The best-so-far ABC (BSF-ABC) is the modified algorithm of artificial bee colony (ABC) which is implemented to solve optimization problems. This algorithm is one of the swarm intelligence algorithms which has recently been introduced in the literature and the results of which show that the BSF-ABC can lead to higher quality solutions with quicker interaction than ordinary ABC or the algorithm based on the currently implemented modern ABC. In this work our goal is to implement the approach based on the BSF-ABC for detecting objects. The aforementioned approach is built on template-based matching using the difference of histograms of RGB level. The latter matches to the target object and template object as an objective function. The results come to prove that the suggested method has successfully been implemented both in object detection and in optimization of time spent on problem solving.