Name
Abstract | ||
|---|---|---|
In this paper, a new algorithm of noise reduction for image based on evidence theory is proposed. The values of all pixels are restricted in interval [0, 1], and set of data in each column is a term of mass function, which can be calculated by D–S composition rule. Judging noise can be achieved by comparing with the value of pixel in middle and of the current one. The noise will be removed by substituting the current value with value computed. An improved accelerated algorithm is also presented by sample window of 2 × 2. As a measure of conflict K with greater value shows that there would be noises within the current sample window. At last, Experiment image “Lena” with additive noise shows as a test sample, that better result can be achieved with the algorithm. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1007/s13042-015-0353-6 | Int. J. Machine Learning & Cybernetics |
Keywords | Field | DocType |
Image noise, Evidence theory, D–S rule | Noise reduction,Computer vision,Value noise,Image based,Salt-and-pepper noise,Algorithm,Image noise,Pixel,Artificial intelligence,Gaussian noise,Mathematics,Gradient noise | Journal |
Volume | Issue | ISSN |
8 | 2 | 1868-808X |
Citations | PageRank | References |
1 | 0.35 | 22 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ye Zhao | 1 | 54 | 11.72 |
| Ju-Sheng Mi | 2 | 2054 | 77.81 |
| Xin Liu | 3 | 287 | 74.92 |
| Xiao-Yun Sun | 4 | 1 | 0.35 |