The objects in optical high-resolution remote sensing images (HRRSIs) are usually tiny, dense, and exist in complex backgrounds, which brings great challenges to accurate object detection. This article presents an improved YOLOv5s network-based technique for remote sensing object recognition to overcome these difficulties. First, unnecessary residual modules are pruned from the cross-stage partial layer of conventional YOLOv5s and a refined residual coordinate attention module is incorporated to enhance the feature representation of the densely packed small objects in HRRSIs by introducing the residual structure and the mix pooling operation instead of the existing average pooling. Second, since various scales of objects are present in HRRSIs, the algorithm of differential evolution is adopted to replace the traditional K-means for generating a variety of anchor boxes in different sizes. Third, we replace the commonly used complete intersection over union (IoU) loss function in YOLOv5s with the AW-IoU loss function based on both α-IoU and wise-IoU. AW-IoU could expedite bounding box regression and focus more on regular anchor boxes. Finally, instead of nonmaximum suppression (NMS), the SCYLLA (S-IoU) soft-NMS is employed to eliminate the redundant duplicate boxes to detect the dense objects in remote sensing images. Experimental results on the NWPU VHR-10 dataset demonstrate that the proposed YOLOv5s method performs well compared with state-of-the-art algorithms.

The detection of remote sensing (RS) objects plays a crucial role in various earth observation tasks. Current RS object detection methods tend to face great challenges due to large-scale variations of object sizes and limited representation of semantic features, which affect the final bounding box regression results in practice. To address the challenges, we propose a novel Multi-branch and Multi-level Feature Enhancement framework (M2FE-YOLO) to refine feature representation learning for improving the performance of object detection. The proposed M2FE-YOLO primarily comprises three components, i.e., Multi-Branch Feature-awareness based CSP (MBFA-CSP) module, Multi-Level Feature Fusion (MLFF) module, and Shape-IoU as a loss function. MBFA-CSP builds on a dual-scale feature-aware mechanism and a serial convolution pathway to dynamically adjust receptive fields, capturing critical contextual patterns in RS images. MLFF resolves the persistent semantic discrepancy between low-level texture features and high-level abstract representations, enabling precise localization of objects in cluttered RS landscapes. Shape-IoU instead of C-IoU incorporates a geometric compatibility factor (i.e., aspect ratio consistency), and is more crucial for bounding box regression of elongated or irregular RS objects. Compared with existing state-of-the-art methods, extensive experiments on RSOD, NWPU VHR-10, and DOTA datasets quantitatively and qualitatively demonstrate the superiority of the proposed M2FE-YOLO method, achieving up to 92.1%, 93.3%, and 72.2% mAP, respectively. Meanwhile, M2FE-YOLO-OBB achieves an excellent detection result of 73.9% mAP on DOTA dataset for oriented object detection task.

The combination of Convolutional neural networks (CNNs) and Vision transformers (ViTs) has garnered considerable attention in hyperspectral image (HSI) classification due to their abilities to enhance the classification accuracy by concurrently extracting local and global features. However, these accuracy improvements come at the cost of significant demands on storage resources, computational overhead, and extensive training samples. To address these challenges, this paper proposes a Multi-scale spatial-spectral shuffling convolution integrated with a three-dimensional lightweight transformer (MSC-3DLT) for HSI classification. This network directly captures three-dimensional structural features throughout the entire feature extraction process, thereby enhancing HSI classification performance even at small sampling rates within a lightweight framework. Specifically, we design a multi-scale spatial-spectral shuffling convolution (MSC) to comprehensively refines spatial-spectral feature granularities and enhance feature interactions by shuffling multi-scale features across different groups. Second, to maximize the exploitation of limited training samples, we rethink transformers from the three-dimensional structural perspective of HSI data and propose a novel lightweight three-dimensional transformer (3DLT). Different from the slicing operation employed in classical transformers, the 3DLT directly extracts the inherent three-dimensional structural features from HSI and mitigates quadratic complexity through a lightweight spatial-spectral pooling cross-attention mechanism (LSPCA). Finally, a novel training strategy is designed to adaptively adjust the learning rate based on multi-metric feedback during the model training process, significantly accelerating the model fitting speed. Extensive experiments demonstrate that the proposed MSC-3DLT method remains highly competitive compared to state-of-the-art methods in terms classification accuracy, model parameters, and FLOPs under small sampling rates. The code of this work is available at https://github.com/H892328874/MSC-3DLT for the sake of reproducibility.

Object detection in high-resolution remote sensing image (HRRSI) faces great challenges of large-scale variations in object size, densely distributed small objects, and complex background interferences. To address these challenges, we propose an innovative single-stage Local-Global Feature Extraction and Coordination Network (LGC-YOLO) to improve the detection accuracy of objects in HRRSIs. LGC-YOLO mainly comprises three modules of Local-Global Spatial Feature Extraction (LGSFE), Gradient Optimized Spatial Information Interaction (GOSII), and Edge-Semantic Feature Coordination Fusion(ESFCF), which synergistically improves the feature extraction and object detection capabilities of LGC-YOLO. Firstly, LGSFE captures local and global features of dense objects through RFAConv and global pooling in multi-branch structure, which effectively alleviates the misalignment between the extracted features of objects and their intrinsic characteristics, thereby providing more accurate and abundant features for subsequent object detection. Secondly, GOSII is designed by dynamically adjusting the weights of each feature channel through combining SRU blocks and SimAM attention mechanism, which are further optimized and embedded into C2f to enhance the representation ability of contextual features. GOSII captures crucial features from complex backgrounds and improves information transmission. Finally, ESFCF integrates the edge and semantic information within shallow feature maps to address the issue of inaccurate localization for small objects, and further improves object detection accuracy by compensating for the loss of edge details in feature extraction. Extensive experiments on three commonly used remote sensing datasets of NWPU VHR-10, VisDrone 2019, and DOTA demonstrate the superiority of our method in object classification and localization compared to other state-of-the-art methods.

Hyperspectral images (HSIs) encapsulate a vast amount of information due to their expansive size and high number of channel dimensions. However, they are insufficiently utilized for ineffective feature extraction, particularly for regions with few samples and predominant edges. To fully leverage the spatial–spectral features of HSIs, a dual-branch multi-scale spatial–spectral residual attention network (MSRAN) that integrates multi-scale feature extraction with residual attention mechanisms is proposed. MSRAN independently extracts spatial and spectral features through dual branches, minimizing the interference between these features and enhancing the focus on feature extraction in different dimensions. Specifically, in the spectral feature extraction branch, diverse-scale 3D convolution kernels capture extended spectral sequence characteristics and neighborhood spectral features. The convolution fusion emphasizes the weight of the central pixel to be classified, followed by the use of spectral residual attention mechanisms to extract enhanced central-pixel spectral features. In the spatial feature extraction branch, multi-level receptive fields are utilized to extract various fine-grained spatial contours, edges, and local detailed features, which are further processed through spatial residual attention to effectively extract spatial composite features. Finally, the convolution fusion module adaptively integrates the center-enhanced spectral features with multi-level fine-grained spatial features for classification. Extensive comparative experiments and ablation studies demonstrate that MSRAN achieves highly competitive results on two classic datasets from Pavia University and Salinas as well as on a novel dataset of WHU-Hi-LongKou.

Hyperspectral image (HSI) possesses distinctive advantages in the classification of materials due to its rich spectral information. Convolutional neural network (CNN) and vision transformer (ViT), as mainstream methodologies, have demonstrated significant success. However, they often ignore subtle spectral differences, leading to the inadequate utilization of the spectral information. In this letter, we present a novel feature extraction and classification method, i.e., dilated deep MPFormer network (DDMN), which takes the inherent advantages of CNN and ViT while enhancing the exploitation of spectral information. First, the dilated depthwise separable convolution (DDSC) is proposed to expand the channel dimension, enabling the capture of subtle spectral differences among similar materials. Second, a sequence of improved transformers, i.e., MPFormer, are adopted to effectively extract spatial-spectral features, in which a new multiscale pooling mixer (MPMixer) is designed to replace the attention module in ViT, resulting in reduced parameter numbers and accelerated training speed. Finally, an adaptive weighted fusion module (AWFM) is developed to improve the interaction between specific texture features in shallow layers and abstract semantic features in deep layers. Extensive experiments demonstrate that the proposed DDMN method achieves improvements in OA of 0.8%, 1.04%, and 1.99% when compared to SOTA methods on three HSI datasets of LK, PU, and HS, respectively.