摘要
基于高等教育自学考试背景,对笔迹鉴别算法进行研究。首先利用多通道的Gabor变换对笔迹纹理图进行特征提取,并通过欧式距离分类器进行分类,在训练样本库中找到与待测样本最相似的5幅笔迹。在此基础上,利用9/7提升小波算法提取出图像高频部分细节特征,对笔迹中的特征单字进行特征提取,综合分析,给出最终鉴别结果。实验结果证明,本算法具有良好的鉴别率。
This paper researches the algorithm of handwriting identification based on the background of self-study examination of higher education. Firstly, multi-chaanel Gabor filter is used in the texture analysis to gain handwriting features. It uses the Euclidean distance as a classifier to classify, then, finds the most similar 5 handwritings with test sample in the training sample library. On tiffs basis, using of the characteristics of 9/7 lifting wavelet algorithm which can extract the minutiae of the high frequent of image extract the features for the single character. Through comprehensive analysis, it gives the final results of the identification. Experimental results show that this algorithm has good identification rates.
出处
《计算机与现代化》
2010年第3期133-137,140,共6页
Computer and Modernization
关键词
单字
GABOR变换
提升小波
笔迹鉴别
特征提取
single character
Gabor transform
lifting wavelet
handwriting identification
feature extraction
