Deep learning approaches have been widely used in building automatic extraction tasks and have made great progress in recent years. However, the missing detection and wrong detection causing by spectrum confusion is still a great challenge. The existing fully convolutional networks (FCNs) cannot effectively distinguish whether the feature differences are from one building or the building and its adjacent non-building objects. In order to overcome the limitations, a building multi-feature fusion refined network (BMFR-Net) was presented in this paper to extract buildings accurately and completely. BMFR-Net is based on an encoding and decoding structure, mainly consisting of two parts: the continuous atrous convolution pyramid (CACP) module and the multiscale output fusion constraint (MOFC) structure. The CACP module is positioned at the end of the contracting path and it effectively minimizes the loss of effective information in multiscale feature extraction and fusion by using parallel continuous small-scale atrous convolution. To improve the ability to aggregate semantic information from the context, the MOFC structure performs predictive output at each stage of the expanding path and integrates the results into the network. Furthermore, the multilevel joint weighted loss function effectively updates parameters well away from the output layer, enhancing the learning capacity of the network for low-level abstract features. The experimental results demonstrate that the proposed BMFR-Net outperforms the other five state-of-the-art approaches in both visual interpretation and quantitative evaluation.

Clouds and snow in remote sensing imageries cover underlying surface information, reducing image availability. Moreover, they interact with each other, decreasing the cloud and snow detection accuracy. In this study, we propose a convolutional neural network for cloud and snow detection, named the cloud and snow detection network (CSD-Net). It incorporates the multi-scale feature fusion module (MFF) and the controllably deep supervision and feature fusion structure (CDSFF). MFF can capture and aggregate features at various scales, ensuring that the extracted high-level semantic features of clouds and snow are more distinctive. CDSFF can provide a deeply supervised mechanism with hinge loss and combine information from adjacent layers to gain more representative features. It ensures the gradient flow is more oriented and error-less, while retaining more effective information. Additionally, a high-resolution cloud and snow dataset based on WorldView2 (CSWV) was created and released. This dataset meets the training requirements of deep learning methods for clouds and snow in high-resolution remote sensing images. Based on the datasets with varied resolutions, CSD-Net is compared to eight state-of-the-art deep learning methods. The experiment results indicate that CSD-Net has an excellent detection accuracy and efficiency. Specifically, the mean intersection over the union (MIoU) of CSD-Net is the highest in the corresponding experiment. Furthermore, the number of parameters in our proposed network is just 7.61 million, which is the lowest of the tested methods. It only has 88.06 GFLOPs of floating point operations, which is less than the U-Net, DeepLabV3+, PSPNet, SegNet-Modified, MSCFF, and GeoInfoNet. Meanwhile, CSWV has a higher annotation quality since the same method can obtain a greater accuracy on it.

Automatic building extraction from high resolution remains challenging for widely applications. Previous studies have reviewed building extraction methods, but rapid advances in deep learning (DL) highlight the need for an updated review focused on HRSI applications. Therefore, we reviewed 243 papers to identify future research trends and highlight five key challenges in HRSI-based building extraction: complex boundary optimization, shape and spectrum variability, insufficient samples, tree and shadow interference, and lightweight model applications. Subsequently, we explore five corresponding opportunities leveraging advanced DL techniques. Existing methods are then categorized into two main groups: non-deep learning (NDL)-based and DL-based approaches. These are further subdivided into four broad categories: traditional methods, machine learning-based methods, semantic segmentation-based methods, and vector-based methods, which are discussed in detail based on their underlying principles. Additionally, we present several practical applications and introduce 10 publicly available benchmarks. We conducted a comprehensive evaluation of nine representative methods alongside our proposed Multiple-parallel vision Mamba network (MVMNet) using three public datasets: WHU, Massachusetts, and WHU Satellite I building datasets. Among these methods, HD-Net and MSSDMAP-Net demonstrated superior performance in addressing challenges associated with complex boundaries, while BuildFormerSegDP and CBRNet showed enhanced effectiveness in mitigating the impact of tree occlusion. Notably, MVMNet outperforms the other methods in addressing the challenges associated with spectral heterogeneity within buildings, achieving optimal Intersection over Union (IoU) values of 0.9076, 0.7376, and 0.6612, respectively. Finally, we explore a range of promising research directions and offer references and insights for future researchers. (Github: https://github.com/329719386/building-review).

Building extraction is significant for the intelligent interpretation of high-resolution remote sensing images (HRSI). However, in some complex scenarios where the features of the building and its adjacent ground objects are similar, the current segmentation model cannot distinguish them effectively. Therefore, we propose a complex scenarios adaptive network (CSA-Net) for building extraction. CSA-Net is comprised of the hierarchical-context feature extraction (HFE) module, the global-local feature interaction (GFI) module, and the multiscale-adaptive feature fusion (MFF) structure. The HFE obtains high-level semantic information at different levels and fuses it with low-level detailed information by skipping connections to enhance the reasoning and perception ability of building structure in complex scenes. Then, the GFI acquires global-local features of buildings and their surrounding environment via dense multiscale dilated convolution. The information can be shared through efficient interaction among features, and irrelevant backgrounds can be suppressed. Then, in the up-sampling process, the MFF alleviates the feature loss and enhances the robustness of the network by using feature fusion after layer-by-layer adaptive weight allocation. Experiments show that CSA-Net outperforms other comparable methods, with intersection over union (IoU) values of 79.99%, 89.75%, and 73.59%, respectively, on the Google Arlinton (GA), WHU, and Massachusetts building datasets. The visual comparison results demonstrate that our method can enhance the accuracy of building extraction in complex scenes. Meanwhile, the efficiency results indicate our approach strikes a balance between calculation parameters and time and achieves high levels of efficiency. (Github: https://github.com/329719386/CSA-Net)

In object-based image analysis of high-resolution images, the number of features can reach hundreds, so it is necessary to perform feature reduction prior to classification. In this paper, a feature selection method based on the combination of a genetic algorithm (GA) and tabu search (TS) is presented. The proposed GATS method aims to reduce the premature convergence of the GA by the use of TS. A prematurity index is first defined to judge the convergence situation during the search. When premature convergence does take place, an improved mutation operator is executed, in which TS is performed on individuals with higher fitness values. As for the other individuals with lower fitness values, mutation with a higher probability is carried out. Experiments using the proposed GATS feature selection method and three other methods, a standard GA, the multistart TS method, and ReliefF, were conducted on WorldView-2 and QuickBird images. The experimental results showed that the proposed method outperforms the other methods in terms of the final classification accuracy.

Accurate semantic segmentation results of the overhead catenary system (OCS) are significant for OCS component extraction and geometric parameter detection. Actually, the scenes of OCS are complex, and the density of point cloud data obtained through Light Detection and Ranging (LiDAR) scanning is uneven due to the character difference of OCS components. However, due to the inconsistent component points, it is challenging to complete better semantic segmentation of the OCS point cloud with the existing deep learning methods. Therefore, this paper proposes a point cloud multi-scale feature fusion refinement structure neural network (PMFR-Net) for semantic segmentation of the OCS point cloud. The PMFR-Net includes a prediction module and a refinement module. The innovations of the prediction module include the double efficient channel attention module (DECA) and the serial hybrid domain attention (SHDA) structure. The point cloud refinement module (PCRM) is used as the refinement module of the network. DECA focuses on detail features; SHDA strengthens the connection of contextual semantic information; PCRM further refines the segmentation results of the prediction module. In addition, this paper created and released a new dataset of the OCS point cloud. Based on this dataset, the overall accuracy (OA), F1-score, and mean intersection over union (MIoU) of PMFR-Net reached 95.77%, 93.24%, and 87.62%, respectively. Compared with four state-of-the-art (SOTA) point cloud deep learning methods, the comparative experimental results showed that PMFR-Net achieved the highest accuracy and the shortest training time. At the same time, PMFR-Net segmentation performance on S3DIS public dataset is better than the other four SOTA segmentation methods. In addition, the effectiveness of DECA, SHDA structure, and PCRM was verified in the ablation experiment. The experimental results show that this network could be applied to practical applications.