The Research Of Image Contrast Enhanced Reversible Data Hiding

Traditional data hiding technology is applied in multimedia.After hiding data,it may destroy the integrity of the multimedia carrier or even permanently distort the carrier,thus causing the multimedia information loses their original use value.Reversible data hiding technology can not only successfully embed secret data,but also correctly recover the the original carrier,without affecting the original value of the carrier.And thus,the technology of reversible data hiding has been applied in some higher requirement fields,such as military,medical,and judicial,which can be used for image authentication,labeling,or tamper detection.Recently,some scholars have proposed reversible data hiding algorithms that enhance the image contrast effect.The proposed algorithm are not to excessively pursue high peak signal-to-noise ratio,but to more emphasizes the sensitivity of the human eye to the image brightness changes and to better understands of image details,i.e.,the effect of image contrast enhancement.Although the peak signal-to-noise ratio of this type of algorithms are not high enough,the quality of the dense image meets the human visual requirements.However,the image contrast of some existing related algorithms has also been improved,but failed to fully consider the local texture of the image,ledding to insufficient embedding capacity,or excessively-enhanced the contrast or even distortion.Therefore,this paper does some related research and improvement on the reversible data hiding algorithm of contrast enhancement effect:(1)Aiming at the problems of insufficient embedding capacity and serious image distortion of existing algorithms,a novel improved embedding algorithm is proposed.First,texture complexity division is improved to evaluate the local texture of the image,and a more accurate texture complexity classification is obtained.Then,the multiple prediction error expansion method is used to modify the image pixels to embed the hidden data,so that each expandable pixel can carry more information.Finally,a threshold,which avoids severe distortion of the image and improves the visual quality of the image,is introduced to control pixels to be over-modified,(2)Considering the particularity and practical applications of medical images,an improved algorithm is proposed for medical images.The algorithm improves the visual effect of medical images and improves the tampering detection results.First,the algorithm uses the classic threshold segmentation method to distinguish the region of interest from the background.Then,the hidden data is embedded in the region of interests to enhance the visual effect of the image.After that,a series of hash code generated from the enhanced effect image are hidden into the background to detect if the region of interest has been tampered with.When the image is tampered,the tampered area content can be detected.Experimental results show that the proposed algorithm has greater embedding capacity,enhanced image contrast effect and reduced contrast distortion than that of those related algorithms.In addition,the contrast of medical images has also been enhanced,algorithm can detect whether images have been tampered with and more accurately identify the tampered area content.