Specular Reflection Detection on Gastroscopic Images
Gastroscope is important in gastric cancer diagnosis.However, the specular reflection is common existed in thegastroscope images and its easily confused with ulcer. In thispaper we develop a method for detecting specular reflection ingastroscopic images. First the Intensity-Saturation jointdistribution of region of interest (ROI) is obtained in HIS colorspace. Then based on the analysis of the Intensity-Saturation (IS)histogram, several templates of different shapes are tested. After the candidate templates are compared on true positive and falsepositive curves from the comparison tests, triangle template isproved more suitable for recognizing pixels of specular reflectionfrom the others in IS histogram. The experiment results from 64gastroscopic images show that the classifier gets a true positiverate (TPR) at 71.9%, false positive rate (FPR) at 0.75% and areaunder the ROC (AUC) at 0.869. There is no confusion with ulcer.It is robust in size of regions and capable of expanding to otherendoscopes.
Rui Yao Yilun Wu Wei Yang Xiaolin Lin Shidan Chen Su Zhang
Dept.of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China School of Biomedical Engineering, Southern Medical University, Guangzhou, China Dept.of Gastroenterology, Ruijin Hospital, Shanghai, China
国际会议
成都
英文
1-4
2010-06-18（万方平台首次上网日期，不代表论文的发表时间）