Image noise positioning method and system

Abstract

The invention discloses an image noise positioning method and system, relates to the field of image analysis, and is used for realizing learning training of a model by constructing sample data, selecting a remote sensing image to be subjected to quality inspection, calculating and analyzing, and judging the severity of strip noise in the image. The image noise positioning method and system has thebeneficial effects that the strip noise in the remote sensing image is detected and positioned quickly and efficiently, the position of the strip noise in the remote sensing image can be positioned objectively and quickly, the severity of the strip noise in the image is judged, and strip noise positioning information is provided for strip denoising quickly. Mass remote sensing data processing isfacilitated, so that the production efficiency of remote sensing image data is improved.

Description

technical field [0001] The invention relates to the field of image analysis, in particular to an image noise positioning method and system. Background technique [0002] Optical remote sensing satellite imaging generally adopts the CCD linear array push-broom method. Due to the difference in response between each detector element of the detector, band noise is likely to be generated during push-broom imaging, which seriously affects the quality of image products and subsequent applications. Strip noise removal and result evaluation for images are important steps in radiation processing and quality inspection. In the process of image radiation quality inspection, the current classic stripe noise evaluation methods, such as stripe coefficient, column mean method, etc., are difficult to accurately quantify and measure the overall stripe noise and removal effect of the image, making the evaluation results of radiation quality inspection sometimes inaccurate Indicates the actual...

AI Technical Summary

Problems solved by technology

In the process of image radiation quality inspection, the current classic stripe noise evaluation methods, such as stripe coefficient, column mean method, etc., are difficult to accurately quantify and measure the overall stripe noise and removal effect of the image, making the evaluation results of radiation quality inspection sometimes inaccurate Characterize the actual quality of the image

In this case, the method of manual visual interpretation is often required to evaluate whether there are bands in the image, which seriously affects the work efficiency, and is easily affected by human subjectivity, which is not conducive to actual work needs.

Currently, there is no related research on band noise location in relevant literature, and most of the existing band noise removal algorithms are based on radiometric calibration parameters to denoise full-scale remote sensing images, and assume that the detector response is linear, regardless of Local differences of different ground features

It is easy to generate local band noise after processing

Images