基于Krawtchouk-RBF的印章图像分类识别  

Classification and recognition of seal image based on Krawtchouk-RBF

作　　者：余彪[1] 万水龙 刘进[3] 王强德[1] 

机构地区：[1]中国人民解放军92493部队 [2]中国人民解放军91960部队 [3]南京航空航天大学

出　　处：《微型机与应用》2014年第6期44-47,共4页Microcomputer & Its Applications

基　　金：文件检验鉴定公安部重点实验室开放课题(10KFKT005)

摘　　要：为了有效地对印章图像进行分类,提出了一种基于Krawtchouk矩和RBF神经网络的印章图像分类识别方法。首先提取标准印章图像的Krawtchouk矩不变量,将其作为RBF神经网络的原始训练样本;然后提取全部待鉴印章图像的Krawtchouk矩不变量,将其作为RBF神经网络的输入量进行分类识别。实验结果表明,使用Krawtchouk矩来描述印章图像的特征并通过RBF神经网络来对其进行分类识别的方法十分有效,与同类的Brushlet-RBF法和KPCA-RBF法相比,其识别率更高,且更准确。In order to effectively categorize the seal image, a method of classification and recognition of seal image based on krawtchouk moment and RBF neural network is proposed. Firstly, krawtchouk moment invariants for part of the standard seal images are calculated to train the RBF neural network. Then characteristics of all the questioned seal images are applied to recognize. Results show that Krawtchouk moments as a feature description of the seal image by RBF neural network to classify the identification method is very effective, its recognition rate is higher, and more accurate, compared to the same kind of Brushlet-RBF and KPCA-RBF.

关 键 词：印章图像 分类识别 KRAWTCHOUK矩 RBF 神经网络 

分 类 号：TN911.73[电子电信—通信与信息系统]