Multispectral mosaic image restoration method

Abstract

The invention relates to a multispectral mosaic image restoration method, which can restore a multispectral mosaic video image shot by adopting a coated video spectrograph to a complete multispectralimage with high spatial resolution and high spectral resolution, and solves the problem of application limitation due to multispectral image real-time transmission. The method comprises the steps of 1) determining a mosaic template according to a spectral band quantity in a pixel matrix block of an original multispectral mosaic image S; 2) extracting single spectral band images S1; 3) performing down-sampling on the single spectral band images S1 to obtain images S2; 4) performing two-fold up-sampling on the images S2 to obtain images S3; 5) performing interpolation operation on the images S3;and 6) judging restoration conditions of the single spectral band images S1, and finishing the restoration of all the single spectral band images.

Description

technical field [0001] The invention relates to a multispectral image processing technology, in particular to a multispectral mosaic image restoration method. Background technique [0002] The research and exploration of super-resolution reconstruction and spectral reconstruction in multispectral image processing has always been an important content in the application and development of multispectral image technology. Multispectral image is a high-resolution remote sensing image that integrates image graphics and spectroscopy, and its spatial resolution and spectral resolution are higher than ordinary remote sensing images. Compared with ordinary digital images, multispectral images have spatial information and spectral information of ground objects, and can simultaneously obtain geometric shape information and spectral characteristic curves of objects. However, due to the large number of spectral segments, high spatial resolution, large amount of data, high information red...

AI Technical Summary

Problems solved by technology

[0004] 1. High spatial resolution and low spectral resolution, suitable for handheld multispectral imaging with fewer types of ground objects and high spatial resolution requirements, such as multispectral camera imaging technology

[0005] 2. Low spatial resolution and high spectral resolution, suitable for airborne multispectral imaging with many types of ground objects and large and regular target geometry, such as: farmland crop detection, forest monitoring, etc.

[0009] However, the image data captured by the coated multispectral video imager cannot be directly applied to analysis and processing, because this type of imaging technology compresses multidimensional multispectral image data into a two-dimensional data format, and the output image has a mosaic effect. The outline is blurred, and the details of the object are seriously lost

The spatial resolution of a single-spectrum image will decrease as the number of spectral segments increases, and the missing spectral information of each pixel will also increase, especially when the number of spectral segments is large, this shortcoming will be more obvious

Images