High-dynamic-range image quality evaluation method based on cumulative gradient similarity

Abstract

The invention relates to a high-dynamic-range image quality evaluation method based on accumulated gradient similarity. the method comprises the following steps: calculating the gradient similarity between a low-dynamic-range image sequence and a high-dynamic-range image, obtaining the accumulated gradient similarity of gradient blocks of the low-dynamic-range image and the high-dynamic-range image, dividing the cumulative gradient similarity into a static image block and a ghost image block by adopting a K-means clustering method, calculating the gradient similarity of the static image block and the ghost image block respectively, and finally, fusing the static image block and the ghost image block to obtain an image quality evaluation result of the high-dynamic-range image. The high-dynamic-range image is divided into the static image block and the ghost image block by adopting the accumulated gradient similarity between the low-dynamic-range image sequence and the high-dynamic-range image, and the gradient similarity is calculated and fused to obtain the image quality evaluation result of the high-dynamic-range image, so the high dynamic range image quality evaluation precision is improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a high dynamic range image quality evaluation method based on cumulative gradient similarity. Background technique [0002] Image quality evaluation is a key issue in the field of image processing. Image quality evaluation methods can be divided into subjective image quality evaluation methods and objective image quality evaluation methods according to whether people participate. Subjective image quality evaluation methods are scored by humans, and the evaluation results are accurate, but the evaluation process is complex, time-consuming, and difficult to be applied in real time. The objective image quality evaluation method does not require human participation, and the image quality is automatically predicted by a specific computer algorithm. According to whether the original undistorted image is used as a reference, the image quality evaluation method can be divided in...

AI Technical Summary

Problems solved by technology

[0003] High dynamic range (HDR) images display a larger brightness range, which can bring better visual experience to viewers, but most of the existing graphics display devices only support displaying 8-bit low dynamic range (LDR) images, so in the production For HDR images, the LDR image sequence needs to be fused into an HDR image using a multi-exposure fusion algorithm, but the dynamic scene in the LDR image sequence causes ghost areas in the HDR image, so a high dynamic range de-ghosting algorithm needs to be used for dynamic sources The exposure-level sequence synthesizes a ghost-free high dynamic range image, and the de-ghosting effect obtained by each algorithm is different

However, little work has been done on image quality assessment of deghosted HDR images

The most reliable method for evaluating ghosted high dynamic range images is subjective evaluation, but subjective evaluation takes a long time and is difficult to be applied in real time. Therefore, it is necessary to develop an objective quality evaluation algorithm for high dynamic range image synthesis and dynamic scene multi-exposure image fusion

Images