VLSI architecture design of moving objects detection using adaptive least-mean-square scheme

This work presents a new model based on least-mean-square (LMS) scheme to train the mask operation on low resolution images. This efficient and real-time method with adaptive least-mean-square scheme (ALMSS) uses the training mask for moving objects detection and tracking on resource-limited systems. However, the scheme of moving objects detection and tracking in a real surrounding environment is a difficult task due to noise issues such as fake motion or noise. The ALMSS approach can effectively reduce the noise with low computing cost in both fake motion and noise environments. In the experiments on real scenes indicate that the proposed ALMSS method is effective for moving objects detection and tracking in real-time. Our approach can be implemented in hardware for the high resolution applications, such as Full-HD images. A prototype VLSI circuit was designed and simulated by TSMC 0.18μm 1P6M process.