Infrared and visible light image synchronous fusion noise reduction method based on side window filtering and multi-scale transformation

Abstract

The invention relates to an image fusion method, in particular to an infrared and visible light image fusion method, and particularly relates to an infrared and visible light image synchronous fusion noise reduction method based on side window filtering and multi-scale transformation. Respectively obtaining corresponding base layer and detail layer information; extracting a base layer salient region of the infrared image by using a self-adaption-based power transformation method; fusing the extracted infrared image base layer region and the visible light image base layer by using a fusion rule to obtain a base layer fusion image; performing synchronous fusion noise reduction processing on the detail layer image by adopting a method based on total variation to obtain a base layer fusion image; and reconstructing the obtained base layer fusion image and the detail layer fusion image to obtain a final fusion image, the contrast of the fusion image is obvious, the salient region is prominent, and the noise is greatly reduced.

Description

technical field [0001] The invention relates to an image fusion method, in particular to an infrared and visible light image fusion method, in particular to a synchronous fusion noise reduction method for infrared and visible light images based on side window filtering and multi-scale transformation. Background technique [0002] Infrared and visible light images contain different information, and image fusion technology can combine complementary information about the same scene captured by different sensors into a more comprehensive image to help subsequent processing. In the process of image acquisition, it is inevitable that the image quality will be degraded due to the interference of noise, which will seriously affect the subsequent tasks. Therefore, image denoising has important practical significance. For source images containing noise, although existing image fusion methods can integrate information for fusion, the resulting fused image is still noisy, so it is not ...

AI Technical Summary

Problems solved by technology

For source images containing noise, although existing image fusion methods can integrate information for fusion, the resulting fused image is still noisy, so it is not conducive to subsequent target recognition.

[0003] At present, if it is necessary to fuse images containing noise, it is necessary to perform noise reduction processing before or after image fusion, which not only takes time and many steps are cumbersome, but also causes the loss of original image details and affects the fusion effect.

Images