Automatic building segmentation from aerial imagery is an important and challenging task because of the variety of backgrounds, building textures and imaging conditions. Currently, research using variant types of fully convolutional networks (FCNs) has largely improved the performance of this task. However, pursuing more accurate segmentation results is still critical for further applications such as automatic mapping. In this study, a multi-constraint fully convolutional network (MC–FCN) model is proposed to perform end-to-end building segmentation. Our MC–FCN model consists of a bottom-up/top-down fully convolutional architecture and multi-constraints that are computed between the binary cross entropy of prediction and the corresponding ground truth. Since more constraints are applied to optimize the parameters of the intermediate layers, the multi-scale feature representation of the model is further enhanced, and hence higher performance can be achieved. The experiments on a very-high-resolution aerial image dataset covering 18 km ${}^2$ and more than 17,000 buildings indicate that our method performs well in the building segmentation task. The proposed MC–FCN method significantly outperforms the classic FCN method and the adaptive boosting method using features extracted by the histogram of oriented gradients. Compared with the state-of-the-art U–Net model, MC–FCN gains 3.2% (0.833 vs. 0.807) and 2.2% (0.893 vs. 0.874) relative improvements of Jaccard index and kappa coefficient with the cost of only 1.8% increment of the model-training time. In addition, the sensitivity analysis demonstrates that constraints at different positions have inconsistent impact on the performance of the MC–FCN.

In this study, we present the Ensemble Convolutional Neural Network (ECNN), an elaborate CNN frame formulated based on ensembling state-of-the-art CNN models, to identify village buildings from open high-resolution remote sensing (HRRS) images. First, to optimize and mine the capability of CNN for village mapping and to ensure compatibility with our classification targets, a few state-of-the-art models were carefully optimized and enhanced based on a series of rigorous analyses and evaluations. Second, rather than directly implementing building identification by using these models, we exploited most of their advantages by ensembling their feature extractor parts into a stronger model called ECNN based on the multiscale feature learning method. Finally, the generated ECNN was applied to a pixel-level classification frame to implement object identification. The proposed method can serve as a viable tool for village building identification with high accuracy and efficiency. The experimental results obtained from the test area in Savannakhet province, Laos, prove that the proposed ECNN model significantly outperforms existing methods, improving overall accuracy from 96.64% to 99.26%, and kappa from 0.57 to 0.86.

In this study, a method based on supervised machine learning is proposed to identify village buildings from open high-resolution remote sensing images. We select Google Earth (GE) RGB images to perform the classification in order to examine its suitability for village mapping, and investigate the feasibility of using machine learning methods to provide automatic classification in such fields. By analyzing the characteristics of GE images, we design different features on the basis of two kinds of supervised machine learning methods for classification: adaptive boosting (AdaBoost) and convolutional neural networks (CNN). To recognize village buildings via their color and texture information, the RGB color features and a large number of Haar-like features in a local window are utilized in the AdaBoost method; with multilayer trained networks based on gradient descent algorithms and back propagation, CNN perform the identification by mining deeper information from buildings and their neighborhood. Experimental results from the testing area at Savannakhet province in Laos show that our proposed AdaBoost method achieves an overall accuracy of 96.22% and the CNN method is also competitive with an overall accuracy of 96.30%.

The automatic extraction of building outlines from aerial imagery for the purposes of navigation and urban planning is a long-standing problem in the field of remote sensing. Currently, most methods utilize variants of fully convolutional networks (FCNs), which have significantly improved model performance for this task. However, pursuing more accurate segmentation results is still critical for additional applications, such as automatic mapping and building change detection. In this study, we propose a boundary regulated network called BR-Net, which utilizes both local and global information, to perform roof segmentation and outline extraction. The BR-Net method consists of a shared backend utilizing a modified U-Net and a multitask framework to generate predictions for segmentation maps and building outlines based on a consistent feature representation from the shared backend. Because of the restriction and regulation of additional boundary information, the proposed model can achieve superior performance compared to existing methods. Experiments on an aerial image dataset covering 32 km${}^2$ and containing more than 58,000 buildings indicate that our method performs well at both roof segmentation and outline extraction. The proposed BR-Net method significantly outperforms the classic FCN8s model. Compared to the state-of-the-art U-Net model, our BR-Net achieves 6.2% (0.869 vs. 0.818), 10.6% (0.772 vs. 0.698), and 8.7% (0.840 vs. 0.773) improvements in F1 score, Jaccard index, and kappa coefficient, respectively.

Multi-source remote sensing imagery has become widely accessible owing to the development of data acquisition systems. In this paper, we address the challenging task of the semantic segmentation of buildings via multi-source remote sensing imagery with different spatial resolutions. Unlike previous works that mainly focused on optimizing the segmentation model, which did not enable the severe problems caused by the unaligned resolution between the training and testing data to be fundamentally solved, we propose to integrate SR techniques with the existing framework to enhance the segmentation performance. The feasibility of the proposed method was evaluated by utilizing representative multi-source study materials: high-resolution (HR) aerial and low-resolution (LR) panchromatic satellite imagery as the training and testing data, respectively. Instead of directly conducting building segmentation from the LR imagery by using the model trained using the HR imagery, the deep learning-based super-resolution (SR) model was first adopted to super-resolved LR imagery into SR space, which could mitigate the influence of the difference in resolution between the training and testing data. The experimental results obtained from the test area in Tokyo, Japan, demonstrate that the proposed SR-integrated method significantly outperforms that without SR, improving the Jaccard index and kappa by approximately 19.01% and 19.10%, respectively. The results confirmed that the proposed method is a viable tool for building semantic segmentation, especially when the resolution is unaligned.