A Motion-Based Tracking System Using the Lucas-Kanade Optical Flow Method
Karl Leyven Leonida, Karla Veronica Sevilla, Cyrel O. Manlises
Abstract
In video monitoring system, many challenges will affect its performance in tracking the movement of people. It was not tested in Lucas-Kanade Optical if it is affected by illumination changes. The basic purpose of this research article is to build and evaluate a proposed system for tracking a person’s movement and motion under various lighting conditions. The system used Lucas-Kanade Optical Flow and Canny Edge Detector. One objective is to use Independent T-Test as its statistical method to compare the processing time of the proposed system concerning motion detection in three different lightings. In the result of data gathering of processing time, the t-statistical value in Sunny and Dim lighting was 12.80, 3.23 t-statistical value in Sunny and Dark lighting, and 11.31 t-statistical value in Dim and Dark lighting. From the gathered data, the researchers fulfilled the research objectives that the proposed system can track a person’s motion even if there is a minimal difference in the environment’s lighting.