Contour detection method based on primary visual cortex fixation micro-motion mechanism

Abstract

The invention aims to provide a contour detection method based on a primary visual cortex fixation micro-motion mechanism. The method comprises the following steps: A, inputting a to-be-detected imageobtained through gray processing, presetting Gaussian first-order derivative functions of a plurality of direction parameters, presetting axial offset of a template center of the Gaussian first-orderderivative functions, presetting offset centers of four Gaussian first-order derivative functions, and calculating and obtaining four corresponding Gaussian first-order derivative offset functions; b, for each pixel point of the to-be-detected image, calculating and obtaining a classic receptive field response of each pixel point; c, presetting a distance weight function and an inhibition coefficient, and performing calculation to obtain inhibition response of each pixel point; and D, subtracting the suppression response of the pixel point from the classical receptive field response of each pixel point to obtain a contour response of each pixel point, and further obtaining a final contour map. The contour detection method overcomes the defects in the prior art, and has the characteristicsof strong simulation and high detection accuracy.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a contour detection method based on a micro-movement mechanism of primary visual cortex. Background technique [0002] Contour detection is a basic task in the field of computer vision. Unlike edges, which are defined as strong brightness changes, contours usually represent the boundaries of one object to other objects. Contour detection models inspired by biology are currently one of the mainstream research directions, but most of them simply simulate a part of the physiological characteristics of the visual system, and the lack of research on the role of the fixation micro-movement mechanism in contour detection has caused a certain degree of Due to the lack of upper contour information and the enhancement of texture information, the integrity of the target contour cannot be well guaranteed. The few schemes that consider the physiological mechanism of fixation fretting focus on...

AI Technical Summary

Problems solved by technology

Contour detection models inspired by biology are currently one of the mainstream research directions, but most of them simply simulate a part of the physiological characteristics of the visual system, and the lack of research on the role of the fixation micro-movement mechanism in contour detection has caused a certain degree of The lack of upper contour information and the enhancement of texture information can not guarantee the integrity of the target contour.

However, the few schemes that consider the physiological mechanism of fixation fretting focus on the effect of texture suppression in the non-classical receptive field area, while ignoring the application in the classical receptive field area of ​​visual cells.

Images