电光与控制, 2013, 20 (12): 51, 网络出版: 2013-12-17
基于图像融合与多尺度分割的目标级变化检测
ObjectLevel Change Detection Based on Image Fusion and MultiScale Segmentation
遥感图像 目标级变化检测 融合 多尺度分割 多值逻辑 remote sensing image objectlevel change detection fusion multiscale segmentation multivalued logic
摘要
传统的像素级变化检测对辐射校正及阈值选择要求较高,因而在实际应用中受到诸多限制。在分析多尺度分割的基础上,提出了一种目标级的变化检测方法。分别利用粗、细尺度对各时相遥感图像的融合图像进行面向对象分割,以获取不同尺寸的目标区域,构造目标的特征进行向量分析得到差异图,并定义变化信息的强度,再利用多值逻辑理论将粗、细尺度下的检测结果进行决策级融合。实验结果表明,与传统的像素级检测方法相比,该方法受辐射差异影响小,检测精度更高,且检测结果对变化强度的衡量准确,能对应于有一定物理意义的目标变化。
Abstract
The traditional pixellevel change detection algorithms have the disadvantage of high requirements for radiometric correction accuracy and threshold selectionwhich may limit their applications.An analysis was made to the multiscale segmentation methodand an objectlevel algorithm was proposed based on it.The fused image of the remote sensing images at different time phases was segmented by using objectoriented method in coarse and fine scales respectively to capture object areas of different sizes.Discrepancy image was obtained after analyzing of characteristics vectorsand the intensity of the changed information was defined.Thenthe decisionlevel fusion was implemented using the measurement results of the coarse and fine scales based on multivalued logic theory.Experiments indicate thatcompared with traditional pixellevel change detection algorithmsthe algorithm is less influenced by radiometric discrepancyhas higher detection accuracyand can evaluate the changed intensity accurately.
吴俊政, 严卫东, 倪维平, 边辉. 基于图像融合与多尺度分割的目标级变化检测[J]. 电光与控制, 2013, 20(12): 51. WU Junzheng, YAN Weidong, NI Weiping, BIAN Hui. ObjectLevel Change Detection Based on Image Fusion and MultiScale Segmentation[J]. Electronics Optics & Control, 2013, 20(12): 51.