Research On Reversible Digital Watermarking Based On Adaptive PVO

As an important branch of information hiding field,digital watermarking technology is a key technical means in digital works copyright authentication and image integrity verification.The reversible digital watermarking technology relies on its ability to hide secret information in the carrier image and can recover the secret information,which contributes to the carrier image without loss in the use stage.It is widely used in military remote sensing images,medical diagnostic images and other application scenes with high original image quality and signal accuracy.Since its concept has been put forward for more than 20 years,the algorithm idea of reversible digital watermarking technology has been constantly innovated through the in-depth research of many scholars.Also,the program performance has been gradually optimized,which further meets the needs of production and life in human society.Currently common reversible watermarking technology solutions can be summarized into three categories: based on the difference expansion,histogram shifting and pixel values ordering,approved by the algorithm design optimization and simulation.Besides,practice confirmed that the watermarking scheme based on pixel value ordering(PVO)has little influence on image distortion at low and middle embedding capacity compared with the former two kinds of schemes.Based on the deep study and research of the existing algorithms,the author finds that the traditional reversible digital watermarking scheme based on PVO still has room for improvement in pixel partitioning strategy and texture complexity calculation.On the one hand,the traditional scheme divides the image into fixed pixel blocks such as 2×2 or 4×4,which is difficult to meet the embedding requirements of different texture complexity in the carrier image at the same time.On the other hand,in the traditional PVO scheme,only two-pixel values within the block are selected to calculate the texture complexity,moreover,the result is prone to deviation from the actual texture complexity of the block.For the above two cases,a reversible digital watermarking scheme based on dynamic adaptive block PVO is proposed.On the one hand,according to the complexity of the local pixel texture of the image,the dynamic adaptive blocking strategy of pixel-by-pixel expansion is adopted,the pixel size could be flexibly controlled according to the complexity of the image.Additionally,this scheme realized with image texture feature is more consistent with the chunking strategy compared with the traditional scheme,which could further improve the efficiency of image pixel resources.On the other hand,this paper proposed that by calculating the sample variance of all the pixels except the pixels in which the watermark information is embedded in the block.The texture complexity calculation method of the complexity of the block texture is represented by the degree of dispersion of the pixel values in the block.Under the premise of ensuring the reversibility of the algorithm,this scheme reflects the complexity of the actual texture more accurately in comparison with the traditional scheme.Meanwhile,it further reduces the distortion of the original carrier image which can improve the performance of the corresponding algorithm in the embedding stage.In conclusion,through the two aspects of design optimization above,simulation results are analyzed to verify the improvement of the performance of the algorithm in comparison with the current mainstream.The proposed algorithm has important reference meaning in dynamic block strategy,texture complexity calculation,threshold combination selection and improving the performance of reversible digital watermarking technology under small and medium embedding capacity.