摘要
针对传统方法存在适应性差、计算量大等特点,提出一种指纹图像置信极值化的处理方法。采用图像均衡化与置信阈值极值化结合对图像进行强化,对每个小区域的像素点进行灰度值求均值和方差,并采用VisualC++平台对算法进行实验验证。实验结果表明,该算法具有处理效果好、适应性强的特点。
In view of the characteristics of the traditional methods, such as poor adaptability and large amount of computation, a method to deal with the confidence extremum of fingerprint images is proposed. Image equalization and confidence threshold extremum are combined to enhance the image. The gray mean value and variance of each pixel in each small area are calculated. The algorithm is verified by Visual C++ platform. The experimental results show that the algorithm has the advantages of good processing effect and strong adaptability.
引文
[1]DAS R K,DE A,PAL C,et al.DSP hardware design for fingerprint binarization and thinning on FPGA[C]//International Conference on Control,Instrumentation,Energy and Communication.IEEE,2014:544-549.
[2]祁亚萍.指纹图像二值化算法的研究[J].科技信息,2011(23):25-30.
[3]MEDEIROS L X,FLORES E L,CARRIJO G A.Optimization of Calculation of Field Orientation Time and Binarization of Fingerprint Images[J].IEEE Latin America Transactions,2011,9(5):868-874.
[4]杨得国,吕怿萌.一种改进的指纹识别算法与研究[J].华中师范大学学报(自然科学版),2015,49(5):668-671.
[5]刘永霞.指纹图像预处理及特征提取算法的研究与实现[D].成都:电子科技大学,2011:32-33.
[6]李智,曲长文,周强,等.一种基于改进SIFT的SAR图像配准算法[J].兵工自动化,2018,37(2):23-28.
[7]田俊青,徐荣青,李亚玲,等.一种改进的指纹图像二值化处理方法[J].微型机与应用,2010,29(18):52-55.
[8]陈戈珩,李文秀.指纹图像二值化优化算法[J].长春工业大学学报,2016,37(5):499-502.
[9]计春雷,冯伟,黎明,等.一种动态阈值加填补的指纹图像二值化算法[J].计算机仿真,2011,28(7):258-261.