Parallel image color enhancing method and device

Abstract

The invention relates to a parallel image color enhancing method, comprising: step 1) loading image data; calculating the three-dimensional look-up table coefficients and the interpolation weight wherein the three-dimensional look-up table coefficients are the index coefficients corresponding to the entries of the look-up tables for each separate color channel; step 2) based on the look-up tables and through the operations on the look-up tables, obtaining the RGB thee-channel look-up table results required for interpolation; and step 3) based on the look-up table results of the step 2 and the interpolation weight, calculating the RGB values of each interpolation points and outputting the result. The invention also discloses a parallel image color enhancing device, which comprises a local memory, an access control unit, a buffer, a parallel arithmetic logic unit, a parallel multiplying accumulator, a state machine and a data and a data interleaving unit. With the method and the device of the invention, the image colors are enhanced; the utilization efficiency of data is increased; the data interaction between a computation component and a peripheral memory is reduced; the pressure on an access broadband is reduced; and reutilization of hardware resources is realized.

Description

technical field [0001] The invention belongs to the field of video image processing, and in particular relates to a parallel image color enhancement method and device. Background technique [0002] At present, one of the mainstream development directions of video technology is ultra-high-definition (4K resolution) display technology. Compared with high-definition (1920*1080) video, the number of pixels of 4K video has been increased from 2M to 8M, so higher requirements are placed on the image quality and performance of image enhancement algorithms. [0003] At present, image color enhancement technology is basically based on three-dimensional interpolation. When performing enhancement, it is necessary to realize color transformation through table lookup and interpolation. For RGB images, each pixel corresponds to three RGB channels, and table lookup 3×8=24 is required times, and 8 weights need to be calculated, and each weight needs to be multiplied twice, which requires a...

AI Technical Summary

Problems solved by technology

[0003] At present, image color enhancement technology is basically based on three-dimensional interpolation. When performing enhancement, it is necessary to realize color transformation through table lookup and interpolation. For RGB images, each pixel corresponds to three RGB channels, and table lookup 3×8=24 is required times, and 8 weights need to be calculated, and each weight needs to be multiplied twice, and the amount of calculation is relatively large. When faced with 4K image processing requirements, it is very likely that the processing capacity is insufficient.

[0004] Therefore, traditional solutions face two difficulties in the ultra-high-definition era. 1. Traditional solutions usually use dedicated hardware circuits with solidified algorithms, making it difficult to flexibly modify the algorithms. When the scheme implements parallel table lookup, it usually uses multiple memory access channels to work in parallel, which increases the hardware cost

Images