HDR video acquisition method and device, electronic equipment and storage medium

Abstract

The invention provides an HDR video acquisition method and device, electronic equipment and a storage medium. The HDR video acquisition method comprises: acquiring a multi-exposure RAW map and performing noise reduction on each RAW map; recovering the RAW image after noise reduction to an RGB image; aligning a plurality of continuous frames of RGB images, and performing gamma correction and linearization on each frame of RGB image to obtain a corresponding gamma RGB image and a linear RGB image; and inputting gamma RGB images and linear RGB images corresponding to continuous multi-frame RGB images into an HDR fusion neural network model to obtain a fused clear HDR image, wherein gamma correction and linearization are performed on each frame of RGB image after the multi-frame RGB images are aligned in combination with data of an accelerometer and a gyroscope to obtain the corresponding gamma RGB images and linear RGB images, and then fusion is performed, so that the problems of artifacts and jitter can be reduced, the generalization ability is improved, and the image quality precision is further improved.

Description

technical field [0001] The present invention relates to the technical field of video image processing, and in particular to an HDR video acquisition method, device, electronic equipment and storage medium. Background technique [0002] With the continuous improvement of imaging system performance, consumers have increasingly strong demands for high-quality High Dynamic Range (High Dynamic Range, HDR) photos and videos at night, HDR photos and videos in backlit environments, and high-quality HDR content Shooting equipment is expensive and requires complex post-processing, which has a high threshold. A lot of research has been done so that consumers can take high-quality HDR photos and videos on their phones. [0003] In the field of digital images, the dynamic range (Dynamic Range) represents the direct ratio between the maximum brightness and the minimum brightness within the displayable range of the image. Among them, the ratio of the maximum luminosity value to the minim...

AI Technical Summary

Problems solved by technology

[0007]Multi-exposure image fusion technology usually performs HDR fusion in the RGB domain after the ISP, which effectively reduces the production cost and difficulty of high dynamic range content. However, the multi-exposure RGB image Acquisition is prone to artifacts and jitter problems due to time delays such as exposure time and data transmission. In addition, the generalization ability of the existing HDR fusion technology is insufficient, resulting in the need to improve the image quality and accuracy, and the calculation of the AI_HDR fusion algorithm is too large. Unable to deploy on mobile

Images