Convolutional neural networks (CNNs) have been increasingly used in remote sensing scene classification/recognition. The conventional CNNs are sensitive to the rotation of the image scene, which will inevitably result in the misclassification of remote sensing scene images that belong to the same category. In this work, we equip the networks with a new pooling strategy, “concentric circle pooling”, to alleviate the above problem. The new network structure, called CCP-net can generate a concentric circle-based spatial-rotation-invariant representation of an image, hence improving the classification accuracy. The square kernel is adopted to approximate the circle kernels in concentric circle pooling, which is much more efficient and suitable for CNNs to propagate gradients. We implement the training of the proposed network structure with standard back-propagation, thus CCP-net is an end-to-end trainable CNNs. With these advantages, CCP-net should in general improve CNN-based remote sensing scene classification methods. Experiments using two publicly available remote sensing scene datasets demonstrate that using CCP-net can achieve competitive classification results compared with the state-of-art methods.

Scene classification is an active research area in the remote sensing (RS) domain. Some categories of RS scenes, such as medium residential and dense residential scenes, would contain the same type of geographical objects but have various spatial distributions among these objects. The adjacency and disjointness relationships among geographical objects are normally neglected by existing RS scene classification methods using convolutional neural networks (CNNs). In this study, a multi-output network (MopNet) combining a graph neural network (GNN) and a CNN is proposed for RS scene classification with a joint loss. In a candidate RS image for scene classification, superpixel regions are constructed through image segmentation and are represented as graph nodes, while graph edges between nodes are created according to the spatial adjacency among corresponding superpixel regions. A training strategy of a jointly learning CNN and GNN is adopted in the MopNet. Through the message propagation mechanism of MopNet, spatial and topological relationships imbedded in the edges of graphs are employed. The parameters of the CNN and GNN in MopNet are updated simultaneously with the guidance of a joint loss via the backpropagation mechanism. Experimental results on the OPTIMAL-31 and aerial image dataset (AID) datasets show that the proposed MopNet combining a graph convolutional network (GCN) or graph attention network (GAT) and ResNet50 achieves state-of-the-art accuracy. The overall accuracy obtained on OPTIMAL-31 is 96.06% and those on AID are 95.53% and 97.11% under training ratios of 20% and 50%, respectively. Spatial and topological relationships imbedded in RS images are helpful for improving the performance of scene classification.

This paper presents a novel approach for automatically detecting land cover changes from multitemporal high-resolution remote sensing images in the deep feature space. This is accomplished by using multitemporal deep feature collaborative learning and a semi-supervised Chan–Vese (SCV) model. The multitemporal deep feature collaborative learning model is developed to obtain the multitemporal deep feature representations in the same high-level feature space and to improve the separability between changed and unchanged patterns. The deep difference feature map at the object-level is then extracted through a feature similarity measure. Based on the deep difference feature map, the SCV model is proposed to detect changes in which labeled patterns automatically derived from uncertainty analysis are integrated into the energy functional to efficiently drive the contour towards accurate boundaries of changed objects. The experimental results obtained on the four data sets acquired by different high-resolution sensors corroborate the effectiveness of the proposed approach.

Accurate geographic data of slums are important for handling urban poverty issues. Previous slum mapping studies using high-resolution or very-high-resolution (HR/VHR) remotely sensed (RS) images are commonly not suitable for city-wide scale tasks. This study aims to efficiently generate a slum map on a city-wide scale using freely accessed multispectral medium-resolution (MR) Sentinel-2 images. Composite slum spectral indices (CSSIs) were initially proposed based on the shapes of spectral profiles of slums and nonslums and directly represent slum characteristics. Specifically, CSSI-1 denotes the normalized difference between the shortwave infrared bands and the red edge band, while CSSI-2 denotes the normalized difference between the blue band and the green band. Furthermore, two methods were developed to test the effectiveness of CSSIs on slum mapping, i.e., the threshold-based method and the machine learning (ML)-based method. Experimental results show that the threshold-based method and the ML-based method achieve intersection over unions (IoU) of 43.89% and 54.45% in Mumbai, respectively. The accuracies of our methods are comparable to or even higher than the accuracies reported by existing methods using HR/VHR images and transfer learning. The threshold-based method exhibits a promising performance in mapping slums larger than 5 ha, while the ML-based method refines mapping accuracies for slum pockets smaller than 5 ha. The threshold-based method and the ML-based method produced the slum map in Mumbai in 2 and 28 min, respectively. Our methods are suitable for rapid large-area slum mapping owing to the high data availability of Sentinel-2 images and high computational efficiency.

Approximately 1 billion people worldwide currently inhabit slum areas. The UN Sustainable Development Goal (SDG 11.1) underscores the imperative of upgrading all slums by 2030 to ensure adequate housing for everyone. Geo-locations of slums help local governments with upgrading slums and alleviating urban poverty. Remote sensing (RS) technology, with its excellent Earth observation capabilities, can play an important role in slum mapping. Deep learning (DL)-based RS information extraction methods have attracted a lot of attention. Currently, DL-based slum mapping studies typically uses three optical bands to adapt to existing models, neglecting essential geo-scientific information, such as spectral and textural characteristics, which are beneficial for slum mapping. Inspired by the geoscience-aware DL paradigm, we propose the Geoscience-Aware Network for slum mapping (GASlumNet), aiming to improve slum mapping accuracies via incorporating the DL model with geoscientific prior knowledge. GASlumNet employs a two-stream architecture, combining ConvNeXt and UNet. One stream concentrates on optical feature representation, while the other emphasizes geo-scientific features. Further, the feature-level and decision-level fusion mechanisms are applied to optimize deep features and enhance model performance. We used Jilin-1 Spectrum 01 and Sentinel-2 images to perform experiments in Mumbai, India. The results demonstrate that GASlumNet achieves higher slum mapping accuracy than the comparison models, with an intersection over union (IoU) of 58.41%. Specifically, GASlumNet improves the IoU by 4.60~5.97% over the baseline models, i.e., UNet and ConvNeXt-UNet, which exclusively utilize optical bands. Furthermore, GASlumNet enhances the IoU by 10.97% compared to FuseNet, a model that combines optical bands and geo-scientific features. Our method presents a new technical solution to achieve accurate slum mapping, offering potential benefits for regional and global slum mapping and upgrading initiatives.

Within-class spectral variation and between-class spectral confusion in remotely sensed imagery degrades the performance of built-up area detection when using planar texture, shape, and spectral features. Terrain slopes and building heights extracted from auxiliary data, such as Digital Surface Models (DSMs) however, can improve the results. Stereo imagery incorporates height information unlike single remotely sensed images. In this study, a new Stereo Pair Disparity Index (SPDI) for indicating built-up areas is calculated from stereo-extracted disparity information. Further, a new method of detecting built-up areas from stereo pairs is proposed based on the SPDI, using disparity information to establish the relationship between two images of a stereo pair. As shown in the experimental results for two stereo pairs covering different scenes with diverse urban settings, the SPDI effectively differentiates between built-up and non-built-up areas. Our proposed method achieves higher accuracy built-up area results from stereo images than the traditional method for single images, and two other widely-applied DSM-based methods for stereo images. Our approach is suitable for spaceborne and airborne stereo pairs and triplets. Our research introduces a new effective height feature (SPDI) for detecting built-up areas from stereo imagery with no need for DSMs.