Image Artificial Degradation

Image artificial degradation is a kind of process to reduce the visual quality of image voluntarily and reversibly, for achieving the business model which called'try and buy'in image business. The degraded image is free for trial to every client, then it would be decrypted or reconstructed if the client decides to pay. This model makes profit without obeying the'free and share'spirit in Internet, which makes it a'win-win'model.This paper investigates and analyzes the image artificial degradation by selective encryption and image noise both. The contribution of this paper is as follow:1. The image artificial degradation algorithm based on selective encryption was introduced and analyzed,which includes bit plane and DCT coefficient selective encryption[1]. The analysis results indicate that the first and second bit plane should not be encryption in the bit plane selective encryption algorithm in degradation model, and also, in DCT coefficient selective encryption, It is better to encrypt the AC amplitude after the run-length coding rather than before it.2. A new selective encryption algorithm for degradation was proposed, which selects some pixels of the image to encrypt. The experiment showed that this algorithm is vulnerable to image filter attack. Then the block-wise improvement is proposed for this defect. The experiment give a positive result to the improve algorithm.3. A fresh scheme based on image noise for image artificial degradation is proposed, which includes both the additive and multiplicative noise. The predictor formulas for the degree of degradation were provided by quantitative analyzing for both two models. Experiment confirms that those formulas fit the actual value.4. The analysis and improvement of the security of the algorithm based on image noise have been done. To withstand image filter attack, the mixed noise was introduced. For defending image averaging attack, the select templet is employed for dual noise.5. At last, the deficiencies of the paper are proposed for further work.The most important job of the paper is the degradation algorithm based on image noise. The algorithm is practicable because it is simple and fast. The more important feature is that the degree of degradation can be accurately calculated. And in the additive noise model, the compression ratio of degraded image can be predicted. Those are necessary conditions in business application. It means that this algorithm which already own certain application value is qualified for business issue,...