A Robust Background Subtraction and Shadow Detection

摘要: This paper presents a novel algorithm for detecting moving objects from static background scene that contains shading and shadows using color images. Although the subtraction technique has been used years in many vision systems as preprocessing step object detection tracking, most of these algorithms are susceptible to both global local illumination changes such highlights. These cause consequent processes, e.g. recognition, etc., fail. problem is underlying motivation our work. We develop robust efficiently computed able cope with change problems, highlights, well changes. Experimental results, which demonstrate system’s performance, also shown.