Fast image processing method for high definition camera

Abstract

The invention provides a fast image processing method for a high-definition camera, which pertains to the field of image processing and comprises the steps of firstly defining and generating a bit adaptive mapping table, an RGB color mapping table and a brightness mapping table; then carrying out statistics to the color and brightness information of the image, determining regulatory factor and brightness of R/G/B, and amending the RGB color mapping table and the brightness mapping table; and then converting the collected image, carrying out R/G/B color mapping, mosaic removing, format conversion and brightness mapping, carrying out filtration to the brightness according to the expected indexes, and generating YCbCr data stream; finally converting the image to a JPEG image for conservation or transmission. The method uses fast table search processing, occupies small resources, is applicable to development of embedded systems, realizes necessary image processing functions, finishes the processing of white balance, color offset remediation, picture contrast adjustment, definition increase, noise filtration and the like fast, increases the reusability of each image algorithm in space and time, and improves the execution efficiency.

Description

technical field [0001] The invention belongs to the field of image processing, and relates to an image processing method, in particular to a fast image processing method for a high-definition camera. Background technique [0002] At present, the monitoring methods for vehicles in the field of intelligent transportation mainly include continuous frame image capture mode and single frame image capture mode. Whether it is the former or the latter, most of them have the following problems: (1) The image processing is complex and slow, and can only rely on PCs for calculations, and the joint control of multiple devices has a large amount of calculations. (2) The lack of necessary image processing functions may cause problems such as color cast, poor contrast, poor clarity and noise pollution in the final photos. (3) Each image algorithm is processed relatively independently, and the execution efficiency is low. (4) The update cost of the image algorithm is high, and the code re...

AI Technical Summary

Problems solved by technology

Whether it is the former or the latter, most of them have the following problems: (1) Image processing is complex and slow, and can only rely on PCs for calculations, and the joint control of multiple devices requires a large amount of calculations

(2) The lack of necessary image processing functions may cause problems such as color cast, poor contrast, poor clarity and noise pollution in the final photos

(3) Each image algorithm is processed relatively independently, and the execution efficiency is low

(4) The cost of image algorithm update is high, and the code reuse rate is small

Some image processing methods also use the table look-up method. For example, the patent application number 200710066718.0 discloses an image processing method, but it does not provide a systematic and rich image processing framework, and it is only proposed for a specific algorithm.

Images