基于局部高程差异的建筑物提取算法
|
摘要
传统算法直接使用数字表面模型(DSM)进行建筑物提取,忽略了地形影响,只适用于地形变化较小的区域。针对该问题,提出了一种基于局部高程差异的建筑物提取算法。利用地物相对于局部区域的高程差异来反映地物高度,再运用基于知识的提取算法提取建筑物。实验结果表明,该算法具有较高的精度,比SVM算法降低了1.18%的误提率和6.51%的漏提率。
The traditional algorithm using DSM to extract building ignores the influence of terrain, and is only applicable to areas with small changes in terrain. In this paper, aiming at this problem, we proposed a building extraction algorithm based on local elevation difference. We used the elevation difference between the object and the local area to reflect the height of the feature. Then, we used the extraction algorithm based on knowledge to extract the buildings. The experimental results show that the proposed algorithm has higher accuracy than SVM, and the commission rate and the omission rate are reduced by 1.18% and 6.51% compared to SVM.
The traditional algorithm using DSM to extract building ignores the influence of terrain, and is only applicable to areas with small changes in terrain. In this paper, aiming at this problem, we proposed a building extraction algorithm based on local elevation difference. We used the elevation difference between the object and the local area to reflect the height of the feature. Then, we used the extraction algorithm based on knowledge to extract the buildings. The experimental results show that the proposed algorithm has higher accuracy than SVM, and the commission rate and the omission rate are reduced by 1.18% and 6.51% compared to SVM.
引文
[1]李德仁,童庆禧,李荣兴,等.高分辨率对地观测的若干前沿科学问题[J].中国科学:地球科学,2012,42(6):805-813
[2]CHENG G,HAN J W.A Survey on Object Detection in Optical Remote Sensing Images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016(117):11-28
[3]CHEN G,Geoffrey J H,Luis M T C,et al.Object-based Change Detection[J].International Journal of Remote Sensing,2012,33(14):4 434-4 457
[4]Ak?ay H G,Aksoy S.Building Detection Using Directional Spatial Constraints[C]//IEEE Geoscience and Remote Sensing Symposium,2010:1 932-1 935
[5]Baltsavias E P.Object Extraction and Revision by Image Analysis Using Existing Geodata and Knowledge:Current Status and Steps Towards Operational Systems[J].ISPRS Journal of Photogrammetry and Remote Sensing,2004,58(3):129-151
[6]张雁.基于机器学习的遥感图像分类研究[D].北京:北京林业大学,2014
[7]陈光洲,岳东杰,许颖,等.基于D S M深度影像的机载LiDAR建筑物边界提取[J].勘察科学技术,2016(1):44-46
[8]段永超.三维航空影像目标地物对象化分类研究[D].北京:中国科学院大学,2016
[9]熊增连,潘建平,阳春花.面向对象的房屋特征信息自动提取方法[J].地理空间信息,2015,13(5):16-18
[2]CHENG G,HAN J W.A Survey on Object Detection in Optical Remote Sensing Images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016(117):11-28
[3]CHEN G,Geoffrey J H,Luis M T C,et al.Object-based Change Detection[J].International Journal of Remote Sensing,2012,33(14):4 434-4 457
[4]Ak?ay H G,Aksoy S.Building Detection Using Directional Spatial Constraints[C]//IEEE Geoscience and Remote Sensing Symposium,2010:1 932-1 935
[5]Baltsavias E P.Object Extraction and Revision by Image Analysis Using Existing Geodata and Knowledge:Current Status and Steps Towards Operational Systems[J].ISPRS Journal of Photogrammetry and Remote Sensing,2004,58(3):129-151
[6]张雁.基于机器学习的遥感图像分类研究[D].北京:北京林业大学,2014
[7]陈光洲,岳东杰,许颖,等.基于D S M深度影像的机载LiDAR建筑物边界提取[J].勘察科学技术,2016(1):44-46
[8]段永超.三维航空影像目标地物对象化分类研究[D].北京:中国科学院大学,2016
[9]熊增连,潘建平,阳春花.面向对象的房屋特征信息自动提取方法[J].地理空间信息,2015,13(5):16-18