摘要
纹理图像的分割过程一般分为特征抽取和特征划分。本文将LAWS纹理能量和分形维数相结合用于纹理图像特征的抽取,获得纹理图像的特征矢量;在特征划分上,本文利用Konhonen神经网络进行特征的划分,完成纹理的分割,实验表明:这两种特征相结合进行纹理图像的分割能够获得较好的分割效果。
There are two important techniques in image segmentation, feature extraction and feature segmentation. In this paper, we combined LAWS energy with fractal dimension for feature extraction. In this method, a feature vector is assembled; feature space is classified based on Konhonen neural net, final segmentation is accomplished. The experiment shows that this method is effective in texture segmentation.
出处
《科技视界》
2012年第19期169-170,共2页
Science & Technology Vision
