Comparison of Swarm Optimization Algorithms for Multi-Target Tracking and Detection

Nowadays the tracking and detection of different objects are important. One of the major applications where we commonly use both of these are the surveillance applications. Here the fastness and the accuracy are very much important and based on those things the technology is searching for new ideas or methods. Most of the algorithms for target detection are currently based on neural networks. The commonly used methods for selecting appropriate hyper-parameters are the random search and grid search methods. But those algorithms have a lot of disadvantages and that makes them inefficient and inaccurate. So, for avoiding those disadvantages we are considering optimization algorithms with swarm intelligence. There are different swarm optimization algorithms, in which the most efficient one is the Glow worm Swarm Optimization (GSO) algorithm. This algorithm will give fast and accurate tracking and detection of objects. We are going to analyze the applicability of the GSO algorithm for the tracking and detection of several moving and stationary targets at the same time and compare other swarm optimization algorithms on the basis of multi-target tracking and detection.