An image preprocessing device suitable for video coding

Abstract

The invention relates to the technical field of image processing, and provides an image preprocessing device suitable for video coding. The device realizes the functions of macroblock zooming, image layer superimposing and thumbnail output of the source image in an online manner, and comprises a data source management module, a two-dimensional zooming module, an image layer superimposing module, an output module and the like. The data source management module pre-generates a source data read instruction required for the output of the current target macroblock row and buffers the source data; The image layer superimposing module reads out and pre-reads the macroblock data of the corresponding superimposed layer according to the coordinate position of each image layer. The output module obtains image data of corresponding size according to a fixed reduction multiple, and then outputs the image data to the video encoding module and writes the image data into the off-chip memory. The device cooperates with the pipeline processing of each module through the block partition and ping-pong storage of the data source, which not only reduces the bandwidth consumption, but also satisfies thereal-time requirement of the high-definition video coding.

Description

technical field [0001] The present invention relates to the field of video coding, in particular to a device for preprocessing a video coding source. Background technique [0002] As the user's demand for high-definition video is getting higher and higher, the bandwidth resource consumed by the image processing process and real-time performance become the bottleneck of the system. [0003] In the existing technology, the two-dimensional image scaling filter can scale the original image in both horizontal and vertical directions. The traditional scaling method, one is to write the intermediate result horizontally first to the off-chip memory, and then from the off-chip memory Read the horizontal scaling result and perform the vertical scaling process. This method is relatively simple to implement, but it requires multiple read and write off-chip memory operations, which will consume more system bandwidth and reduce system processing performance. One is to realize the two-di...

AI Technical Summary

Problems solved by technology

This method is relatively simple to implement, but it requires multiple read and write off-chip memory operations, which will consume more system bandwidth and reduce system processing performance.

One is to realize the two-dimensional zoom operation in one step. This method requires a lot of line buffers to store the original image pixel information required by the filter, which will use a lot of on-chip memory, especially in the field of video coding. Image processing is based on macro Block format output, in order to meet the fast output of data in a macroblock, if a traditional method, such as a 2nd-order vertical filter, needs to buffer 32 lines in the worst case, it will consume a lot of hardware resources, and this press The method of zooming the entire row cannot meet the real-time requirements of video encoding

Images