Image denoising based on the dyadic wavelet transform and improved threshold. (English) Zbl 1172.94317
Summary: Wavelet transform has shown a strong capability to filter out noise that existed in images. The main problem to improve denoising performance lies in the fact that it is difficult to select suitable wavelet transform and related thresholds. In this paper, we completely explore characteristics of dyadic wavelet transform and its possible use to image denoising. An improved threshold method is elaborately designed to establish the self-adjusted layer threshold function and wavelet reconstruction. Theoretical analysis shows that denoising precision can be improved by way of adopting different thresholds according to the different scales of the wavelet coefficients of image and noise. Experiment results demonstrated that our method is obviously superior to that of fixed threshold regarding the effect of image denoising.
MSC:
| 94A08 | Image processing (compression, reconstruction, etc.) in information and communication theory |
References:
| [1] | DOI: 10.1142/S0219691305000701 · Zbl 1104.68811 · doi:10.1142/S0219691305000701 |
| [2] | DOI: 10.1142/S0219691306001452 · Zbl 1141.68592 · doi:10.1142/S0219691306001452 |
| [3] | DOI: 10.1142/S0219691307002117 · Zbl 1197.42029 · doi:10.1142/S0219691307002117 |
| [4] | DOI: 10.1142/S0219691308002227 · Zbl 1148.68462 · doi:10.1142/S0219691308002227 |
| [5] | DOI: 10.1142/S0219691308002513 · Zbl 1157.94318 · doi:10.1142/S0219691308002513 |
| [6] | DOI: 10.1002/cpa.3160410705 · Zbl 0644.42026 · doi:10.1002/cpa.3160410705 |
| [7] | Mallat S., IEEE Trans. Pattern Anal. Mach. Intell. 11 pp 374– |
| [8] | DOI: 10.1109/29.45554 · doi:10.1109/29.45554 |
| [9] | Zhang Z., J. Electron. Inf. Technol. 11 pp 1080– |
| [10] | DOI: 10.1109/34.142909 · doi:10.1109/34.142909 |
| [11] | DOI: 10.1109/18.382009 · Zbl 0820.62002 · doi:10.1109/18.382009 |
| [12] | Zhang W., J. Xidian Univ. 31 pp 57– |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
