Tracking of the moving object using novel CAM-shift algorithm

In this paper, we propose an algorithm that controls the expansion of the search window due to similar color background and moving objects, which is a disadvantage of CAM-shift. The proposed method designates the moving object to be tracked by using the search window designation method of the existing CAM-shift, extracts the feature points of the moving object by dividing the upper body and the lower body inside the designated search window, and finds the histogram. The feature points extracted from the upper and lower bodies are compared with the pixels of the histogram to determine the final feature points of the object by dividing the upper and lower bodies. Finally, we apply the RANSAC algorithm to find the histogram average of the extracted feature points and create a new search window. and constantly updates new search windows to keep track of objects. Finally, several experiments demonstrate the feasibility and applicability of the proposed method. Copyright © The Korean Institute of Electrical Engineers.