|
| 基于Harris角点检测和聚类算法的掌纹图像ROI提取方法 |
| ROI extraction method of palmprint based on the Harris point detection and clustering algorithm |
| 投稿时间：2017-12-31 |
| DOI：10.3969/j.issn.1005-5630.2018.05.005 |
| 中文关键词: 掌纹识别 Harris算法 聚类算法 感兴趣的区域 |
| 英文关键词:palmprint recognition Harris algorithm clustering algorithm region of interest |
| 基金项目: |
|
| 摘要点击次数: 3424 |
| 全文下载次数: 3655 |
| 中文摘要: |
| 为了实现从掌纹图像中提取出稳定的感兴趣的区域（ROI）图像的目标，设计了一种新的掌纹ROI提取方法。使用Harris角点检测算法对二值化后的掌纹图像进行角点提取，收集相应区域的角点，利用聚类算法得到角点簇的中心点坐标，从中寻找关键点建立坐标系，提取ROI。分别利用该方法在不同掌纹数据库中进行了ROI提取实验。实验结果表明，该方法对不同的掌纹数据库图像均能保持很好的提取效果，成功率均达到了99%以上。 |
| 英文摘要: |
| In order to achieve the goal of extracting a stable region of interest(ROI) image from palmprint images,a new method for extracting palmprint ROI was proposed.Harris point detection algorithm is used to extract the angular point of the palmprint image after binarization and collect the angular points in the corresponding region.Clustering algorithm is used to obtain the central coordinates of the angular cluster.The key points are found in the cluster center to establish coordinate system to extract the ROI.The ROI extraction experiment was carried out in the different palmprint database.The results show that the algorithm can effectively extract the ROI in the different palmprint database,with the accuracy rate of more than 99%. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
