科研之友
微信
新浪微博
Facebook
分享链接
摘要
Blob features are usually used as texture descriptors for object classification and most traditional blob detectors nowadays are luminance-based However, in many applications we could require a blob detector for the color domain In this paper, we propose a novel and efficient framework for the extension from scalar-signals to vector-signals of multi-scale blob detection to prevent information loss due to gray scale transformation Then, we present a blob detector for stamp image classification to distinguish the stamp types In our experiments, we compare our method to other approaches The experimental results demonstrate the effectiveness of our proposed detector
- 出版日期2010-10
- 单位北京邮电大学
