首页 | 杂志简介 | 征稿简则 | 投稿指南 | 常见问题 | 名词解释 | 刊物订阅 | 联系我们 | English
基于可变阈值和坐标变换融合的红外刑侦重叠手印目标提取算法
投稿时间:2021-06-08  修订日期:2021-06-22  点此下载全文
引用本文:李俊芳,李凯臣,于晓.基于可变阈值和坐标变换融合的红外刑侦重叠手印目标提取算法[J].红外,2021,42(10):33~44
摘要点击次数: 50
全文下载次数: 54
作者单位E-mail
李俊芳 天津理工大学电气工程与自动化学院 lijunfang_tjut@163.com 
李凯臣 天津理工大学电气工程与自动化学院  
于晓 天津理工大学电气电子工程学院 yx_tjut@163.com 
基金项目:天津理工大学教学基金项目(YB20-05);国家自然科学基金项目(61502340);天津市自然科学基金项目(18JCQNJC01000);天津市教委科研计划项目(2018KJ133);天津市复杂系统控制理论与应用重点实验室开放基金项目(TJKL-CATCS-201907)
中文摘要:红外重叠手印图像是刑侦领域比较复杂的一类红外图像,其本身具有信息不确定性和手印重叠性。如何准确提取重叠的手印目标是刑侦领域处理此类图像的一大难题。针对该问题,提出了一种基于可变阈值和坐标变换融合的红外刑侦重叠手印目标提取算法。首先,采用阈值算法提取整体手印图和手印重叠部分图;然后通过坐标变换方法滤除两部分图像的无关区域;最后利用轮廓融合算法将两部分图像融合,得到整体手印目标的提取结果。为了使提取结果具有更好的可视性,又采用图像旋转操作后的坐标变换方法来提取单个手印目标。实验结果表明,与其他一些目标提取方法相比,本文方法在提取整体和单独手印目标时都具有较高的准确性和完整性。
中文关键词:红外重叠手印图像  目标提取  阈值分割  坐标变换  图像融合  图像旋转
 
Target Extraction Algorithm Based on Variable Threshold and Coordinate Transformation for Infrared Overlapped Fingerprint of Criminal Investigation
Abstract:Infrared overlapped fingerprint image is a kind of complex infrared image in the field of criminal investigation. The information of the image is uncertain,and the fingerprints are overlapped. How to accurately extract the overlapped target is a difficult problem in the field of criminal investigation. To solve this problem, an infrared criminal investigation overlapped fingerprint target extraction algorithm based on the fusion of variable threshold value and coordinate transformation is proposed in this paper. Firstly, the threshold algorithm is used to extract the whole and the overlapped fingerprint image. Then the coordinate transformation method is used to filter the independent regions of the two parts of the images. Finally, the contour fusion algorithm is used to fuse the two parts of the image, so as to obtain the extraction result of the whole fingerprint target. In order to make the extraction results more visual, the coordinate transformation method after the image rotation operation to extract a single fingerprint target is adopted in this paper. Experimental results show that, compared with other target extraction methods, this method has higher accuracy and integrity in the whole and single fingerprint target extraction.
keywords:infrared overlapped fingerprint image  target extraction  threshold segmentation  coordinate transformation  image fusion  image rotation
查看全文  HTML  查看/发表评论  下载PDF阅读器

版权所有:《红外》编辑部

北京勤云科技发展有限公司