A method and system for relative radiometric correction of remote sensing images

Abstract

The invention discloses a method and system for relative radiation correction of remote sensing images. First, a target image is acquired, and a reference image corresponding to the target image is determined based on the target image; secondly, the reference image and the target image are pre-set respectively. After processing, the reference image pixel point set and the target image pixel point set are obtained; then based on the kernel canonical correlation analysis method, the nonlinear radiation control points are determined according to the reference image pixel point set and the target image pixel point set; finally, according to the The nonlinear radiation control points correct the pixels at each stage of the target image to obtain the surface reflectance value. This invention determines nonlinear radiation control points based on the kernel canonical correlation analysis method, and overcomes the traditional problem of large radiation differences in the images themselves caused by large differences in time phases and geometric differences in sensor imaging, making it difficult to ensure radiation consistency between images. , greatly improving the utilization rate of remote signaling data.

Description

technical field [0001] The invention relates to the field of audio and video technology, in particular to a method and system for relative radiation correction of remote sensing images. Background technique [0002] With the rapid development of remote sensing technology, the use of multi-sensor and multi-temporal remote sensing data for land use and land cover change monitoring, global resource and environment analysis, and climate change monitoring has become more and more widely used, broadening the scope of remote sensing data in time and space. Utilization, such large-area and large-dimensional data applications often have very high requirements for data set consistency, even higher than the accuracy requirements for a single image. [0003] For multi-scene data processing, due to the processing error of each scene data, the difference in acquisition time, and the difference in sensors, etc., after splicing, a "boundary" phenomenon appears near the data splicing line. T...

AI Technical Summary

Problems solved by technology

[0004] Although the existing relative radiation correction method can better ensure the radiation consistency between images, there are strict constraints on the image phase: the imaging time of the reference image and the target image are required to be similar, that is, the requirement can be different years, However, the month is required to be within one month, and the imaging time is within one hour. This constraint greatly limits the application of remote sensing data on a national or even global scale for land use and resource analysis. Due to differences and other reasons, the radiation of the image itself is quite different, and it is difficult to ensure the consistency of radiation between images, thus greatly reducing the utilization rate of remote signaling data

Images