Agricultural greenhouses (AGs) are an important component of modern facility agriculture, and accurately mapping and dynamically monitoring their distribution are necessary for agricultural scientific management and planning. Semantic segmentation can be adopted for AG extraction from remote sensing images. However, the feature maps obtained by traditional deep convolutional neural network (DCNN)-based segmentation algorithms blur spatial details and insufficient attention is usually paid to contextual representation. Meanwhile, the maintenance of the original morphological characteristics, especially the boundaries, is still a challenge for precise identification of AGs. To alleviate these problems, this paper proposes a novel network called high-resolution boundary refined network (HBRNet). In this method, we design a new backbone with multiple paths based on HRNetV2 aiming to preserve high spatial resolution and improve feature extraction capability, in which the Pyramid Cross Channel Attention (PCCA) module is embedded to residual blocks to strengthen the interaction of multiscale information. Moreover, the Spatial Enhancement (SE) module is employed to integrate the contextual information of different scales. In addition, we introduce the Spatial Gradient Variation (SGV) unit in the Boundary Refined (BR) module to couple the segmentation task and boundary learning task, so that they can share latent high-level semantics and interact with each other, and combine this with the joint loss to refine the boundary. In our study, GaoFen-2 remote sensing images in Shouguang City, Shandong Province, China are selected to make the AG dataset. The experimental results show that HBRNet demonstrates a significant improvement in segmentation performance up to an IoU score of 94.89%, implying that this approach has advantages and potential for precise identification of AGs.

Remote sensing nighttime lights (NTLs) offers a unique perspective on human activity, and NTL images are widely used in urbanization monitoring, light pollution, and other human-related research. As one of the payloads of sustainable development science Satellite-1 (SDGSAT-1), the Glimmer Imager (GI) provides a new multi-spectral, high-resolution, global coverage of NTL images. However, during the on-orbit testing of SDGSAT-1, a large number of stripes with bad or corrupted pixels were observed in the L1A GI image, which directly affected the accuracy and availability of data applications. Therefore, we propose a novel destriping algorithm based on anomaly detection and spectral similarity restoration (ADSSR) for the GI image. The ADSSR algorithm mainly consists of three parts: pretreatment, stripe detection, and stripe restoration. In the pretreatment, salt-pepper noise is suppressed by setting a minimum area threshold of the connected components. Then, during stripe detections, the valid pixel number sequence and the total pixel value sequence are analyzed to determine the location of stripes, and the abnormal pixels of each stripe are estimated by a clustering algorithm. Finally, a spectral-similarity-based method is adopted to restore all abnormal pixels of each stripe in the stripe restoration. In this paper, the ADSSR algorithm is compared with three representative destriping algorithms, and the robustness of the ADSSR algorithm is tested on different sizes of GI images. The results show that the ADSSR algorithm performs better than three representative destriping algorithms in terms of visual and quantitative indexes and still maintains outstanding performance and robustness in differently sized GI images.

Coastal aquaculture areas are some of the main areas to obtain marine fishery resources and are vulnerable to storm-tide disasters. Obtaining the information of coastal aquaculture areas quickly and accurately is important for the scientific management and planning of aquaculture resources. Recently, deep neural networks have been widely used in remote sensing to deal with many problems, such as scene classification and object detection, and there are many data sources with different spatial resolutions and different uses with the development of remote sensing technology. Thus, using deep learning networks to extract coastal aquaculture areas often encounters the following problems: (1) the difficulty in labeling; (2) the poor robustness of the model; (3) the spatial resolution of the image to be processed is inconsistent with that of the existing samples. In order to fix these problems, this paper proposes a novel semi-/weakly-supervised method, the semi-/weakly-supervised semantic segmentation network (Semi-SSN), and adopts 3 data sources: GaoFen-2 image, GaoFen-1(PMS)image, and GanFen-1(WFV)image with a 0.8 m, 2 m, and 16 m spatial resolution, respectively, and through experiments, we analyze the extraction effect of the model comprehensively. After comparing with other the-state-of-art methods and verifying on an open remote sensing dataset, we take the Fujian coastal area (mainly Sanduo) as the experimental area and employ our method to detect the effect of storm-tide disasters on coastal aquaculture areas, monitor the production, and make the distribution map of coastal aquaculture areas.

High-resolution thermal infrared (TIR) remote sensing images can more accurately retrieve land surface temperature and describe the spatial pattern of urban thermal environment. The Thermal Infrared Spectrometer (TIS), which has high spatial resolution among spaceborne thermal infrared sensors at present, and global data acquisition capability, is one of the sensors equipped in the SDGSAT-1. It is an important complement to the existing international mainstream satellites. In order to produce standard data products, rapidly and accurately, the automatic registration and geometric correction method needs to be developed. Unlike visible–visible image registration, thermal infrared images are blurred in edge details and have obvious non-linear radiometric differences from visible images, which make it challenging for the TIR-visible image registration task. To address these problems, homomorphic filtering is employed to enhance TIR image details and the modified RIFT algorithm is proposed to achieve TIR-visible image registration. Different from using MIM for feature description in RIFT, the proposed modified RIFT uses the novel binary pattern string to descriptor construction. With sufficient and uniformly distributed ground control points, the two-step orthorectification framework, from SDGSAT-1 TIS L1A image to L4 orthoimage, are proposed in this study. The first experiment, with six TIR-visible image pairs, captured in different landforms, is performed to verify the registration performance, and the result indicates that the homomorphic filtering and modified RIFT greatly increase the number of corresponding points. The second experiment, with one scene of an SDGSAT-1 TIS image, is executed to test the proposed orthorectification framework. Subsequently, 52 GCPs are selected manually to evaluate the orthorectification accuracy. The result indicates that the proposed orthorectification framework is helpful to improve the geometric accuracy and guarantee for the subsequent thermal infrared applications.

It is crucial to detect and classify industrial heat sources for sustainable industrial development. Sustainable Development Science Satellite 1 (SDGSAT-1) thermal infrared spectrometer (TIS) data were first introduced for detecting industrial heat source production areas to address the difficulty in identifying factories with low combustion temperatures and small scales. In this study, a new industrial heat source identification and classification model using SDGSAT-1 TIS and Landsat 8/9 Operational Land Imager (OLI) data was proposed to improve the accuracy and granularity of industrial heat source recognition. First, multiple features (thermal and optical features) were extracted using SDGSAT-1 TIS and Landsat 8/9 OLI data. Second, an industrial heat source identification model based on a support vector machine (SVM) and multiple features was constructed. Then, industrial heat sources were generated and verified based on the topological correlation between the identification results of the production areas and Google Earth images. Finally, the industrial heat sources were classified into six categories based on point-of-interest (POI) data. The new model was applied to the Beijing–Tianjin–Hebei (BTH) region of China. The results showed the following: (1) Multiple features enhance the differentiation and identification accuracy between industrial heat source production areas and the background. (2) Compared to active-fire-point (ACF) data (375 m) and Landsat 8/9 thermal infrared sensor (TIRS) data (100 m), nighttime SDGSAT-1 TIS data (30 m) facilitate the more accurate detection of industrial heat source production areas. (3) Greater than 2~6 times more industrial heat sources were detected in the BTH region using our model than were reported by Ma and Liu. Some industrial heat sources with low heat emissions and small areas (53 thermal power plants) were detected for the first time using TIS data. (4) The production areas of cement plants exhibited the highest brightness temperatures, reaching 301.78 K, while thermal power plants exhibited the lowest brightness temperatures, averaging 277.31 K. The production areas and operational statuses of factories could be more accurately identified and monitored with the proposed approach than with previous methods. A new way to estimate the thermal and air pollution emissions of industrial enterprises is presented.

Massive and diverse remote sensing data provide opportunities for data-driven tasks in the real world, but also present challenges in terms of data processing and analysis, especially pixel-level image interpretation. However, the existing shallow-learning and deep-learning segmentation methods, bounded by their technical bottlenecks, cannot properly balance accuracy and efficiency, and are thus hardly scalable to the practice scenarios of remote sensing in a successful way. Instead of following the time-consuming deep stacks of local operations as most state-of-the-art segmentation networks, we propose a novel segmentation model with the encoder–decoder structure, dubbed XANet, which leverages the more computationally economical attention mechanism to boost performance. Two novel attention modules in XANet are proposed to strengthen the encoder and decoder, respectively, namely the Attention Recalibration Module (ARM) and Attention Fusion Module (AFM). Unlike current attention modules, which only focus on elevating the feature representation power, and regard the spatial and channel enhancement of a feature map as two independent steps, ARM gathers element-wise semantic descriptors coupling spatial and channel information to directly generate a 3D attention map for feature enhancement, and AFM innovatively utilizes the cross-attention mechanism for the sufficient spatial and channel fusion of multi-scale features. Extensive experiments were conducted on ISPRS and GID datasets to comprehensively analyze XANet and explore the effects of ARM and AFM. Furthermore, the results demonstrate that XANet surpasses other state-of-the-art segmentation methods in both model performance and efficiency, as ARM yields a superior improvement versus existing attention modules with a competitive computational overhead, and AFM achieves the complementary advantages of multi-level features under the sufficient consideration of efficiency.