基于谱直方图和SOFM网络的纹理分类

An Algorithm for Classification of Texture Image Based on Spectrum Histogram and Self-organized Feature Mapping Network

提出了一种基于纹理谱直方图和自组织特征映射网络的纹理分类方法。引入像素的八近邻离散付氏变换犤5犦,随机选取局部纹理区域,计算所选纹理区域的纹理谱并量化得到谱直方图,将其作为自组织特征映射(SOFM)网络的特征模式输入并训练网络。训练结束后的拓扑输出层对应于纹理的不同类别。算法简单有效,对6类Brodatz纹理进行测试,得到了良好的分类效果。

This paper suggests an algorithm for classification of texture image based on spectrum histogram and self-or-ganized feature mapping network.It induces the idea of pixel's8-neighbor Fourier series,randomly chooses the local re-gion of texture and computes the spectrum of it.After quantizing the spectrum,the spectrum histogram of the local area is extracted and then provided to the SOFM network as a feature vector to train the net.The neurons in the topological output layer correspond to different textures when the training process is finished.Experiments on6samples of Brodatz textures demonstrate the simplicity and efficiency of this algorithm.