Run-length coding FPGA (field programmable gate array) implementing method in JPEG-LS (joint photographic experts group-lossless standard)

Abstract

The invention discloses a run-length coding FPGA (field programmable gate array) implementing method in a JPEG-LS (joint photographic experts group-lossless standard). The method includes: on the basis of a JPEG-LS implementing method, setting up, storing and looking up a mapping table according to index values so as to obtain an accumulated sum array, comparing run lengths with values in the accumulated sum array for five times according to a binary comparison method so as to obtain cycle indexes to achieve run-length coding, and simply and effectively achieving index value updating in two steps. By the method, the problem of complex updating structures of the index values is solved, run-length coding is achieved in short period, a mapping table structure is simplified by accumulated operations, and hardware resources are saved. Besides, a fully-pipelined structure is adopted, hardware resource consumption is less, and the method is real-time and easy to implement a field programmable gate array.

Description

technical field [0001] The present invention relates to the technical field of image processing, and further relates to a field-programmable gate array FPGA (Field-Programmable Gate Array) implementation method in the JPEG_LS (Joint Photographic Experts Group-lossless) image compression system in the image compression field. The present invention provides an array for binary comparison on the basis of establishing a summation table for traversing all index values, updates index values ​​through table lookup and binary comparison, and realizes run-length coding. The present invention can be applied to image compression coding with the same gray value in a large area in an image compression system. Background technique [0002] The implementation method of JPEG_LS image compression coding is widely used in the field of image compression. Since its run-length coding is completed by a loop structure, and each cycle needs to update data, it has been difficult to implement run-len...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is: in order to complete the encoding mapping operation, four mapping tables are introduced, and each table must store a large amount of data information, which requires a lot of hardware resources such as memory

The shortcoming of this method is that each pixel that meets the conditions is compared, and the output result is compared every time, and the comparison and update operation is performed every cycle, resulting in a large amount of calculation and slow processing speed.

Images