Method for reducing internal storage consumption inedge enhancement for multi-output pixed device

Abstract

The invention relates to a method for reducing internal memory consumption as making edge enhancement on multi-output pixel device, utilizing the four existing static random access memories and n+4 FIFO buffers, when first to nth input cells read next bit data, removing first column of bit data of the first to nth buffers, and shifting leftward various colums of bit data by one bit each, but for the first colum of bit data of the (n+1)th to (n+4)th buffers, storing all of them into the positions at one end of each of the first to fourth memories, respectively, and extracting a colum of bit data from the other end, respectively, in sequence storing them into the end positions of the first to fourth buffers, and storing the read next bit data into the fifth to (n+4)th buffers, thus able to simultaneously extract n 9*5 tables from the FIFO buffers for the edge enhancing algorithm. And it can largely reduce hardwre resource consumption, avoding influencing processing speed by repeatedly reading data.

Description

technical field [0001] The present invention relates to a method for reducing hardware resource consumption when performing edge enhancement, in particular to a method for reducing memory consumption when performing edge enhancement on a multi-output pixel device. Background technique [0002] In the current field of image processing, most research and development focuses on improving whether the visual effect produced by an image output device (such as a printer) after image output is consistent with the real object. Edges of original image objects, such as fonts, curves, three-dimensional objects, etc., can be approximated by linear equations or natural logarithms to make the edges smoother. When these objects are digitized, they need to be converted into a matrix format that can be processed by an image output device, and their edges must conform to the grid of the device. [0003] However, when the resolution of these image output devices is low, the edges of the image ...

AI Technical Summary

Problems solved by technology

[0008] However, when performing edge-enhancing calculations, since the sequence data must be read sequentially, the first input unit 141 and the second input unit 142 cannot be accessed randomly, and the data to be processed must be read separately. FIFO buffer 120 of large size, and static random access memory 170, and is equivalent to repeatedly reading data twice from the outside, which will cause a large amount of consumption of memory storage capacity used for calculation in the system, especially with For the use of bandwidth of the internal signal transmission bus (bus) and the CPU (Central Processing Unit) resources consumed during data transmission, whether it is edge enhancement of odd-numbered row sequence data or even-numbered row sequence data, To construct an individual 9×5 form 110, the data must be read many times, which seriously affects the execution of other processing programs. Although it can achieve the advantages of faster printing speed and better output effect, it relatively pays more hardware costs. , especially memory and CPU resource consumption

Images