论文部分内容阅读
针对遥感影像变化检测中同物异谱、异物同谱导致的波段间敏感性差异、孤立噪声干扰等问题研究的不足,文章提出了一种基于模糊C均值聚类的波段修正法变化检测算法:采用差值/比值复合法构造分波段对比差异影像,增强了复合影像的振幅及结构信息;基于差异影像的邻域熵权修正多波段联合的邻域互信息量建立修正影像,较好整合了影像不同波段间的反射差异和邻域空间信息;最后利用改进的模糊C均值算法对修正影像进行变化检测。多组实验结果表明,波段修正法缓解了单波段敏感性差异对变化检测的影响,有效避免了局部最优、孤立噪声干扰等问题,检测结果更接近客观实际。
Aiming at the shortcomings of the research on the differences between the same material and the different spectrum in the detection of the change of the remote sensing image, the sensitivity difference between the bands caused by the foreign matter in the same spectrum, and the isolated noise interference, a change detection algorithm based on fuzzy C-means clustering is proposed. The differential / ratio composite method was used to construct the sub-band contrast difference image to enhance the composite image’s amplitude and structure information. Based on the difference image’s neighborhood entropy weight, the multi-band joint neighborhood information was modified to build a modified image, The difference of reflection between different wave bands of image and the neighborhood spatial information; finally, the improved fuzzy C-means algorithm is used to detect the change of the corrected image. The experimental results show that the band correction method alleviates the influence of single-band sensitivity difference on the change detection, effectively avoids the problems of local optimum and isolated noise interference, and the detection results are closer to the objective reality.