Semantic segmentation is a fundamental task in remote sensing image analysis (RSIA). Fully convolutional networks (FCNs) have achieved state-of-the-art performance in the task of semantic segmentation of natural scene images. However, due to distinctive differences between natural scene images and remotely-sensed (RS) images, FCN-based semantic segmentation methods from the field of computer vision cannot achieve promising performances on RS images without modifications. In previous work, we proposed an RS image semantic segmentation framework SDFCNv1, combined with a majority voting postprocessing method. Nevertheless, it still has some drawbacks, such as small receptive field and large number of parameters. In this paper, we propose an improved semantic segmentation framework SDFCNv2 based on SDFCNv1, to conduct optimal semantic segmentation on RS images. We first construct a novel FCN model with hybrid basic convolutional (HBC) blocks and spatial-channel-fusion squeeze-and-excitation (SCFSE) modules, which occupies a larger receptive field and fewer network model parameters. We also put forward a data augmentation method based on spectral-specific stochastic-gamma-transform-based (SSSGT-based) during the model training process to improve generalizability of our model. Besides, we design a mask-weighted voting decision fusion postprocessing algorithm for image segmentation on overlarge RS images. We conducted several comparative experiments on two public datasets and a real surveying and mapping dataset. Extensive experimental results demonstrate that compared with the SDFCNv1 framework, our SDFCNv2 framework can increase the mIoU metric by up to 5.22% while only using about half of parameters.

In recent years, remote sensing techniques such as satellite and drone-based imaging have been used to monitor Pine Wilt Disease (PWD), a widespread forest disease that causes the death of pine species. Researchers have explored the use of remote sensing imagery and deep learning algorithms to improve the accuracy of PWD detection at the single-tree level. This study introduces a novel framework for PWD detection that combines high-resolution RGB drone imagery with free-access Sentinel-2 satellite multi-spectral imagery. The proposed approach includes an PWD-infected tree detection model named YOLOv5-PWD and an effective data augmentation method. To evaluate the proposed framework, we collected data and created a dataset in Xianning City, China, consisting of object detection samples of infected trees at middle and late stages of PWD. Experimental results indicate that the YOLOv5-PWD detection model achieved 1.2% higher mAP compared to the original YOLOv5 model and a further improvement of 1.9% mAP was observed after applying our dataset augmentation method, which demonstrates the effectiveness and potential of the proposed framework for PWD detection.

Focusing on problems of blurred detection boundary, small target miss detection, and more pseudo changes in high-resolution remote sensing image change detection, a change detection algorithm based on Siamese neural networks is proposed. Siam-FAUnet can implement end-to-end change detection tasks. Firstly, the improved VGG16 is utilized as an encoder to extract the image features. Secondly, the atrous spatial pyramid pooling module is used to increase the receptive field of the model to make full use of the global information of the image and obtain the multi-scale contextual information of the image. The flow alignment module is used to fuse the low-level features in the encoder to the decoder and solve the problem of semantic misalignment caused by the direct concatenation of features when the features are fused, so as to obtain the change region of the image. The experiments are trained and tested using publicly available CDD and SZTAKI datasets. The results show that the evaluation metrics of the Siam-FAUnet model are improved compared to the baseline model, in which the F1-score is improved by 4.00% on the CDD and by 7.32% and 2.62% on the sub-datasets of SZTAKI (SZADA and TISZADOB), respectively; compared to other state-of-the-art methods, the Siam-FAUnet model has improved in both evaluation metrics, indicating that the model has a good detection performance.

Accurate daily surface soil moisture (SSM) mapping at high spatial resolution (e.g., 30 m) remains challenging due to individual satellite sensor limitations. Although passive microwave sensors provide frequent coarse-resolution observations and synthetic aperture radar (SAR) offers high-resolution data intermittently, achieving both simultaneously requires sensor synergy. This paper introduces the microwave–optical multi-stage synergistic downscaling framework (MMSDF) to generate daily 30 m SSM products. The framework integrates SMAP L4 (9 km), MODIS data (500 m–1 km), harmonized Landsat Sentinel-2 (HLS, 30 m), radiometric terrain corrected Sentinel-1 (RTC-S1, 30 m), and auxiliary geographic data. It comprises three stages: (1) downscaling SMAP L4 to 1 km via random forest; (2) calibrating Sentinel-1 water cloud model (WCM) using intermediate 1 km SSM to retrieve 30 m SSM without in situ calibration; and (3) fusing daily 1 km SSM and intermittent 30 m WCM-derived retrievals using the spatial–temporal fusion model (ESTARFM) to generate seamless daily 30 m SSM maps. Validation against in situ measurements from 16 sites in Hunan Province, China (summer 2024) yielded R of 0.54 and RMSE of 0.045 ${\mathrm{cm}}^3/{\mathrm{cm}}^3$. Results demonstrate the framework’s capability to synergize multi-source data for high-resolution daily SSM estimates valuable for hydrological and agricultural applications.

The surface fragmentation of karst landscapes leads to a high degree of coupling between fire scar site boundaries and topographic relief. However, the applicability of spatio-temporal data fusion methods for fire scar extraction in such geomorphological areas remains systematically unevaluated. This study developed a spatial–temporal adaptive fusion model integrating Landsat 30-m data with MODIS daily observations to generate continuous high-precision dNBR datasets. Using a typical karst fire region in Guizhou and Yunnan, China, as a case study, we validated the method’s effectiveness for fire trace extraction in fragmented landscapes. The proposed fusion technique addresses cloud cover limitations in humid climates by constructing continuous NBR time series, enabling precise fire boundary delineation and severity quantification. We comparatively implemented multiple fusion approaches (FSDAF, STARFM, and STDFA) and evaluated their performance through both spectral (RMSE, AD, and PSNR) and spatial (Edge, LBP, and SSIM) metrics. Key findings include the following: (1) FSDAF outperformed other methods in spectral consistency and spatial adaptation, particularly for heterogeneous mountainous terrain with fragmented vegetation. (2) Comparative experiments demonstrated that pre-calculating vegetation indices before temporal fusion (Strategy I) produced superior results to post-fusion calculation (Strategy II). Moreover, in our karst landscape study area, our proposed Hybrid Strategy selection framework can dynamically optimize the fusion process of multi-source satellite data, which is significantly better than a single fusion strategy. (3) The dNBR-based extraction achieved 90.00% producer accuracy, 69.23% user accuracy, and a Kappa coefficient of 0.718 when validated against field data. This study advances fire monitoring in karst regions by significantly improving both the spatio-temporal resolution and accuracy of burn scar detection compared to conventional approaches. The framework provides a viable solution for fire impact assessment in topographically complex landscapes under cloudy conditions.

Grain size analysis is used to study grain size and distribution. It is a critical indicator in sedimentary simulation experiments (SSEs), which aids in understanding hydrodynamic conditions and identifying the features of sedimentary environments. Existing methods for grain size analysis based on images primarily focus on scenarios where grain edges are distinct or grain arrangements are regular. However, these methods are not suitable for images from SSEs. We proposed a deep learning model incorporating histogram layers for the analysis of SSE images with fuzzy grain edges and irregular arrangements. Firstly, ResNet18 was used to extract features from SSE images. These features were then input into the histogram layer to obtain local histogram features, which were concatenated to form comprehensive histogram features for the entire image. Finally, the histogram features were connected to a fully connected layer to estimate the grain size corresponding to the cumulative volume percentage. In addition, an applied workflow was developed. The results demonstrate that the proposed method achieved higher accuracy than the eight other models and was highly consistent with manual results in practice. The proposed method enhances the efficiency and accuracy of grain size analysis for images with irregular grain distribution and improves the quantification and automation of grain size analysis in SSEs. It can also be applied for grain size analysis in fields such as soil and geotechnical engineering.