Optical Flow Analysis Based on Cellular Neural Networks

The optical flow is widely used in detection and dynamic analysis of moving objects, but it has the heavy computational burden, so it has not been able to in the real-time progress needs. Cellular Neural Network is a kind of neural network with outstanding image processing capabilities, and can be utilized in real-time applications. This paper presents an optical flow analysis method based on CNN, and for extract the interested moved object accurately, an approach which can be realized by CNN was given. Simulation results confirm the validity of the approach developed herein.