Computational autonomous visual perception using cellular neural networks
Wei‐Song Lin, An-Te Liu, Chun-Hsiung Fang · 2005
Visual attention, which is a static feature process, and visual alerting, which is a dynamic feature process, constitutes the main functions of autonomous visual perception. They enable biological vision to detect rapidly the interesting parts in a scene and automatically select the attentive spots. The selected attentive spots usually correlate with the conspicuous parts of the scene, which may be the potential targets. This work has developed a computational model to extract and aggregate static and dynamic features for autonomous visual perception. Intention on optical intensity, color, and object location are considered in the procedure to determine the priority of the selected attentive spots. The huge computational burden in the Gaussian and Gabor filters of the autonomous visual perception is relaxed by employing the Cellular Neural Networks. Experimental results of the autonomous visual perception design are shown.