Biological visual processing for hybrid-order texture boundary detection with CNN-UM

This paper investigates a novel biological visual processing for hybrid-order texture boundary detection. The texture boundary detection is based on the first- and second-order features to model the pre-attentive stage of a human visual system. This system is implemented by a cellular neural network universal machine (CNN-UM) with 3×3 templates to approximate desired filter transfer functions. The system design can process a 64×64 gray-scale image. The proposed algorithm can successfully be performed by CNN-UM and detect the texture boundary in a given image.