基于多源遥感影像的高后果区变化检测方法

近年来,随着城镇化进程的持续推进,部分管道的周边环境不断变化,相应管道的高后果区范围也随之发生变化,准确、高效地掌握环境变化数据以持续更新高后果区信息对管道企业的安全管理工作具有重要意义。由于短期内管道沿线具有自然地物变化较少、人工地物变化频率较高的特点,利用高分辨率正射影像与多光谱影像,提出一种基于多源遥感影像的高后果区变化检测方法。该方法在超像素分割基础上,以LBP-HOG融合特征和光谱梯度差方法对多源数据的变化信息进行提取。实验表明,此方法对管道周边人工地物的变化检测具有较高的准确性,可有效提高高后果区管理的工作效率。

Detection method for change of high-consequence areas based on multi-source remote sensing image

With the continuous promotion of urbanization in China in recent years,the surrounding environment of some pipelines is changing constantly,and the scope of the high-consequence areas of the pipeline section is also changing.Accurately and efficiently mastering the data of environment change to update the information of the high-consequence areas is very important for the safety management of the pipeline companies.According to the characteristics of less change of natural features but frequent change of artificial features along the pipeline in short term,a method for detection of the change of the high-consequence areas based on the multi-source remote sensing image with high-resolution orthoimage and multispectral image was put forward.In this method,the change information of multi-source data is extracted with the LBP-HOG fusion feature and spectral gradient difference method based on the super-pixel segmentation.Experiments show that,the change of artificial features along the pipeline can be detected accurately with this method,so that the efficiency for management of high-consequence areas could be improved effectively.