Lossy compression coding method, device and system-on-chip

Abstract

The invention discloses a lossy compression coding method, a device and a system-on-chip, and belongs to the technical field of image processing. The method comprises the steps of determining a regionof interest in a to-be-coded image, wherein the region of interest is an image region where a target object is located in the to-be-coded image; carrying out local noise reduction processing on the to-be-coded image to obtain a to-be-coded image after local noise reduction processing, the to-be-coded image after local noise reduction processing comprising a noise reduction region and a non-noisereduction region, the non-noise reduction region comprises the region of interest; and performing lossy compression coding on the to-be-coded image after the local noise reduction processing. The device comprises a determination module used for determining a region of interest in a to-be-coded image; a processing module which is used for carrying out local noise reduction processing on the to-be-coded image; and an encoding module which is used for carrying out lossy compression encoding on the image to be encoded after local noise reduction processing. According to the method, the size of thecoded file can be reduced after lossy compression coding; and meanwhile, the definition of the region of interest can be reserved after lossy compression decoding.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a lossy compression encoding method, device and system-on-chip (System on Chip, SOC for short). Background technique [0002] The lossy compression coding method is widely used in the compression processing of image and video data. Its basic principle is to lose some information in the compression process in exchange for a larger compression ratio and reduce the file size after compression. For example, Joint Photographic Experts Group (JPEG for short) is a lossy compression encoding method based on discrete cosine transform (Discrete Cosine Transform, DCT for short). In surveillance applications, cameras capture images based on specific trigger events. The captured images are usually encoded with JPEG first, and then stored or transmitted. [0003] Certain monitoring scenarios will generate a large number of snapshots, for example, the vehicle monitoring bay d...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a lossy compression encoding method, device and system-level chip, which can solve the problem that although lossy compression encoding methods such as JPEG can reduce the size of the encoded image, the quality of the decoded image is relatively poor

Images