Abstract | ||
---|---|---|
The conventional histogram intersection (HI) algorithm computes the intersected section of the corresponding color histograms in order to measure the matching rate between two color images. Since this algorithm is strictly based on the matching between bins of identical colors, the final matching rate can be easily affected by color variation caused by various environment changes. In this paper, a Gaussian weighted histogram intersection (GWHI) algorithm is proposed to facilitate the histogram matching via taking into account matching of both identical and similar colors. The weight is determined by the distance between two colors. The algorithm is applied to license plate classification. Experimental results show that the proposed algorithm produces a much lower intra-class distance and a much higher inter-class distance than previous HI algorithms for tested images which are captured under various illumination conditions. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1109/ICPR.2006.596 | ICPR (3) |
Keywords | Field | DocType |
license plate classification,color image,color variation,final matching rate,gaussian weighted histogram intersection,corresponding color histogram,algorithm compute,matching rate,conventional histogram intersection,proposed algorithm,histogram matching,image classification,gaussian processes,color histogram | Histogram,Computer vision,Color histogram,Pattern recognition,Histogram matching,Adaptive histogram equalization,Artificial intelligence,Balanced histogram thresholding,Image histogram,Histogram equalization,Color normalization,Mathematics | Conference |
ISSN | ISBN | Citations |
1051-4651 | 0-7695-2521-0 | 8 |
PageRank | References | Authors |
0.59 | 8 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wenjing Jia | 1 | 325 | 45.08 |
Huaifeng Zhang | 2 | 240 | 18.84 |
Xiangjian He | 3 | 932 | 132.03 |
Qiang Wu | 4 | 304 | 40.42 |