摘要
虹膜诊断学是一门通过分析眼睛虹膜纤维组织的色彩与结构,来判断身体组织健康状况的学科,不同人眼虹膜的纤维结构的密度存在差异,可用于判断身体组织健康状况。文中介绍了一种基于Tamura纹理特征中的粗糙度、对比度和方向度的虹膜纤维结构密度计算方法。在使用Tamura算法之前对图片进行归一化、直方图均衡和局部二值化三步预处理,以减小直接应用Tamura算法的背景、光照不均匀的影响,减小结果误差。选取3幅典型图像代表三个等级的虹膜纤维结构密度,进而计算3幅典型图像的Tamura纹理特征以确定虹膜纤维结构密度与所选取特征间的关系。最后将该算法应用到沈阳工业大学视觉检测技术研究所自建虹膜库中的60幅图片样本中,结果表明该方法与主观粗糙度判断一致,切实有效。
Iris diagnostics is a subject to judge the body health by analyzing color and structure of iris fiber tissue. The densities of different fiber structures of human eyes are different, which can be used to judge body health. A calculation method of iris fiber structure based on roughness, contrast and the direction of Tamura texture features is introduced. The image preprocessing is carded on including normalization, histogram equalization and local binarization before utilization of Tamura algorithm, to reduce the influence of uneven illumination by using Tamura algorithm directly and the result error. Three typical images are selected to represent three levels iris fiber structure density ,calculating Tamura texture features of three typical images to determine the relationships between iris fiber structure density and selected characteristics. Finally ,the algorithm is applied to 60 image samples in iris library built by visual inspection techniques institute in Shenyang University of Technology, and the results show that the method is consistent with the subjective judgment of roughness, which is practical and effective.
出处
《计算机技术与发展》
2016年第3期36-39,共4页
Computer Technology and Development
基金
国家自然科学基金资助项目(61271365)
辽宁省博士启动基金项目(20131078)
