Method for scene matching region selection and reference image optimization of image matching system

Abstract

The invention relates to a method for scene matching region selection and reference image optimization of an image matching system, which comprises the steps of performing initialized blocking on a remote sensing satellite image, extracting edge features of each region according to an edge feature extraction algorithm, and screening block regions with centralized edge gradient points; calculating the autocorrelation degree of each region by using a repetition pattern index measurement method, sorting the regions from small to large according to an intracorrelation index, and outputting an optimal candidate region; building a spatial feature vector of a reference image of the candidate region by adopting a spatial distribution description method; building a scene matchability measurement evaluation index set, analyzing the relevance between the evaluation index set and the matching probability, and outputting a matching performance evaluation index set; and statistically analyzing matched and correlated aspects of the reference image and the satellite image, and optimizing the reference image according to a reference image optimization method. High matching performance and high reliability are realized by using the optimized reference image when the matching system operates in an unknown environment, and the image matching system is guaranteed to complete a task successfully.

Description

technical field [0001] The invention relates to the technical field of image processing system performance evaluation, in particular to a scene matching area selection and reference map optimization method of an image matching system. Background technique [0002] In the scene matching system, the essence of the matching process is to calculate the similarity between the real-time image and the reference image. Therefore, the quality of the reference image is closely related to the scene matching accuracy. As the pre-planning information, the reference map mainly comes from remote sensing images. There are many differences in time and imaging system, which seriously affect the quality of the reference map. It is of great significance to improve the performance of the scene matching system to measure the evaluation index and criterion of high matching, and to select the matching area and optimize the reference map according to these indexes. [0003] The specific process of ...

AI Technical Summary

Problems solved by technology

However, in an actual unknown environment, it is difficult to obtain homogeneous infrared image sequences, while heterogeneous images of visible light are very easy to obtain, such as satellite images and aerial images, and the analysis errors caused by this imaging system are less concerned. , these lead to a decrease in the matching probability

This requires the establishment of a reference image selection and evaluation method for heterogeneous images, which can adapt to the preparation conditions of heterogeneous images, and at the same time have the function of optimizing reference images in advance, so as to promote the preparation of reference images to approach the goal of high quality. The method of preparing and optimizing the reference map has not been published yet

Images