Image Analysis on Corneal Opacity:A Novel Method to Estimate Postmortem Interval in Rabbits

作     者：周兰 刘艳 刘良 卓荦 梁曼 杨帆 任亮 朱少华 

作者机构：Department of Forensic MedicineTongji Medical CollegeHuazhong University of Science and Technology Key Laboratory of Evidence Science(China University of Political Science and Law)Ministry of Education 

出 版 物：《Journal of Huazhong University of Science and Technology(Medical Sciences)》 (华中科技大学学报（医学英德文版）)

年 卷 期：2010年第30卷第2期

页      面：235-239页

核心收录：

学科分类：090603[农学-临床兽医学] 0831[工学-生物医学工程（可授工学、理学、医学学位）] 0710[理学-生物学] 1001[医学-基础医学(可授医学、理学学位)] 08[工学] 080203[工学-机械设计及理论] 09[农学] 0906[农学-兽医学] 0805[工学-材料科学与工程（可授工学、理学学位）] 0802[工学-机械工程] 0703[理学-化学] 0836[工学-生物工程] 

主　　题：forensic medicine postmortem interval corneal opacity image analysis 

摘      要：Corneal opacity is one of the most commonly used parameters for estimating postmortem interval （PMI）. This paper proposes a new method to study the relationship between changes of corneal opacity and PMI by processing and analyzing cornea images. Corneal regions were extracted from images of rabbits＇ eyes and described by color-based and texture-based features, which could represent the changes of cornea at different PMI. A KNN classifier was used to reveal the association of image features and PMI. The result of the classification showed that the new method was reliable and effective.