摘要
文章给出了一种基于Gabor小波纹理特征的磨粒图像识别新方法,主要是利用Gabor小波设计了一种多通道小波滤波器,对磨粒图像直接进行小波变换,用Gabor小波变换系数的模的平均值和其标准方差来表示抽取的图像特征。把获得的小波特征归一化后输入到改进的BP神经网络分类器进行分类识别。最后,对磨粒图像进行了一系列的仿真实验,结果表明,识别正确率在 91%以上,并且识别速度很快。
The novel method of wear debris recognition based on Gabor wavelet texture feature is proposed. Mostly multi-channels wavelet filters is designed using Gabor wavelet,and wear debris image is directly transformed by wavelet filters. The feature of extracting gray wear debris image is denoted by the coefficients of Gabor wavelet transform and its standard variance.The wavelet feature is normalized and input into improved BP neural networks to classify. Finally,a series of imitate experimentations are conducted. The results indicate that the identifying accuracy is more than 91%,and the identifying speed is very fast.
出处
《激光与红外》
CAS
CSCD
北大核心
2005年第3期190-192,共3页
Laser & Infrared