Vision Based Real Time Monitoring System for Elderly Fall Event Detection Using Deep Learning

作     者：G.Anitha S.Baghavathi Priya 

作者机构：Department of Information TechnologyRajalakshmi Engineering CollegeChennai602105India Department of Computer Science and EngineeringRajalakshmi Engineering CollegeChennai602105India 

出 版 物：《Computer Systems Science & Engineering》 (计算机系统科学与工程（英文）)

年 卷 期：2022年第42卷第7期

页      面：87-103页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：Computer vision elderly people fall detection deep learning metaheuristics object detection parameter optimization 

摘      要：Human fall detection plays a vital part in the design of sensor based alarming system,aid physical therapists not only to lessen after fall effect and also to save human *** and timely identification can offer quick medical ser-vices to the injured people and prevent from serious *** vision-based approaches have been developed by the placement of cameras in diverse everyday *** present times,deep learning(DL)models par-ticularly convolutional neural networks(CNNs)have gained much importance in the fall detection *** this motivation,this paper presents a new vision based elderly fall event detection using deep learning(VEFED-DL)*** proposed VEFED-DL model involves different stages of operations namely pre-processing,feature extraction,classification,and parameter ***-ily,the digital video camera is used to capture the RGB color images and the video is extracted into a set of *** improving the image quality and elim-inate noise,the frames are processed in three levels namely resizing,augmenta-tion,and min–max based ***,MobileNet model is applied as a feature extractor to derive the spatial features that exist in the preprocessed *** addition,the extracted spatial features are then fed into the gated recur-rent unit(GRU)to extract the temporal dependencies of the human ***,a group teaching optimization algorithm(GTOA)with stacked autoenco-der(SAE)is used as a binary classification model to determine the existence of fall or non-fall *** GTOA is employed for the parameter optimization of the SAE model in such a way that the detection performance can be *** order to assess the fall detection performance of the presented VEFED-DL model,a set of simulations take place on the UR fall detection dataset and multi-ple cameras fall *** experimental outcomes highlighted the superior per-formance of the presented method over the recent methods.