Encoding method, image processing device and storage device based on intra-frame prediction

Abstract

The present application discloses an encoding method based on intra-frame prediction, an image processing device, and a storage device. The method includes: obtaining an original luminance block to be encoded; determining at least two candidate sub-block division methods of the original luminance block to be encoded; calculating The cost value of each candidate sub-block segmentation method; select the one whose cost value is not the highest from at least two candidate sub-block segmentation methods, as the selected sub-block segmentation method; use the selected sub-block segmentation method to convert the original brightness to be encoded The block is divided to obtain at least two sub-blocks to be coded; the at least two sub-blocks to be coded are respectively coded to obtain coded sub-blocks. Through the above method, the present application can effectively improve prediction accuracy and compression performance.

Description

technical field [0001] The present application relates to the technical field of image encoding, in particular to an encoding method based on intra-frame prediction, an image processing device and a storage device. Background technique [0002] As users have higher and higher requirements for video clarity, the amount of video image data is also increasing. In order to reduce the burden of transmitting video image data, video encoding is performed on video image data before transmission to compress the size of video image data. . The process of video coding includes prediction, transformation, quantization, coding, etc. [0003] The currently mainly used prediction modes include intra-frame prediction, which is to predict the pixel of the current pixel point in a frame of image according to the pixels of the surrounding pixel points (ie, reference pixels). Among them, sub-block partition (IntraSub-Partitions, ISP) is an important part of intra prediction. [0004] In the ...

AI Technical Summary

Problems solved by technology

Predictive accuracy and compressibility are limited

[0005] In addition, in the prior art, when predicting the remaining sub-blocks in the current CU block except the first sub-block, it is necessary to wait for the reconstruction of the previous adjacent sub-block to complete, and use the reconstruction of the previous sub-block The pixel is used as the reference pixel of the current sub-block for prediction, and the operation efficiency is low

Images