Deep Learning Based Smart Monitoring of Indoor Stadium Video Surveillance
Wang Yanbo, B. Bizu and V. Praveena
In the ongoing world, modern deep learning with computer interface technology has proven superior to previous learning methods. Video surveillance in smart monitoring has been broadly researched because of its wide range of applications. Though there are many significant challenges in successful data frame processing in the spatial and temporal domain, efficient classification of high-dimensional moving and live video frames remains needed research. To overcome such challenging problems, An Integrated Deep Learning-Based Smart Video Monitoring Module (DLS- VMM) for effective indoor stadium surveillance has been suggested. In particular, to analyze frame features, a significant storage network and access network have been introduced. To obtain automated indoor stadium monitoring and surveillance, the DLSVMM is processed to convert high-dimensional features into a class of lower-dimensional features Forgetting the timely frames of videos indoors. Such stored data are trained in the sub-spaces and integrated to encrypt the initial retrieval influenced by error clearance performance using digital communication codes to implement further improvements. The study findings demonstrate the effectiveness of this method. The high efficiency with detailed studies includes implementation and validation experimentally. The results revealed that the accuracy, F-measure, sensitivity, delay, and validation ratio of the DLSVMM achieves comparable accuracy with state-of-the-art methods.
Keywords: Safety measurements, security, video, surveillance, smart monitoring, deep learning