Segment anything model (SAM) has been widely applied to various downstream tasks for its excellent performance and generalization capability. However, SAM exhibits three limitations related to remote sensing (RS) semantic segmentation task: 1) the image encoders excessively lose high-frequency information, such as object boundaries and textures, resulting in rough segmentation masks; 2) due to being trained on natural images, SAM faces difficulty in accurately recognizing objects with large-scale variations and uneven distribution in RS images; and 3) the output tokens used for mask prediction are trained on natural images and not applicable to RS image segmentation. In this article, we explore an efficient paradigm for applying SAM to the semantic segmentation of RS images. Furthermore, we propose multiscale enhanced SAM (MeSAM), a new SAM fine-tuning method more suitable for RS images to adapt it to semantic segmentation tasks. Our method first introduces an inception mixer into the image encoder to effectively preserve high-frequency features. Second, by designing a mask decoder with RS correction and incorporating multiscale connections, we make up the difference in SAM from natural images to RS images. Experimental results demonstrated that our method significantly improves the segmentation accuracy of SAM for RS images, outperforming some state-of-the-art (SOTA) methods. The code will be available at https://github.com/Magic-lem/MeSAM .

Convolutional neural networks (CNNs) have recently achieved outstanding performance for hyperspectral (HS) and multispectral (MS) image fusion. However, CNNs cannot explore the long-range dependence for HS and MS image fusion because of their local receptive fields. To overcome this limitation, a transformer is proposed to leverage the long-range dependence from the network inputs. Because of the ability of long-range modeling, the transformer overcomes the sole CNN on many tasks, whereas its use for HS and MS image fusion is still unexplored. In this article, we propose a spectral–spatial transformer (SST) to show the potentiality of transformers for HS and MS image fusion. We devise first two branches to extract spectral and spatial features in the HS and MS images by SST blocks, which can explore the spectral and spatial long-range dependence, respectively. Afterward, spectral and spatial features are fused feeding the result back to spectral and spatial branches for information interaction. Finally, the high-resolution (HR) HS image is reconstructed by dense links from all the fused features to make full use of them. The experimental analysis demonstrates the high performance of the proposed approach compared with some state-of-the-art (SOTA) methods.

Recently, deep learning (DL)-based industrial applications have attracted broad attention due to their advanced performance. However, the limited computational resource in portable devices always makes big DL models inapplicable in the industry. DL-based single-image super-resolution also encounters this problem because of its large computations. Besides, most lightweight convolutional-neural-network-based methods utilize features insufficiently, which restricts their capability for industrial reconstruction. To alleviate this problem, we present a progress interaction-learning network (PILN) to refine features at different levels: at the global level, we employ a progressive interaction-learning strategy to integrate hierarchical features in temporal and spatial dimensions; at the mediate level, enhanced interaction-learning units, adopting the enhanced interactive study, significantly boost the reconstruction performance; at the local level, employing pixelwise learning, residual cells are raised to search for an optimal information flow by weight distribution. Extensive experiments demonstrate that the PILN outperforms other state-of-the-art methods.

Recent developments and the widespread adoption of remote sensing data (RSD) gave rise to various hyperspectral imaging applications (HSI). With its detailed spectral information, HSI has been adopted for use in various agriculture-related applications. These applications demand more accurate classification, highlighting the significance of HSI in plant disease detection, crop classification, etc. Despite existing models for RSD-based agricultural applications, such models lack generalizability for plant classification. This task is challenging for the UNet-based architectures due to non-linear combinations of the pixels in the hyperspectral image, reduced and diverse information of each pixel, and high dimensions. Furthermore, a shortage of labeled data makes achieving high classification accuracy challenging. This study proposes an improved 3D UNet architecture based on a modified convolutional neural network that uses spatial and spectral information. This approach solves the limitations of existing UNet-based models, which suffer to deal with non-linear combinations and reduced and diverse information of small pixels. The proposed model employs a semantic segmentation strategy with modified architecture for more accurate classification and segmentation. The studies employ widely recognized benchmark HSI datasets, such as the Indian Pines (IP), Salinas, Pavia University (PU), Honghu, and Xiong'an datasets. These datasets are assessed using average accuracy, overall accuracy, and the Kappa coefficient. The suggested model demonstrated exceptional classification accuracy, achieving 99.60% for the Indian Pines dataset and 99.67% for the Pavia University dataset. The proposed approach is further validated and proven to be superior and robust through comparisons with existing models.