论文部分内容阅读
针对城区分类,利用颜色特征构造一个新颖的无监督的分类框架.首先,基于最近提出的极化合成孔径雷达(PolSAR)数据的四分量分解模型,计算了常用的颜色空间:YUV,RGB,HSI和CIELab,通过引入颜色熵量化的选择颜色特征;然后,联合纹理特征和扩展的散射功率熵,用自适应的均值漂移算法分割PolSAR图像;最后,根据基于G0分布的距离测度合并聚簇为较为匀质的地物类别.通过L波段AIRSAR数据和C波段Radarsat-2的PolSAR数据验证了提出算法的有效性,分类正确率表明,相比于已有的工作,提出的算法对于城区有较好的区分能力.
A novel unsupervised classification framework is constructed for urban classification using color features.Firstly, based on the recently proposed four-component decomposition model of Polarimetric Synthetic Aperture Radar (PolSAR) data, commonly used color space is calculated: YUV, RGB, HSI And CIELab, the color feature is quantified by the introduction of color entropy. Secondly, the PolSAR image is segmented by the adaptive mean shift algorithm by combining the texture features with the extended entropy of scattering power. Finally, the clustering is better based on the distance measure based on G0 distribution Homogeneous object classification.The validity of the proposed algorithm is verified by the L-band AIRSAR data and the PolSAR data of the C-band Radarsat-2, and the classification accuracy shows that the proposed algorithm is better than the existing one The ability to distinguish.