Contour Detection Method Based on Nonclassical Receptive Field and Linear and Nonlinear Modulation

Abstract

The present invention provides a contour detection method based on non-classical receptive fields and linear nonlinear modulation, including: A, inputting the image to be detected after grayscale processing; B, performing Gabor filtering on the image to be detected to obtain Gabor energy in each direction of each pixel C. Construct the simulation model of X cells and Y cells in retinal ganglion cells; D. Calculate the response of each central neuron corresponding to the X cell to the stimulus response of the non-classical receptive field; E. Calculate the response of each central neuron corresponding to the Y cell Neurons are stimulated by non-classical receptive fields; F, respectively calculate the stimulus responses of the central neurons of X and Y cells modulated by the classic and non-classical receptive fields, and add them up as the corresponding contour values; G, The contour value of each pixel is processed to obtain the final contour value. The method overcomes the defect of low contour recognition rate in the prior art, and has the characteristics of good simulation effect and high contour recognition rate.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a contour detection method based on nonclassical receptive field and linear nonlinear modulation. Background technique [0002] The contour defines the shape of the target, and the contour is one of the important tasks in target recognition, and obtaining the target contour from the cluttered scene is an important and quite difficult task, mainly because there are usually a large number of textured background edges around the contour, so This work mainly needs to exclude nonsensical edges due to textured regions while preserving object contours. The key to improving the detection rate is to optimize the integration of local information into consistent global features based on context. The human visual system has the ability to quickly and effectively extract contour features from complex scenes, which effectively promotes the development of contour detection algorithms inspired b...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims to provide a contour detection method based on non-classical receptive field and linear nonlinear modulation, which overcomes the defects of poor simulation effect and low contour recognition rate in the prior art, and has the characteristics of good simulation effect and high contour recognition rate

Images