Research On Anti-counterfeiting Of Halftone Information Hiding

In recent years,with the rapid development of the commodity economy and the significant improvement in the demand for property rights maintenance and genuine security,the halftone information anti-counterfeiting scheme has been rapidly developed and widely used for its low cost,simple operation and no space occupation.In this paper,halftone image is taken as the main research object.By studying the halftone information hiding algorithm based on dot shape and the anti-counterfeiting scheme based on hard copy output watermark,the hidden performance and extraction effect of anti-counterfeiting information in halftone image are analyzed,and different anti-counterfeiting schemes are proposed.The research content mainly includes the following aspects:Aiming at the halftone information hiding scheme based on dot shape,this paper explores the best matching rule of different shape dot combinations under the premise of ensuring that the hidden graphic information is invisible to the human eye.Round,square,and diamond-shaped dots commonly used in print production are combined to obtain different matching schemes.The hidden information is binarized and used to control the modulated signal of the screening.In the process of image halftone processing,the hidden information part adopts a mesh type screening,and the original graphic part adopts another kind of network dot screening to achieve the information hiding effect.When extracting,use the dot template to filter out the location of the hidden information dot and reconstruct the hidden information.The subjective and objective evaluation results show that the image generated by the circular-diamond dot combination screening is the most similar to the original image,and the concealment and extraction are superior.Studying the dual anti-counterfeiting effect of the halftone information hiding method combined with the micro-art art network,the threshold matrix of H-shape and circle dots is designed.Then,hidden information is scrambled and generated control signals to screening by encrypted information.In the halftone processing of carrier image,disparate regions of image is screened by control signals using the different dot shapes of threshold matrix,and halftone image containing both micro-art dots and anti-counterfeiting information is obtained.Template matching method is used to extract the anti-counterfeiting information.Experiment utilized the circular dot template to obtain the scrambled hiding information,and then get the original hiding information after reverse scrambling.In the specific area of screening images,the H-type artistic dots are clearly,so the printed products are authenticated effectively.The halftone image is similar to the carrier image,and the security of the anti-counterfeiting information is strong.FSIM values of extracted hidden information are all over 0.9.Based on the digital watermark anti-counterfeiting scheme,the paper explores the influence of paper surface performance on watermark anti-counterfeiting information extraction.Theexperiment tests the roughness,whiteness,opacity,basis weight,glossiness of 7 commonly used printing papers,and prints and scans watermarks.The normalized correlation coefficient(NC)is selected as the characterization parameter of watermark extraction quality.Combined with the correlation statistical analysis theory,the correlation between the extraction quality of watermark and the paper parameters under hard copy output is evaluated.The results show that the quality of watermark extraction of different papers is different.When the scanning resolution is 600 dpi,the correlation between paper whiteness,roughness,smoothness and NC value is greater than 0.8,the gloss correlation value is slightly greater than 0.6,and the opacity value is less than 0.4.So the previous three influencing factors have a strong correlation with the quality of watermark extraction,the influence of gloss is normal,and the influence of opacity is small.In order to balance the conflict between watermark robustness and invisibility,and optimize the performance of watermark anti-attack,a double-transform domain watermarking algorithm based on image texture features is proposed.The texture features of the carrier image are extracted by the wavelet coefficient analysis method,and the original watermark image is combined to generate the reconstructed watermark.Wavelet transform is performed on the three color components of the RGB carrier image,and the low-frequency components of each component are cut to form a pure quaternion matrix.Then the reconstructed watermark replaced the real-region intermediate frequency coefficients of quaternion.By inverse quaternion transformation and wavelet reconstruction,experiment implemented the embedding of the watermark.The results of experiment showed that the proposed algorithm can effectively resist conventional image processing attacks,and algorithm has strong robustness.