数字图像处理技术在并殖吸虫自动分类中的应用研究  

Studies on application of digital image processing technique to automatic taxonomy of Paragonimus

作　　者：付承彬[1] 詹希美[1] 伍小明[2] 李卓雅[1] 王轶[1] 何蔼[1] 

机构地区：[1]中山医科大学病原生物学部寄生虫学教研室,广州510089 [2]中山大学数学系

出　　处：《中国人兽共患病杂志》2002年第1期39-43,共5页Chinese Journal of Zoonoses

基　　金：国家自然科学基金项目 (No .39770 6 72;30 170 837);广东省 2 11重点学科建设基金项目

摘　　要：目的　建立并殖吸虫自动分类系统 ,并研制出并殖吸虫自动分类程序软件。方法　应用数字图像处理技术 (digi talimageprocessingtechnique)建立起并殖吸虫图像采集、处理及识别系统 ,并在此基础上研制出并殖吸虫自动分类程序软件。结果　得到高度清晰的并殖吸虫数字图像 ,分辨率高达 130多万象素 ;利用该技术能获并殖吸虫较为精确的分类参数实际值 ;研制出并殖吸虫自动分类程序软件 ,实际应用证明 :所建立的并殖吸虫自动分类系统能较好地对来自不同国家、地区的 7种并殖吸虫进行自动分类鉴定 ,准确率达 89.0 4%。结论　数字图像处理技术在Aim To establish the automatic system for taxonomy of Paragonimus using digital image processing technique.Methods The system of digital image analysis in this study was made up of a stereomicroscope,a charge coupled device (CCD) camera and a personal computer.The process includes image acquisition and image processing,measurement of trematodes morphology,establishment of classification criteria,exploiting the program software of the automatic system for taxonomy of Paragonimid Trematodes and program software test.Results The high clear digital image database of Paragonimus,the accurate database of classification feature of Paragonimus and the automatic system for taxonomy of Paragonimus were established.Classification based on the eight most important features gave an overall correct classification of 89.04%.The program run accurately,objectively and fast.Conclusion This study provided the theoretical basis and experimental data for further establishing the automatic system for taxonomy of Trematodes.

关 键 词：数字图像处理技术 并殖吸虫 自动分类 

分 类 号：R383.23[医药卫生—医学寄生虫学] R319[医药卫生—基础医学]