1. A Color HSV Image Edge Detection Method Based on Gradient Extreme Value
- Author
-
Jing-xiu Zhao and Ming-yan Liu
- Subjects
Color histogram ,Color image ,business.industry ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Color balance ,Color gradient ,Color quantization ,Edge detection ,Computer Science::Computer Vision and Pattern Recognition ,Computer vision ,Artificial intelligence ,business ,Image gradient ,ComputingMethodologies_COMPUTERGRAPHICS ,Hue ,Mathematics - Abstract
In order to be better for color image edge detection,this paper analyze the existing problems of gradient algorithm with HSV color image characteristics and the separation by color hue, saturation and brightness information and proposes a color image edge detection method based on gradient extreme value of the local region. The algorithm extracts the edge pixels directly in the local area boundary detection window, first calculate the distance of hue, saturation and brightness between two pixels and then take their sum as the final color distance. The experimental results indicate that this method could obtain more real edges with better continuity.
- Published
- 2008
- Full Text
- View/download PDF