In recent years, the amount of remote sensing imagery data has increased exponentially. The ability to quickly and effectively find the required images from massive remote sensing archives is the key to the organization, management, and sharing of remote sensing image information. This paper proposes a high-resolution remote sensing image retrieval method with Gabor-CA-ResNet and a split-based deep feature transform network. The main contributions include two points. (1) For the complex texture, diverse scales, and special viewing angles of remote sensing images, A Gabor-CA-ResNet network taking ResNet as the backbone network is proposed by using Gabor to represent the spatial-frequency structure of images, channel attention (CA) mechanism to obtain stronger representative and discriminative deep features. (2) A split-based deep feature transform network is designed to divide the features extracted by the Gabor-CA-ResNet network into several segments and transform them separately for reducing the dimensionality and the storage space of deep features significantly. The experimental results on UCM, WHU-RS, RSSCN7, and AID datasets show that, compared with the state-of-the-art methods, our method can obtain competitive performance, especially for remote sensing images with rare targets and complex textures.

Recently, there have been rapid advances in high-resolution remote sensing image retrieval, which plays an important role in remote sensing data management and utilization. For content-based remote sensing image retrieval, low-dimensional, representative and discriminative features are essential to ensure good retrieval accuracy and speed. Dimensionality reduction is one of the important solutions to improve the quality of features in image retrieval, in which LargeVis is an effective algorithm specifically designed for Big Data visualization. Here, an extended LargeVis (E-LargeVis) dimensionality reduction method for high-resolution remote sensing image retrieval is proposed. This can realize the dimensionality reduction of single high-dimensional data by modeling the implicit mapping relationship between LargeVis high-dimensional data and low-dimensional data with support vector regression. An effective high-resolution remote sensing image retrieval method is proposed to obtain stronger representative and discriminative deep features. First, the fully connected layer features are extracted using a channel attention-based ResNet50 as a backbone network. Then, E-LargeVis is used to reduce the dimensionality of the fully connected features to obtain a low-dimensional discriminative representation. Finally, L2 distance is computed for similarity measurement to realize the retrieval of high-resolution remote sensing images. The experimental results on four high-resolution remote sensing image datasets, including UCM, RS19, RSSCN7, and AID, show that for various convolutional neural network architectures, the proposed E-LargeVis can effectively improve retrieval performance, far exceeding other dimensionality reduction methods.

Many high-performance networks are not designed with lightweight application scenarios in mind from the outset, which has greatly restricted their scope of application. This paper focuses on ConvNeXt as the research subject. It incorporates the Cross Stage Partial Connections mechanism and a series of optimizations. These improvements significantly reduce the parameter scale and network complexity. The new network is named E-ConvNeXt, which can maintain high accuracy performance under different complexity configurations. The three core contributions of E-ConvNeXt are : (1) Integrating the Cross Stage Partial Network (CSPNet) with ConvNeXt and adjusting the network structure, which reduces the model's network complexity by up to 80\%; (2) Optimizing the Stem and Block structures to enhance the model's feature expression capability and operational efficiency; (3) Replacing Layer Scale with channel attention. Experimental validation on ImageNet classification demonstrates E-ConvNeXt's superior accuracy-efficiency balance. Transfer learning tests on object detection tasks further confirm its generalization capability.

To design efficient neural networks, many studies focus on reducing redundancy. However, such feature discarding is often random and may remove useful information, which leads to degraded network performance. To address the challenge of insufficient feature extraction under limited computational budgets, we decouple feature modeling into channel-wise and spatial dimensions, which substantially reduces computational overhead. To enable efficient spatial modeling, we propose Learnable Wavelet Spatial Attention (LWSA), which leverages the wavelet transform to decompose the feature map into multiple subbands of varying informativeness and performs spatial attention selectively on the most informative components, enabling effective feature extraction at low computational cost. By cascading the ESE module and LWSA, we propose a Cascaded Channel–Spatial Attention mechanism for feature extraction. Building upon this, we further introduce WCANet, a lightweight architecture that achieves effective feature modeling under low computational budgets across a variety of vision tasks. For example, on the ImageNet-1K, WCANet significantly outperforms FasterNet in accuracy while achieving 40% higher throughput. WCANet-T4 runs 2.9× faster than StarNet-S1 with 1.4% higher accuracy, and also surpasses MobileMamba-T2 by 1.3% in Top-1 accuracy while maintaining high efficiency.