研究了指纹特征提取的原理及算法实现流程，它依次可分为指纹图像归一化、分割和二值化、图像增强滤波、图像细化、特征提取等几个步骤。在分割和二值化阶段提出了一种改进的基于图像分块和阈值迭代的二值化分割算法，新算法能有效提取局部图像的最优二值化阈值，可提高对明暗分布不均的图像的二值化分割的效果和指纹图像分割的质量。在图像增强和细化阶段引入细化邻域方法进行图像增强和细化，最后对指纹特征进行提取。利用 MATLAB 工具进行了仿真，结果表明，采用改进的算法进行指纹特征提取基本能准确、快速的找出指纹图像的特征。%The principle of fingerprint feature extraction and concluding algorithmic flows were studied in this paper. Fingerprint features have been extracted from fingerprint image normalization,segmentation, binarization,image enhancement aspects filtering,image thinning and feature extraction. An improved binary segmentation algorithm was proposed in the study based on image block and threshold iteration, which has solved uneven distribution of light and shade,extracting the optimal threshold value of the local image and improved the quality of image segmentation. The implementation process and steps of the a-bove algorithm was explained in the paper and was verified through MATLAB. The results showed that the algorithm could accurately and quickly identifies the characteristics of the fingerprint image.
展开▼
