Image interference degree evaluation method based on edge strength similarity

Abstract

The invention discloses an image disturbed degree evaluation method based on the edge intensity similarity, and the method comprises the steps: collecting a disturbed image through a CCD camera, carrying out the preprocessing of the collected image, and converting the image into a two-dimensional numerical matrix from an RGB space; step 3, extracting a target image from the obtained interfered image, carrying out partitioning processing on the whole image, and enabling the size of each small image to be the same as the corresponding size of the target image extracted in the step 2; calculatinga direction coefficient of each image small block; calculating the edge intensity of each pixel point in the target image and each small image; and on the basis of the calculation step, calculating the edge intensity similarity ESM of each small image and the whole image. The method can achieve the quick evaluation of the interference degree of the image under a complex background condition, is good in robustness, and is high in calculation precision.

Description

technical field [0001] The invention relates to an image interference degree evaluation technology, in particular to an image interference degree evaluation method based on edge strength similarity. Background technique [0002] As the "eye" of modern information acquisition, the main task of photoelectric imaging is to collect image information in a specific area and provide data information for subsequent analysis and processing of specific targets in the image. At the same time, due to complex conditions, photoelectric imaging systems are particularly vulnerable to Interference from the external environment, such as: sunlight, lights, various stray lights, and various artificial strong lights, etc. These interferences can cause local unclear images, and heavy interference can cause large-scale pixel saturation of the image, thereby This makes subsequent target analysis and processing difficult. How to judge the degree of interference of an image is the premise of studyin...

AI Technical Summary

Problems solved by technology

These evaluation methods are generally only suitable for describing the size of clutter interference in photoelectric images with relatively simple scenes. For photoelectric images with complex interference backgrounds, the calculation accuracy drops seriously, so it is impossible to quantitatively evaluate the photoelectric interference image quality.

Images