Wide-range remote-sensing image automatic dodging and color uniformizing method based on standard color library

Abstract

A wide-range remote-sensing image automatic dodging and color uniformizing method based on a standard color library comprises the steps that according to subregions formed by dividing a whole range, images which cover all the subregions in all seasons and are obtained through remote sensing of sensors are collected; the images are grouped according to the sensors, resolution and time phase information, and the images of the same kind of sensors, the same kind of resolution and a certain time phase in all the subregions are in one group; absolute radiation processing is carried out, and the images different with the adjacent images in color are adjusted; all the standard color images in all the groups are spliced respectively, and all spliced full-width standard images form the standard color library; firstly, dodging and color statistics are carried out on all single-width images to be processed, then standard color information of related time phases in corresponding geographical ranges under proper space scales and sensor kinds is extracted from the standard color library, the hue of each single-width image is adjusted according to the standard color information, and finally the adjusted images are spliced to form a wide-range image.

Description

technical field [0001] The invention belongs to the field of surveying and mapping science and technology, and relates to an automatic light and color uniform method for large-scale remote sensing images based on a standard color library, which is mainly applied to the production of large-scale high-quality digital orthophoto images and the production of large-scale high-quality seamless remote sensing image libraries Create and other fields. Background technique [0002] With the continuous development of remote sensing technology, remote sensing images play an increasingly important role in many fields. Due to the limited coverage of a single image, it cannot effectively meet the task requirements in many cases. Therefore, it is necessary to stitch multiple images or even multiple heterogeneous images to form a large-scale image. However, during splicing, it can be found that due to the influence of factors such as the imaging method, imaging time, light source, and terra...

AI Technical Summary

Problems solved by technology

The linear transformation method considers the color consistency of multiple images in the area as a whole, and the processing results do not depend on the order of the images, but there are still color differences in local areas, and color distortion is prone to occur

For example, the uniform color processing based on the Wallis filter is a local image transformation, which adjusts the variance and mean value of the gray value of different images or different positions of the image to make them have approximately equal values. When the measurement area uses Wallis When the operator adjusts the color difference, it usually selects a representative image in the measurement area as the reference image of the color tone. The difference is very large, resulting in color cast of the processed image

The theoretical basis of the variance-mean method is that the two images have a gray level difference in the sense of least squares, which usually reduces the local contrast of the image to be processed

Histogram matching, also known as histogram normalization, is to statistically adjust the gray distribution of two images to make them as close as possible, but if the distribution of some gray levels of the image is too concentrated, color distortion is prone to occur

Generally speaking, the phenomenon of color deviation of remote sensing images can be solved to some extent after the above-mentioned two-step method, but they also have some problems. Special areas, such as water bodies, need to be manually edited to obtain satisfactory results, and the degree of automation is not high. Second, some existing algorithms often select a picture in the measurement area when performing color uniformity between images. or several images with representative hues as tone reference images, the result is likely to lead to the accumulation of color deviation and the appearance of systematic color deviation in the process of color uniformity; the third is when performing color uniformity of multiple images, especially heterogeneous images , the lack of standard color information, the selection of different control images will have different results, leading to the freedom of color uniformity, which not only affects the quality of color uniformity, but also is not conducive to the subsequent splicing of images in different regions, hindering the integration of large-scale images generate

Images