Performance Study of Some Recent Optimization Techniques for Energy Minimization in Surveillance Video Synopsis Framework

In the age of the smart city, each activity is under surveillance. The employment of plentiful surveillance video cameras produces the gigantic amount of redundant video data. For ease of investigations, video synopsis competently shrinks the length with the preservation of all activities presents in the original video. The outcome of the video synopsis technology greatly depends on the central module, the optimization framework, and its minimization. This paper evaluates the performance of various optimization techniques, namely simulated annealing (SA), NSGA II, cultural algorithm (CA), teaching-learning-based optimization (TLBO), graywolf optimizer (GWO), forest optimization algorithm (FOA), JAYA algorithm, elitist-JAYA algorithm, self-adaptive multi-population-based JAYA algorithm (SAMP-JAYA), to minimize the energy in the field of object-based surveillance video synopsis. The experimental results and analysis direct the need for an optimization algorithmwhich can efficiently and consistently solve the minimization problem in connection to video synopsis.