基于形态学与支持向量机的虹膜坑洞纹理检测

发布时间：2019-08-17 16:19

【摘要】：坑洞纹理是虹膜表面上一种重要的特征纹理。对可见光虹膜图像而言,如何能不受睫毛、眼睑、光斑及光照不均匀等干扰因素的影响,并能快速、准确地提取出该特征纹理仍然是目前一个亟待解决的难题。提出一种基于形态学和支持向量机(SVM)的可见光虹膜坑洞纹理检测方法。首先使用灰度形态学和二值形态学相结合的方法提取出所有目标纹理;然后使用区域生长方法定位所有目标纹理并计算各个目标纹理的特征向量;最后再使用SVM和定义约束条件的方法提取出最终的坑洞纹理。通过实验证明:该方法能较好地克服光斑等干扰的影响,对坑洞状纹理的检出率高于其他同类方法。
[Abstract]:Pit texture is an important feature texture on iris surface. For visible iris images, how to be unaffected by eyelashes, eyelids, spot and uneven light, and how to extract the feature texture quickly and accurately is still an urgent problem to be solved. A visible iris hole texture detection method based on morphology and support vector machine (SVM) is proposed. Firstly, the gray morphology and binary morphology are used to extract all the target texture, then the region growth method is used to locate all the target texture and calculate the feature vector of each target textures. finally, the final hole texture is extracted by SVM and the method of defining constraints. The experimental results show that this method can overcome the influence of spot and other interference, and the detection rate of pit texture is higher than that of other similar methods.
【作者单位】：沈阳工业大学视觉检测技术研究所;沈阳化工大学计算机科学与技术学院;
【基金】：国家自然科学基金(61271365)项目资助
【分类号】：R770.4

【相似文献】