法医学杂志

法医学杂志 ›› 2021, Vol. 37 ›› Issue (3): 325-331.DOI: 10.12116/j.issn.1004-5619.2020.400506

所属专题： 法医昆虫学

• 专题 • 上一篇    下一篇

基于翅脉的图像数字化分析进行嗜尸性蝇类种类鉴定

尚艳杰1， 潘鹏亮2， 李香蓉1， 李轲1， 林江1， 郭亚东1   

  1. 1. 中南大学基础医学院法医系，湖南 长沙 410013； 2. 信阳农林学院，河南 信阳 464000
  • 发布日期:2021-06-25 出版日期:2021-06-28
  • 通讯作者: 郭亚东，男，教授，主要从事法医昆虫学和法医病理学研究；E-mail：gdy82@126.com
  • 作者简介:尚艳杰（1993—），男，博士，主要从事法医昆虫学和法医病理学研究；E-mail：990799582@qq.com
  • 基金资助:
    国家自然科学基金资助项目（81772026，82072114）

Species Identification of Sarcosaprophagous Flies Based on Vein Digital Image Analysis

SHANG Yan-jie1, PAN Peng-liang2, LI Xiang-rong1, LI Ke1, LIN Jiang1, GUO Ya-dong1   

  1. 1. Department of Forensic Science, School of Basic Medical Sciences, Central South University, Changsha 410013, China; 2. Xinyang Agriculture and Forestry University, Xinyang 464000, Henan Province, China
  • Online:2021-06-25 Published:2021-06-28

PDF (PC)

518

摘要： 目的 基于翅脉的图像数字化分析对常见嗜尸性蝇类进行种类鉴定，为法医昆虫学中嗜尸性蝇类快速、准确的物种鉴定提供新的思路。 方法 用腐肉野外随机诱捕棕尾别麻蝇、绯角亚麻蝇、酱亚麻蝇、拟东方辛麻蝇、红尾粪麻蝇、巨尾阿丽蝇以及裸芒综蝇7种常见的嗜尸性蝇类雌雄成虫共226只，对每只苍蝇右翅的17个标志点进行数字化处理和图像分析，利用置换检验评估异速生长效应的影响，典型变量分析（canonical variate analysis，CVA）对7种嗜尸性蝇类物种间以及雌性物种间的翅形变化情况进行分析，交叉判别验证对分类的可靠性进行评价。 结果 在7种嗜尸性蝇类物种间和雌性成虫物种间，异速生长的影响具有统计学意义（P<0.05）。CVA结果表明在7种嗜尸性蝇类物种间和雌性物种之间，翅膀形状变化具有明显的差异性，前2个典型变量占据了翅脉形状总变异的82.9%和84.1%。利用翅脉的图像数字化分析可以区分这7种常见的嗜尸性蝇类，总体种类判别准确率为81.2%~100.0%，7种雌性蝇类的种类判别准确率为75.0%~100.0%。 结论 基于翅脉的图像数字化分析是一种较为简便、可靠的昆虫物种鉴定的方法，可用于常见嗜尸性蝇类的种类鉴定。

关键词: 法医昆虫学, 嗜尸性蝇类, 翅脉, 图像数字化分析

Abstract: Objective To identify species of common sarcosaprophagous flies based on digital image analysis of veins, in order to provide new idea for fast and accurate species identification of sarcosaprophagous flies in forensic entomology. Methods Random trapping of 226 male and female sarcosaprophagous flies that comprised of 7 common species, including Sarcophaga peregrina, Parasarcophaga ruficornis, Sarcophaga dux, Seniorwhitea reciproca, Bercaea cruentata, Aldrichina grahami, and Synthesiomysia nudiseta with carrion in the field was conducted. The 17 landmarks on the right wing of each fly were digitally processed and the images were analyzed. The effects of allometry were evaluated using a permutation test. Wing shape variations among 7 sarcosaprophagous fly species and female species was analyzed using canonical variate analysis （CVA）. Additionally, cross-validation test was used to evaluate the reliability of classification. Results Among 7 sarcosaprophagous fly species and female species, the effect of allometry had statistical significance （P<0.05）. The CVA results showed that among 7 sarcosaprophagous fly species and female species, differences in the wing shape were significant, and the first two canonical variates accounted for 82.9% and 84.1% of the total variation of vein shape. Vein digital image analysis can be used to separate the 7 common sarcosaprophagous flies, with an overall species identification accuracy of 81.2%-100.0%, and with a species identification accuracy of 75.0%-100.0% to distinguish the female flies of the 7 sarcosaprophagous flies species. Conclusion Vein digital image analysis is a relatively convenient and reliable method for identification of insect species, which can be used for species identification of common sarcosaprophagous flies.

Key words: forensic entomology, sarcosaprophagous flies, vein, digital image processing

中图分类号: