Name
Title | ||
|---|---|---|
Automatic parameter selection for denoising algorithms using a no-reference measure of image content. |
Abstract | ||
|---|---|---|
Across the field of inverse problems in image and video processing, nearly all algorithms have various parameters which need to be set in order to yield good results. In practice, usually the choice of such parameters is made empirically with trial and error if no "ground-truth" reference is available. Some analytical methods such as cross-validation and Stein's unbiased risk estimate (SURE) have been successfully used to set such parameters. However, these methods tend to be strongly reliant on restrictive assumptions on the noise, and also computationally heavy. In this paper, we propose a no-reference metric Q which is based upon singular value decomposition of local image gradient matrix, and provides a quantitative measure of true image content (i.e., sharpness and contrast as manifested in visually salient geometric features such as edges,) in the presence of noise and other disturbances. This measure 1) is easy to compute, 2) reacts reasonably to both blur and random noise, and 3) works well even when the noise is not Gaussian. The proposed measure is used to automatically and effectively set the parameters of two leading image denoising algorithms. Ample simulated and real data experiments support our claims. Furthermore, tests using the TID2008 database show that this measure correlates well with subjective quality evaluations for both blur and noise distortions. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/TIP.2010.2052820 | IEEE Transactions on Image Processing |
Keywords | Field | DocType |
tid2008 database show,computationally heavy,image content,noise distortion,image processing,quantitative measure,matrix algebra,noise distortions,video processing,image denoising algorithms,image denoising,sharpness,leading image,analytical method,no-reference metric,random noise,proposed measure,automatic parameter selection,visually salient geometric features,gradient methods,tid2008 database,denoising,parameter optimization,stein unbiased risk estimation,denoising algorithm,cross-validation,true image content,local image gradient matrix,blur distortions,singular value decomposition,svd,no-reference measure,noise measurement,parameter estimation,inverse problem,noise reduction,ground truth,estimation,cross validation | Noise reduction,Image gradient,Noise measurement,Image processing,Artificial intelligence,Inverse problem,Estimation theory,Computer vision,Singular value decomposition,Pattern recognition,Algorithm,Cross-validation,Mathematics | Journal |
Volume | Issue | ISSN |
19 | 12 | 1941-0042 |
Citations | PageRank | References |
138 | 5.01 | 12 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xiang Zhu | 1 | 264 | 10.86 |
| Peyman Milanfar | 2 | 3284 | 155.61 |