School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China.
Collaborative Innovation Center of Geospatial Technology, Wuhan University, Wuhan 430079, China.
Sensors (Basel). 2019 Mar 31;19(7):1557. doi: 10.3390/s19071557.
The identification and monitoring of buildings from remotely sensed imagery are of considerable value for urbanization monitoring. Two outstanding issues in the detection of changes in buildings with composite structures and relief displacements are heterogeneous appearances and positional inconsistencies. In this paper, a novel patch-based matching approach is developed using densely connected conditional random field (CRF) optimization to detect building changes from bi-temporal aerial images. First, the bi-temporal aerial images are combined to obtain change information using an object-oriented technique, and then semantic segmentation based on a deep convolutional neural network is used to extract building areas. With the change information and extracted buildings, a graph-cuts-based segmentation algorithm is applied to generate the bi-temporal changed building proposals. Next, in the bi-temporal changed building proposals, corner and edge information are integrated for feature detection through a phase congruency (PC) model, and the structural feature descriptor, called the histogram of orientated PC, is used to perform patch-based roof matching. We determined the final change in buildings by gathering matched roof and bi-temporal changed building proposals using co-refinement based on CRF, which were further classified as "newly built," "demolished", or "changed". Experiments were conducted with two typical datasets covering complex urban scenes with diverse building types. The results confirm the effectiveness and generality of the proposed algorithm, with more than 85% and 90% in overall accuracy and completeness, respectively.
从遥感图像中识别和监测建筑物对于城市化监测具有重要价值。在检测具有复合结构和地形位移的建筑物变化时，存在两个突出问题，即外观不均匀和位置不一致。本文提出了一种新的基于补丁匹配的方法，使用密集连接的条件随机场（CRF）优化来从多时相航空图像中检测建筑物变化。首先，将多时相航空图像进行组合，使用面向对象的技术获取变化信息，然后使用基于深度卷积神经网络的语义分割来提取建筑物区域。利用变化信息和提取的建筑物，应用基于图割的分割算法生成多时相变化建筑物提案。接下来，在多时相变化建筑物提案中，通过相位一致性（PC）模型集成角点和边缘信息进行特征检测，并使用称为定向 PC 直方图的结构特征描述符进行基于补丁的屋顶匹配。通过基于 CRF 的协同细化，收集匹配的屋顶和多时相变化建筑物提案，从而确定建筑物的最终变化，并将其进一步分类为“新建”、“拆除”或“变化”。实验采用了两个典型的数据集，涵盖了具有多种建筑物类型的复杂城市场景。结果表明，所提出的算法具有有效性和通用性，总体准确率和完整性分别超过 85%和 90%。
