Star map background suppression method and device, computer equipment and storage medium

Abstract

The invention relates to a star map background suppression method and device, computer equipment and a storage medium. The method comprises the steps that after an original image is divided into blocks, pixel points with the iterative cutting result equal to the gray value of the original image are selected through a sigma iterative cutting algorithm to serve as fitting seed points, fitting parameters are obtained according to a fitting seed point set, the star map background is estimated through polynomial fitting, and the star map background is obtained. Carrying out primary suppression on the background of the original image by utilizing the estimated star map background; calculating a minimum gradient map of the preliminary background suppression image, determining an extraction threshold according to the minimum gradient map and a local region pixel value of any pixel point, and determining a dark local space channel of the preliminary background suppression image according to the extraction threshold; and performing enhanced suppression on the background of the preliminary background suppression image according to the dark local space channel to obtain an enhanced background suppression image. According to the method, the purposes of suppressing the background and improving the signal-to-noise ratio of the target can be achieved, and the effect of the method is insensitive to suppression parameters.

Description

technical field [0001] The present application relates to the technical field of image analysis, and in particular, to a method, device, computer equipment and storage medium for suppressing the background of a star map based on polynomial fitting and dark local space channels. Background technique [0002] Star map background suppression is a key preprocessing technique that determines the detection of space debris in optical images. A good background suppression effect is the premise to ensure the effective detection of weak and small space debris. [0003] The background of the star map is mainly composed of stray light such as moonlight, mist, artificial light pollution, and earth-atmosphere light (space-based). Compared with the background stars, debris targets, and radiation noise, the background appears in the airspace as a large fog with gentle fluctuations. The patches are mainly low frequency components in the frequency domain. Since space debris targets are typic...

AI Technical Summary

Problems solved by technology

Specifically, the threshold segmentation method is simple to calculate, but it is only suitable for occasions with small background fluctuations, and it is easy to mistakenly suppress weak and fragmented targets. This method is usually used in conjunction with other methods; Sigma iterative clipping is actually a variable threshold segmentation method. This method alleviates the shortcomings of the fixed threshold segmentation method to a certain extent, but it also has the problem of wrongly suppressing weak and small fragment targets; both median filtering and morphological filtering belong to statistical sorting filters, which are relatively more effective. Commonly used star map background suppression methods, this type of method needs to set a suitable filter kernel size in advance, and its suppression effect is very sensitive to the filter kernel size: if the filter kernel size is too large, it will greatly increase the amount of calculation, and the suppression effect is not good, and the filter kernel is too small Debris or star targets will be suppressed, and this type of method cannot preserve the grayscale and edge characteristics of the target well, which will affect the centroid positioning accuracy in the later stage

To sum up, the currently commonly used star map background suppression methods mainly have the following two shortcomings: 1. It is difficult to balance the contradiction between background suppression and the signal-to-noise ratio improvement of weak and small debris targets, that is, when the background suppression degree is large, it is easy to cause the energy of debris targets to The loss and even some weak targets are directly suppressed; Second, the suppression effect is too sensitive to the suppression parameters, which limits the universality and intelligence of the algorithm

Images