摘要
提出了一种改进的Gabor变换的图像纹理增强算法,将空域的纹理图像变换到联合空间频率域并将联合空间频率域的能量分布作为掌纹的特征,Gobar滤波器在频域内对子块其主方向上频率能量分布进行滤波,特征向量进行了匹配识别,增强特征纹线信息并对核心区域进行多方向空间频率能量滤波合成,同时也对滤波算法进行了优化,有效地减少了运算量。实验证明,该套增强算法运行稳定,效果好,可有效提取纹理图像特征纹线信息。
This paper proposes one kind of improved Gabor transformation based image texture enhancement algorithm, which transforms the texture picture of the space domain to the united spatial frequency domain, and takes the energy distribution as the characteristic of the hand print. The range to the subblock its principal direction Gobar fi]ter carries on in the frequency domain the filtering of frequency energy distribution in the subblock＇s principal direction. The eigen vector/feature, vector/proper vector carries on the match recognition, the multi-director space frequency energy filter synthesis, simultaneously has the characteristic print line information is enhanced, and the multi-direction space frequency energy filtering synthesis of the core region is realized. At the same time, the filtering algorithm is optimized, thus the computing complexity is reduced. Experiment indicates that this set of enhanced algorithm operates well.
出处
《通信技术》
2008年第12期259-261,共3页
Communications Technology
