Innovative Edge Computing for Real-Time Video Surveillance and Taekwondo Training Enhancement

The constantly growing volume of data created globally makes it impossible for the centralised cloud computing method to provide low-latency, high-efficiency surveillance camera services.In order to alleviate transmission pressure, the load on the main cloud server, and the end to end latency of the video surveillance system, a distributed computing architecture is developed that immediately analyzes peripheral video data.By lowering the probability of tracker drift or malfunction in the videos, the suggested Enhanced Multiple Instance Learning with Whale Optimization Technique (EMIL-WOM) enables the classifier to extract the features with lower computing costs and shorter computation times.For various scenarios, the optimised neural network generates computation models, which are then logically placed in edge devices.The level of Taekwondo is chosen to address the uneven teaching quality for the goal of real-time analysis as society develops.To solve the teaching challenges in the Taekwondo learning process and improve the calibre of Taekwondo, the researcher conducted particular study in relation to the tactile learning theory.This research uses scientific and technological resources as a guide to assess technical actions and strategies and apply them to specific educational experiments for testing.This work analyses and recommends a way for making innovative services based on the edge computing paradigm.This experimental technique eliminates several interoperability and service scalability issues with conventional design.The suggested EMIL-WOM achieves 96.5% accuracy, 56.1% computational complexity, 32.4% RMSE, 24.1% RAE, 30% MAE, and 45.3 seconds of response time when compared to existing approaches.