Based On The Visual Receptive Field Properties Of The Object Contour Extraction Algorithm

Contour extraction requests to identify objects in complex scenes by eliminating texture and distinguishing contours from non-contour edges. Fortunately, the human visual system (HVS) can extract this contour feature effectively. The broad region outside the classical receptive field (CRF) of a neuron in the primary visual cortex (V1), namely non-CRF (NCRF), exerts robust modulatory effects on the responses to visual stimuli presented within the CRF. This modulating effect is mostly suppressive which plays important roles in visual information processing. One possible role is to extract object contours from disorderly background textures.In this work we propose two contour detection methods based on visual mechanisms. The first one is based on an improved Edge preserving smoothing and surround inhibition model. A butterfly-formed surrounding area is employed for the computation of inhibition term, which could provide a flexible inhibitory effect for the NCRF modulation on CRF. Comparisons with other visual contour detection models show that the proposed model can suppress texture effectively and extract contour more correctly.Inspired by the inhibitory interactions between CRF and NCRF of V1 neurons, the other model is based on dynamic inhibition. The kernel idea is that the side and end sub-regions of NCRF work in different manners, i.e., while the strength of side inhibition is consistently calculated just based on the local features in the side-regions at a fine spatial scale, the strength of end inhibition adaptively varies in accordance with the local features in both end and side-regions at both fine and coarse scales. Simulation results demonstrate that by introducing such an adaptive mechanism into the model, the non-meaningful texture elements are removed dramatically, and at the same time, the object contours are extracted effectively. Besides the superior performance in contour detection over other inhibition based models, this model provides a better understanding on the roles of NCRF and has potential applications in computer vision and pattern recognition.