Video Object Tracking is currently an active area of research in several images processing applications suchlike traffic controlling, video compression, surveillance, human-computer interaction, and video editing. Object tracking algorithms strive to detect and track a determined moving object through a sequence of images. Regularly, these object tracking algorithms use and correlate many pre-processing tasks, such as motion estimation and image segmentation. However, researchers struggle with several challenging problems like changes in illumination in the scene, severe occlusion, or abrupt changes in object motion which affect the object tracking task.

Existing object tracking algorithms are used successfully to track a moving object through a sequence of images with a low grade of complexity. Meaning by low grade of complexity the spatial or temporal smoothness presented by a given set of images which causes that a moving object to be easily detected. Additionally, the presence of non-uniform illumination in a sequence of images, which is a phenomenon commonly presented in outdoor and indoor environments, is rarely dealt by existing algorithms. Therefore, there exists the need to explore a new methodology which tackles this problem. The primary contribution of this work is a methodology to deal with object tracking tasks in sequence of frames exposed to non-uniform illumination conditions.