In this paper, we incorporate Fourier transform with Log-Polar transform to achieve rotation invariant texture analysis.
通过将对数极坐标和傅立叶变换结合,完成旋转不变的傅立叶纹理特征提取。
An uniform sampling image can be transformed to a space variant sampling image by log-polar coordinate mapping which reduces the searching space and redundant pixels.
通过采用对数极坐标映射,可将空间均匀取样得到的图像变换成非均匀取样的图像,从而减小冗余像素,压缩了搜索空间。
This paper proposes a kernel function - hierarchical log-polar matching kernel which making use of the feature spatial information for building classification in remote sensing images.
提出了一种利用图像特征空间信息的核函数——层次对数极坐标匹配核,用于遥感图像建筑物目标的分类。
Results show highter compnaion ratio were obtained using polar-log cooulinatemappmg as preprocessor.
结果表明以极对数映射作预处理器压缩比高。
Results show highter compnaion ratio were obtained using polar-log cooulinatemappmg as preprocessor.
结果表明以极对数映射作预处理器压缩比高。
应用推荐