generated by Internet of Things(IoT)devices,which can significantly improve the reliability and safety of these systems.In this paper,we propose a hybrid autoencoder model,called ConvBiLSTMAE,which combines convolutional neural network(CNN)and bidirectional long short-term memory(BiLSTM)to more effectively train complex temporal data patterns in anomaly detection.On the hardware-in-the-loopbased extended industrial control system dataset,the ConvBiLSTM-AE model demonstrated remarkable anomaly detection performance,achieving F1 scores of 0.78 and 0.41 for the first and second datasets,respectively.The results suggest that hybrid autoencoder models are not only viable,but potentially superior alternatives for unsupervised anomaly detection in complex industrial systems,offering a promising approach to improving their reliability and safety.

A Smart Heart Disease Diagnostic System Using Deep Vanilla LSTM

Effective smart healthcare frameworks contain novel and emerging solutions for remote disease diagnostics,which aid in the prevention of several diseases including heart-related abnormalities.In this context,regular monitoring of cardiac patients through smart healthcare systems based on Electrocardiogram(ECG)signals has the potential to save many lives.In existing studies,several heart disease diagnostic systems are proposed by employing different state-of-the-art methods,however,improving such methods is always an intriguing area of research.Hence,in this research,a smart healthcare system is proposed for the diagnosis of heart disease using ECG signals.The proposed framework extracts both linear and time-series information on the ECG signals and fuses them into a single framework concurrently.The linear characteristics of ECG signals are extracted by convolution layers followed by Gaussian Error Linear Units(GeLu)and time series characteristics of ECG beats are extracted by Vanilla Long Short-Term Memory Networks(LSTM).Following on,the feature reduction of linear information is done with the help of ID Generalized Gated Pooling(GGP).In addition,data misbalancing issues are also addressed with the help of the Synthetic Minority Oversampling Technique(SMOTE).The performance assessment of the proposed model is done over the two publicly available datasets named MIT-BIH arrhythmia database(MITDB)and PTB Diagnostic ECG database(PTBDB).The proposed framework achieves an average accuracy performance of 99.14%along with a 95%recall value.

Intelligent Prediction Approach for Diabetic Retinopathy Using Deep Learning Based Convolutional Neural Networks Algorithm by Means of Retina Photographs

Retinopathy is a human eye disease that causes changes in retinal blood vessels that leads to bleed,leak fluid and vision impairment.Symptoms of retinopathy are blurred vision,changes in color perception,red spots,and eye pain and it cannot be detected with a naked eye.In this paper,a new methodology based on Convolutional Neural Networks(CNN)is developed and proposed to intelligent retinopathy prediction and give a decision about the presence of retinopathy with automatic diabetic retinopathy screening with accurate diagnoses.The CNN model is trained by different images of eyes that have retinopathy and those which do not have retinopathy.The fully connected layers perform the classification process of the images from the dataset with the pooling layers minimize the coherence among the adjacent layers.The feature loss factor increases the label value to identify the patterns with the kernel-based matching.The performance of the proposed model is compared with the related methods of DREAM,KNN,GD-CNN and SVM.Experimental results show that the proposed CNN performs better.

Long Short Term Memory Networks Based Anomaly Detection for KPIs

In real-world many internet-based service companies need to closely monitor large amounts of data in order to ensure stable operation of their business.However,anomaly detection for these data with various patterns and data quality has been a great challenge,especially without labels.In this paper,we adopt an anomaly detection algorithm based on Long Short-Term Memory(LSTM)Network in terms of reconstructing KPIs and predicting KPIs.They use the reconstruction error and prediction error respectively as the criteria for judging anomalies,and we test our method with real data from a company in the insurance industry and achieved good performance.

Anomaly Based Camera Prioritization in Large Scale Surveillance Networks

Digital surveillance systems are ubiquitous and continuously generate massive amounts of data,and manual monitoring is required in order to recognise human activities in public areas.Intelligent surveillance systems that can automatically identify normal and abnormal activities are highly desirable,as these would allow for efficient monitoring by selecting only those camera feeds in which abnormal activities are occurring.This paper proposes an energy-efficient camera prioritisation framework that intelligently adjusts the priority of cameras in a vast surveillance network using feedback from the activity recognition system.The proposed system addresses the limitations of existing manual monitoring surveillance systems using a three-step framework.In the first step,the salient frames are selected from the online video stream using a frame differencing method.A lightweight 3D convolutional neural network(3DCNN)architecture is applied to extract spatio-temporal features from the salient frames in the second step.Finally,the probabilities predicted by the 3DCNN network and the metadata of the cameras are processed using a linear threshold gate sigmoid mechanism to control the priority of the camera.The proposed system performs well compared to state-of-theart violent activity recognition methods in terms of efficient camera prioritisation in large-scale surveillance networks.Comprehensive experiments and an evaluation of activity recognition and camera prioritisation showed that our approach achieved an accuracy of 98%with an F1-score of 0.97 on the Hockey Fight dataset,and an accuracy of 99%with an F1-score of 0.98 on the Violent Crowd dataset.

A Triplet-Branch Convolutional Neural Network for Part-Based Gait Recognition

Intelligent vision-based surveillance systems are designed to deal with the gigantic volume of videos captured in a particular environment to perform the interpretation of scenes in form of detection,tracking,monitoring,behavioral analysis,and retrievals.In addition to that,another evolving way of surveillance systems in a particular environment is human gait-based surveillance.In the existing research,several methodological frameworks are designed to use deep learning and traditional methods,nevertheless,the accuracies of these methods drop substantially when they are subjected to covariate conditions.These covariate variables disrupt the gait features and hence the recognition of subjects becomes difficult.To handle these issues,a region-based triplet-branch Convolutional Neural Network(CNN)is proposed in this research that is focused on different parts of the human Gait Energy Image(GEI)including the head,legs,and body separately to classify the subjects,and later on,the final identification of subjects is decided by probability-based majority voting criteria.Moreover,to enhance the feature extraction and draw the discriminative features,we have added soft attention layers on each branch to generate the soft attention maps.The proposed model is validated on the CASIA-B database and findings indicate that part-based learning through triplet-branch CNN shows good performance of 72.98%under covariate conditions as well as also outperforms single-branch CNN models.

An efficient deep learning-assisted person re-identification solution for intelligent video surveillance in smart cities

Innovations on the Internet of Everything(IoE)enabled systems are driving a change in the settings where we interact in smart units,recognized globally as smart city environments.However,intelligent video-surveillance systems are critical to increasing the security of these smart cities.More precisely,in today’s world of smart video surveillance,person re-identification(Re-ID)has gained increased consideration by researchers.Various researchers have designed deep learningbased algorithms for person Re-ID because they have achieved substantial breakthroughs in computer vision problems.In this line of research,we designed an adaptive feature refinementbased deep learning architecture to conduct person Re-ID.In the proposed architecture,the inter-channel and inter-spatial relationship of features between the images of the same individual taken from nonidentical camera viewpoints are focused on learning spatial and channel attention.In addition,the spatial pyramid pooling layer is inserted to extract the multiscale and fixed-dimension feature vectors irrespective of the size of the feature maps.Furthermore,the model’s effectiveness is validated on the CUHK01 and CUHK02 datasets.When compared with existing approaches,the approach presented in this paper achieves encouraging Rank 1 and 5 scores of 24.6% and 54.8%,respectively.

Call for Papers Special Issue on Deep Learning and Evolutionary Computation for Satellite Imagery

Satellite images are humungous sources of data that require efficient methods for knowledge discovery.The increased availability of earth data from satellite images has immense opportunities in various fields.However,the volume and heterogeneity of data poses serious computational challenges.The development of efficient techniques has the potential of discovering hidden information from these images.This knowledge can be used in various activities related to planning,monitoring,and managing the earth resources.Deep learning are being widely used for image analysis and processing.Deep learning based models can be effectively used for mining and knowledge discovery from satellite images.