Visual perception method and system based on biological neural network and stochastic resonance

Abstract

The invention discloses a visual perception method and system based on a biological neural network and stochastic resonance. The stochastic resonance principle of an integral discharge neural network of synaptic conductance and a basic biophysical process of visual formation are combined. Two types of nerve cells, namely rod cells and cone cells, are known to be mainly distributed on the retina; the sight rod cells are mainly responsible for distinguishing the rough contour of an object but cannot distinguish the color, and the sight cone cells are sensitive in light and have high color distinguishing capacity. In order to enhance the contrast ratio of a color image, a biological neural network is used for simulating the cooperation effect of a rod cell cluster in a visual perception process, and a new color image enhancement method is developed. According to the method, the image contrast and brightness can be remarkably improved, the details of the dark area are also remarkably enhanced, the edge information is also remarkably improved, and a better visual effect is achieved; and compared with a classical single-scale Retinex method and an HE algorithm, the method also shows obvious superiority.

Description

technical field [0001] The invention belongs to the technical field of image enhancement, and in particular relates to a visual perception method and system based on a biological neuron network and stochastic resonance. Background technique [0002] Visual perception technology or image enhancement technology is currently a hot research topic in the field of image processing, and it is widely used in military night vision, road traffic, surveillance video, brain-computer interface and other engineering fields related to weak signal detection. [0003] Most of the traditional image enhancement methods focus on "denoising", but under the strong background noise, some useful information of the weak image is often weakened in the process of noise reduction, resulting in the image enhancement algorithm for the purpose of noise reduction often There are inevitable flaws. The principle of stochastic resonance provides a new way of using noise to highlight the characteristics of we...

AI Technical Summary

Problems solved by technology

[0004] However, the existing stochastic resonance algorithms for image enhancement mainly consider the enhancement of grayscale images, and most of the existing algorithms lack the details of algorithm implementation and do not have sufficient biological interpretability

For example, visual perception algorithms based on simple threshold models or overdamped bistable models do not give key parameters such as threshold selection details and evaluation indicators for optimal target images; based on the combination of singular value decomposition and overdamped bistable models image enhancement algorithm, also lacks key details, and the stochastic resonance algorithm based on total variation regularization and overdamped bistable model is inconvenient to use and understand because the algorithm structure is too complex

Images