SIFT算法在钳剪痕迹中的同一认定研究  

Research on identification of SIFT algorithm in clipper marks

作　　者：蒋小龙 王震[1] 李德盛 JIANG Xiaolong;WANG Zhen;LI Desheng(Criminal Investigation Police University of China,Shenyang 110035,China)

机构地区：[1]中国刑事警察学院,辽宁沈阳110035

出　　处：《现代电子技术》2024年第19期62-66,共5页Modern Electronics Technique

基　　金：公安部科技基础工作专项项目(2023JC07)。

摘　　要：随着数字图像处理技术的进步,钳剪痕迹的同一认定成为了一个关键的研究领域。文中提出一种基于尺度不变特征变换(SIFT)的方法对钳剪痕迹进行同一认定。首先,利用SIFT算法从钢丝钳剪切的断头痕迹图像中提取关键点,将已知剪切痕迹检材与待识别痕迹样本进行匹配;然后,根据匹配点的数量和距离信息,评估痕迹图像之间的相似性和一致性;最后,通过RANSAC算法进行一致性验证和错误匹配剔除,得到最终的匹配结果。实验结果表明,该方法能够在较多的实验样本中进行初步的同一认定,为钳剪痕迹的识别提供了一种新的有效途径。With the development of digital image processing technology,identification of clipper marks has become a key research area.In this paper,a scale-invariant feature transform(SIFT)based method is proposed for the identity identification of the clipper marks.The SIFT algorithm is used to extract the key points from the image with severed head marks generated by cutting of wire cutters,and then the known material with cut marks is matched with the sample of the marks to be identified.The similarity and consistency between the images of the marks are evaluated on the basis of the number of the points to be matched and the distance information.The RANSAC(random sample consensus)algorithm is used for consistency verification and error matching elimination,and the final matched result is obtained.The experimental results show that the proposed method can be used for the preliminary identification of more experimental samples,and is a new and effective way for the recognition of clipper marks.

关 键 词：尺度不变特征变换 RANSAC算法 钳剪痕迹 特征匹配 同一认定 一致性验证 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP399[电子电信—信息与通信工程]