Research On Digital Image Compression Coding And Digital Image Watermarking Algorithms

On the basis of analysis and summary of relevant research, in-depth study is carried out on key technology in digital image compression coding and digital image watermark. Several algorithms are proposed by optimizing algorithm structure and improving encoding and watermarking procedure, etc. The main innovative achievements are listed as follows:In view of the features of the digital image transform coefficients, the image coding technology in transform domain is studied in this paper. A new image coding method is proposed based on wavelet transform and color correlation. The correlation of color components is used in the method, which is especially suitable for progressive transmission system at low bit rates. A new image coding method based on wavelet transform and distributed arithmetic coding is proposed. The method can make the decoded image has good visual effect with high compression ratio and high PSNR, which is especially suitable for the medical image lossless compression. And then, in order to capture the essential edge structure information in the image data, a new image compression coding method based on total variation model and bandelet transform is proposed, which greatly reduces the blocking effects of the DCT image compression and is especially suitable for texture image compression.In order to solve the problem of ROI image compression, two new ROI coding algorithms for medical image by progressive transmission are introduced in this paper. A new image coding algorithm is proposed based on active contour model image segmentation. The method can obtain ROI through modified active contour model and realize the image data compression combined with wavelet transform. The decoders can achieve near-lossless compression image with higher compression ratio and PSNR. Another image coding algorithm is proposed based on fuzzy C-means clustering image segmentation and the method is simple and easy to implement. The lossless compression is achieved by using the modified rectangle split-merge coding to compress image data. The experiment result shows that the two methods can be used to compress medical image data with high compression ratio and fidelity.The image watermarking technology in transform domain is studied in this paper. A new digital image watermarking method based on multiple description coding and wavelet transform is introduced. Multiple descriptions coding in the channel coding technology is used to embed the watermark into the wavelet domain by the interaction block frequency hopping spectrum spread strategy. The experiment result shows that the proposed method has better performance of anti-attack of noise and cutting. A digital image watermark algorithm based on bandelet transform is presented to enhance the safety of the watermarking system and the imperceptibility of the watermark. An Arnold transform, chaotic encryption modulation and BCH coding is carried out on the watermark which is embedded into the bandelet coefficients. The experiment result shows Bandlet domain inherits the advantages of wavelet domain, which has strong ability against cropping and compression assault, and improves the anti-noise ability of the watermarking system. This algorithm proved to be with better robustness.According to the medical image application background, a new medical image blind watermarking algorithm based on ROI and contourlet transform is proposed. The image is segmented through modified active contour model and the double digital watermarking is embedded by using the modified difference technique and the method which comparing the contourlet coefficients with its neighborhood mean respectively. Watermark higher PSNR at the decoders can be extracted which is able to protect the copyright and hide patients’ private information. The method can determine and repair the tampered region effectively, which has better robustness for common image processing operations.