Method for improving latent image boundary based on edge detection error diffusion algorithm

Abstract

The invention discloses a method for improving a latent image boundary based on an edge detection error diffusion algorithm, comprising the following steps: step (1), inputting a target image for color separation; step (2), performing amplitude screening; step (3), obtaining a latent image of hidden information; and step (4), adjusting dots at the boundary of and inside the latent image in a halftone picture of an invisible channel, including a primary adjustment and a secondary adjustment; and synthesizing to obtain a halftone picture of the target image embedded with the hidden information.The method adopts the difference dynamics of the dots of the boundary of the latent image to obtain an error dispersion coefficient, quantizes the error of the gray value according to the coefficientand performs dispersion on the surrounding dots, and associates the contour feature of the hidden information with the dispersion of the error, which may effectively improve the problem of anti-counterfeiting failure caused by that the outline of the hidden information reveals.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a latent image boundary improvement method based on an edge detection error diffusion algorithm. Background technique [0002] The grating anti-counterfeiting technology based on the Moiré effect is an important branch and research direction in the field of information hiding technology research. It embeds some anti-counterfeiting information directly into the digital carrier without affecting the use value of the original carrier, and it is not easy to be detected and modified again. But by using the detection grating, the anti-counterfeit information can be recognized and recognized by the user. The purpose of anti-counterfeiting is achieved by hiding the anti-counterfeiting information (hidden information) in the digital carrier (target image). [0003] However, in the anti-counterfeiting information of the prior art, the problem that often occurs in the ...

AI Technical Summary

Problems solved by technology

In the improvement of the dot percentage, combined with the visual masking effect, the adaptation coefficient is set to 0.25. Although this uniform adaptive change of all edge dots can improve the concealment to a certain extent, the scope of adaptation is narrow and can only be used in Better results on images with flatter tones

[0005] Xie Shiqi from Xi'an University of Technology pointed out in "Study on Screening Error Analysis Based on Analog Printing Concept" that the traditional error diffusion method is to process error diffusion from the perspective of transmission error, and the proportional coefficients in each direction are fixed. The scale factor in the vertical direction is always larger than in other directions, which in some cases can cause cumulative errors, resulting in contour distortion

Based on the above, some studies have made further improvements, referring to the error diffusion method to improve the dot percentage on the boundary. Although the adaptive scale coefficients are different, their scale coefficients in the horizontal and vertical directions are always larger than those in other directions. A variety of test results show that such a dot percentage improvement method is still not applicable to the target image in all situations, and will cause serious contour distortion

[0006] The Chinese invention patent application with the publication number CN106891632A discloses a method for making anti-counterfeit marks by combining printed matter gratings and image carriers. By segmenting the graphic information that needs to be hidden, the segmented parts are placed on the carrier image and the visible and hidden grating respectively. , use the explicit and hidden grating with graphic information to extract the complete graphic and text information in the corresponding digital image, and the effect of complete image reproduction, the visible and hidden grating that does not match the digital image cannot achieve the effect of extracting complete graphic and text information, Therefore, the anti-counterfeiting performance of the product is improved, but in the application of the target image with complex texture, the hidden information is very easy to be perceived by the human eye

Images