This paper presents a new approach, direct LL-mask band scheme (DLLBS), for the detection and tracking of moving objects using a low resolution image. Moving object detection is an important basic task for intelligent video surveillance systems, because it provides a focus of attention for post-processing. However, the successful detection of moving objects in a real environment is a difficult task, due to noise cause by fake motion, such as the motion of leaves in trees. Many methods have been developed in constrained environment, for the detection and tracking of moving objects. The DLLBS method can effectively reduce this noise, with low computing cost, in both indoor and outdoor environments. For circumstances where occlusions occur, we propose a new approach, characteristic point recognition (CPR). Together with DLLBS and CPR, the problems associated with occlusions are alleviated. The experimental results indicate that the proposed DLLBS method (for 320 x 240 and 640 x 480 image frames) can provide higher precision in the detection and tracking of moving objects, and multiple moving objects where occlusion is a problem, for real-time intelligent video surveillance applications.
International Journal of Innovative Computing, Information and Control 8(7) pt.A, pp.4451-4468