虹膜人工特征选择和标注方法的初步研究

目的本文主要研究基于虹膜图像的人工特征选择和标注方法,探寻虹膜人工鉴定研究方向,探讨虹膜技术作为一种新的刑事技术在诉讼中应用的可行性。方法首先,作者结合眼解剖学与虹膜基础理论,对虹膜特征进行了粗分类,分为五大类型:放射状沟线、向心沟、卷缩轮、隐窝和色素点;其次,作者结合已有的虹膜算法和图像处理方法对虹膜图像特征的提取和分析方法进行研究;最后,通过专用软件辅助,对虹膜图像进行区域切割、归一化、特征定位和标记、特征信息提取等系列处理。结果初步实现了两张虹膜图像在同一尺度下的人工特征选取和标注。结论本文研究的虹膜人工特征选择和标注方法,是虹膜识别技术在检验鉴定领域应用的初步探索,为后续虹膜人工比对鉴定的深入研究奠定基础。

Tentative exploration into artificial selection and labeling of iris feature

Objective To explore the artificial selection and labeling of iris feature based on the relevant image dissection, so as to probe into the research direction of iris identification and discuss the feasibility of iris feature as new evidence for litigation. Methods From the anatomic knowledge and theory of eye and iris, the iris features were first of all roughly classified into five general types: radial sulcus line, centripetal sulcus, collarette, crypt and pigment spot. The present available iris algorithms and image processing approaches were then analytically adopted to extract and scrutinize iris image features. Finally, the iris images, having been localized and adjusted under assistance of special software, were subjected to serial tackling of region cutting, image normalization, feature positioning and marking, characteristic information extracting and other processing. Results Two iris images tested were preliminarily realized of artificial selection and labeling of their features at the same scale. Conclusion The here-attempted method is one preliminary exploration on artificial selection and labeling of iris feature, and one tentative application of iris recognition technology in the inspection and identification, having supposedly laid foundation for artificial comparison and identification of iris to go into further depth. © 2021, Editorial Office of Forensic Science and Technology. All rights reserved.