Jinlong Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jinlong Zhang has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Jiahang Liu
College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
3 shared publications - Xinpeng Ling
College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2 shared publications - Zexian Duan
College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2 shared publications - Zhiheng Liu
College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
2 shared publications - Le Wang
Xi’an Jiaotong University
1 shared publication - Sanping Zhou
Xi’an Jiaotong University
1 shared publication
Latest Publications
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-15
Adaptive Receptive Field Network for Small Objects Detection in Remote Sensing Images
Small object detection in remote sensing images is severely hampered by the significant scale variation even among small objects. Conventional methods often rely on a static receptive field to incorporate context, leading to a critical scale mismatch: oversized fields submerge tiny objects in background noise, while undersized fields fail to capture sufficient context for larger ones. To resolve this dilemma, we propose the adaptive receptive field network based on the mixture of experts (ARFN-MoEs), a novel architecture that dynamically tailors receptive fields to specific object scales. Our framework is built upon a set of scale-specialized experts, each expert module with a distinct receptive field optimized for a particular scale range. To orchestrate these experts, we introduce a grid-level router that intelligently dispatches local feature regions to the most appropriate expert, ensuring a precisely scale-matched contextual feature extraction. Furthermore, a grid-level balance loss is designed to encourage the uniform utilization of all experts during training, preventing model degradation. Extensive experiments on multiple public benchmark datasets demonstrate the effectiveness and superiority of our proposed method for small object detection in remote sensing images. The source code will be made available at: https://github.com/GDragon986/ARFN-MoE
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-15
Adaptive Receptive Field Network for Small Objects Detection in Remote Sensing Images
Small object detection in remote sensing images is severely hampered by the significant scale variation even among small objects. Conventional methods often rely on a static receptive field to incorporate context, leading to a critical scale mismatch: oversized fields submerge tiny objects in background noise, while undersized fields fail to capture sufficient context for larger ones. To resolve this dilemma, we propose the adaptive receptive field network based on the mixture of experts (ARFN-MoEs), a novel architecture that dynamically tailors receptive fields to specific object scales. Our framework is built upon a set of scale-specialized experts, each expert module with a distinct receptive field optimized for a particular scale range. To orchestrate these experts, we introduce a grid-level router that intelligently dispatches local feature regions to the most appropriate expert, ensuring a precisely scale-matched contextual feature extraction. Furthermore, a grid-level balance loss is designed to encourage the uniform utilization of all experts during training, preventing model degradation. Extensive experiments on multiple public benchmark datasets demonstrate the effectiveness and superiority of our proposed method for small object detection in remote sensing images. The source code will be made available at: https://github.com/GDragon986/ARFN-MoE
- Open Access
- 2 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 15721-15734
Small-Object Detection in Remote Sensing Images With Super-Resolution Perception
Small objects are widely distributed on remote sensing images (RSIs), and most of them are achieved by super resolution (SR) reconstruction followed by detection. However, due to the independent training of the SR network and the detection network, the lack of interaction between them leads to the limited performance of small object detection (SOD). Furthermore, time accountability is increased since the SR task is performed before the detection task. To address these problems, we develop a new SOD network to improve the SOD performance in RSIs, which embeds the SR task into the SOD task. First, a channel attention weighting module is proposed before the backbone to assign weights to different channels of the input image, allowing the network to selectively focus on different channels. Secondly, a self-attention encoding module is designed between the backbone and neck to add self-attention weight to the features extracted from the backbone and enhance the feature representation ability of small objects. Finally, the SR perceptual branch and perceptual loss are designed so that the SR task and the detection task can be associated through the SR perceptual loss, and the SR perceptual branch can guide the backbone network to learn high-resolution features through joint training, thus improving the detection performance of small objects. In the inference phase, the SR perceptual branch has been removed to improve the speed. Extensive experimental results on VEDAI and DOTA datasets show that the proposed method achieves an accuracy of 82.63% and 78.63%.
- 2 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 15721-15734
Open AccessSmall-Object Detection in Remote Sensing Images With Super-Resolution Perception
Small objects are widely distributed on remote sensing images (RSIs), and most of them are achieved by super resolution (SR) reconstruction followed by detection. However, due to the independent training of the SR network and the detection network, the lack of interaction between them leads to the limited performance of small object detection (SOD). Furthermore, time accountability is increased since the SR task is performed before the detection task. To address these problems, we develop a new SOD network to improve the SOD performance in RSIs, which embeds the SR task into the SOD task. First, a channel attention weighting module is proposed before the backbone to assign weights to different channels of the input image, allowing the network to selectively focus on different channels. Secondly, a self-attention encoding module is designed between the backbone and neck to add self-attention weight to the features extracted from the backbone and enhance the feature representation ability of small objects. Finally, the SR perceptual branch and perceptual loss are designed so that the SR task and the detection task can be associated through the SR perceptual loss, and the SR perceptual branch can guide the backbone network to learn high-resolution features through joint training, thus improving the detection performance of small objects. In the inference phase, the SR perceptual branch has been removed to improve the speed. Extensive experimental results on VEDAI and DOTA datasets show that the proposed method achieves an accuracy of 82.63% and 78.63%.
- Open Access
- 14 May 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (10) , 1741
ERNet: A Rapid Road Crack Detection Method Using Low-Altitude UAV Remote Sensing Images
The rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently, this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can provide images with a centimeter-level or even subcentimeter-level ground resolution, which provides a new, efficient, and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we proposed a real-time edge reconstruction crack detection network (ERNet) that adopted multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and the background. To capture global dependencies across spatial and channel levels, we proposed an efficient bilateral decomposed convolutional attention module (BDAM) that combined depth-separable convolution and dilated convolution to capture global dependencies across the spatial and channel levels. To enhance the accuracy of crack detection, we used a coordinate-based fusion module that integrated spatial, semantic, and edge reconstruction information. In addition, we proposed an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. The experimental results demonstrated that our network achieved the best balance between accuracy and inference speed compared to six established models.
- 14 May 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (10) , 1741
Open AccessERNet: A Rapid Road Crack Detection Method Using Low-Altitude UAV Remote Sensing Images
The rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently, this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can provide images with a centimeter-level or even subcentimeter-level ground resolution, which provides a new, efficient, and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we proposed a real-time edge reconstruction crack detection network (ERNet) that adopted multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and the background. To capture global dependencies across spatial and channel levels, we proposed an efficient bilateral decomposed convolutional attention module (BDAM) that combined depth-separable convolution and dilated convolution to capture global dependencies across the spatial and channel levels. To enhance the accuracy of crack detection, we used a coordinate-based fusion module that integrated spatial, semantic, and edge reconstruction information. In addition, we proposed an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. The experimental results demonstrated that our network achieved the best balance between accuracy and inference speed compared to six established models.
- Open Access
- 2 April 2024
- Published by MDPI AG in Preprints.org
ERNet: A Rapid Road Cracks Detection Method from Low-Altitude UAV Remote Sensing Image
Rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can obtain images with centimeter or even subcentimeter ground resolution, which provides a new efficient and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we propose a real-time edge reconstruction crack detection network (ERNet) which adopts multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and background. To capture global dependencies across spatial and channel levels, we propose an efficient bilateral decomposed convolutional attention module (BDAM) that combines depth separable convolution and dilated convolution to capture global dependencies across spatial and channel levels. To enhance the accuracy of crack detection, we use a coordinate-based fusion module that integrates spatial, semantic, and edge reconstruction information. In addition, we propose an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. Experimental results demonstrate that our network achieves the best balance between accuracy and inference speed.
- 2 April 2024
- Published by MDPI AG in Preprints.org
Open AccessERNet: A Rapid Road Cracks Detection Method from Low-Altitude UAV Remote Sensing Image
Rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can obtain images with centimeter or even subcentimeter ground resolution, which provides a new efficient and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we propose a real-time edge reconstruction crack detection network (ERNet) which adopts multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and background. To capture global dependencies across spatial and channel levels, we propose an efficient bilateral decomposed convolutional attention module (BDAM) that combines depth separable convolution and dilated convolution to capture global dependencies across spatial and channel levels. To enhance the accuracy of crack detection, we use a coordinate-based fusion module that integrates spatial, semantic, and edge reconstruction information. In addition, we propose an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. Experimental results demonstrate that our network achieves the best balance between accuracy and inference speed.