摘要
利用独立分量分析提取字符特征,以信息理论中负熵作为估计输出分量之间独立性的目标函数,并在此基础上对待识别字符进行重建,通过对重建模型的误差分析进行字符识别.对3000个车牌字符的识别实验,取得了较高的识别率,证明其算法的有效性和鲁棒性.
Independent component analysis is used to extract features of characters, and negentropy is applied to estimate statistical independence between output components, then character recognition is conducted based on the error analysis of reconstructed models＇. The proposed algorithm is tested on 3000 characters, a high recognition rate can be obtained, and the experimental results demonstrate the algorithm is feasible, robust and applicable.
出处
《四川大学学报（自然科学版）》
CAS
CSCD
北大核心
2006年第6期1259-1263,共5页
Journal of Sichuan University(Natural Science Edition)
关键词
独立分量分析
特征提取
重建模型
FASTICA算法
independent component analysis
feature extraction
model reconstruction
Fast ICA algorithm
