Two-Level Content-Based Endoscope Image Retrieval
Quan Zhang, Xiaoying Tai, Yihong Dong, Shanliang Pan, Xin Luo, Kenji Kita · 2008
Based on the analysis of endoscope image, in this paper, a new color quantification method is proposed to extract improved CCV and the V component shape invariant moment achieving image feature base. Inspiring from general information searching, the two-level content-based endoscope image retrieval is represented using the improved CCV and V component shape invariant moment guaranteeing the first retrieval recall. Experiments prove the efficiency of these methods.