Research On A New Algorithm Based On Image Feature Matching

Halftone screening algorithm is the simulation of continuous tone image by modulating the shape or position of the halftone dot. This paper presents a new algorithm to make the halftone image content features with the original continuous tone images to achieve the best matching, and solving the key problem of the prepress information processing of picture. The anti-counterfeiting super line could be print clearly, and the line pattern get a high quality, the print quality has been greatly improved too. In the case of non use of special anti-counterfeiting ink or anti-counterfeiting paper, and the printing image has a certain anti-false effect itself.Through in-depth researching on the halftone screening algorithms and halftone screening theory, this paper proposes a new algorithm for the printing of anti-countering super lines. In the processing of printing anti-countering lines, usually using AM or FM algorithm, because the anti-countering lines should be real and with black bottom, but considering the characteristics of amplitude modulation or frequency modulation, making the printed anti-countering lines become breezing. So It is difficult to achieve the desired effect.This paper uses a new halftone screening algorithm based on vector mapping, firstly cutting the binary image into several part, for example 8*8 or 16*16 pixels as a part, and then printing every parts. And the image preprocessing stage of binarization and the median filter algorithm to do the optimization, improve the image quality.By using the new halftone screening algorithm, the printed anti-countering lines get more clear and higher quality under the same printer level. Reducing the requirements for high quality prints. And the halftone images which is printed out by this method is difficult to be copied, and has good effect of anti-countering.