Abstract | ||
---|---|---|
This paper describes a method that can improve the visibility of features in shadowed regions in urban aerial images while retaining non-shadowed regions and the natural tint of shadowed regions automatically. The method first calculates the posteriori probabilities of each pixel belonging to shadowed class or non-shadowed class using color features from a low frequency component image, and then compensates the intensity and saturation of each pixel in shadowed regions according to the posteriori probabilities. Experimental results indicate the effectiveness of proposed method. |
Year | DOI | Venue |
---|---|---|
2000 | 10.1109/ICPR.2000.905339 | ICPR |
Keywords | Field | DocType |
posteriori probability,color feature,natural tint,shadowed region,non-shadowed region,spatial analysis,non-shadowed class,dynamic shadow compensation,low frequency component image,shadowed class,image analysis,probability,image processing,ambient intelligence,image segmentation,geography,remote sensing,low frequency,vents,pixel | Computer vision,Shadow,Visibility,Pattern recognition,Computer science,Image processing,Artificial intelligence,Pixel,Color image | Conference |
ISSN | Citations | PageRank |
1051-4651 | 4 | 0.74 |
References | Authors | |
2 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
akira suzuki | 1 | 4 | 0.74 |
akio shio | 2 | 15 | 3.05 |
Hiroyuki Arai | 3 | 64 | 13.25 |
Sakuichi Ohtsuka | 4 | 21 | 4.80 |