Method used for generating high dynamic range image with single low dynamic range image

Abstract

The invention proposes a method used for generating a high dynamic range image with a single low dynamic range image. The method comprises the steps that based on the human visual characteristics, the luminance component and the chrominance component of the low dynamic range image are separated; gamma correcting and edge-preserving filtering are carried out on the luminance component; the basic layer of the luminance component is extracted; traverse operating is carried out on the basic layer and the luminance component to acquire the detail layer of the luminance component; an anti-tone mapping function is constructed; anti-tone mapping operation is carried out on the detail layer and the gamma-corrected luminance image to acquire the respective anti-color collimated image; the luminance component after anti-tone mapping and the detail layer are merged to acquire a new luminance component; the chrominance component and the new luminance component are merged; and contrast stretching and denoising are carried out on the merged image to acquire the final high dynamic range image. According to the invention, the high dynamic range image is acquired through the single low dynamic range image, and the method has the advantages of good processing effect, high operability and wide applicability.

Description

technical field [0001] The invention relates to image processing technology, in particular to a method for generating a high dynamic image of a single image by layered extraction and re-merging based on detailed information of a single low dynamic range image. [0002] technical background [0003] Over the past two decades, high dynamic range (HDR) image capture, introduced by the field of computer graphics, has revolutionized this and other fields such as photography, virtual reality, visual imaging, video games etc. HDR imaging can directly capture and utilize the actual physical value of brightness. During the imaging process of the image, very dark and very bright areas may be recorded in the same frame of image or the same video at the same time. Using the actual physical value of brightness to represent the image can avoid the appearance of underexposed and overexposed areas in the image. Traditional image acquisition methods do not use the actual physical value of b...

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to solve the problem of generating high dynamic range images from low dynamic range images, and provide a method for directly generating high dynamic range images from a single low dynamic range image, through which a high dynamic range image can be obtained

Images