Jin Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jin Zhou has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Dan Jia
Institute of Optics Electronics, Chinese Academy of Sciences, Chengdu, China
3 shared publications - Nian Pan
National Laboratory on Adaptive Optics, Chengdu 610209, China
2 shared publications - Dengyu Yin
AVIC Chengdu Aircraft Design & Research Institute, Chengdu 610091, China
1 shared publication - Di Wang
AVIC Chengdu Aircraft Design & Research Institute, Chengdu 610091, China
1 shared publication - Haiying Wu
Huayin Ordance Test Center, Huayin 714200, China
1 shared publication Hechen Zhang
Institute of Optics Electronics, Chinese Academy of Sciences, Chengdu, China
1 shared publication
Latest Publications
- Open Access
- 28 January 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (3) , 423
HD-BSNet: A Plug-and-Play Dual-Mechanism Synergistic Enhancement Framework for Small Object Detection
In remote sensing and low-altitude unmanned aerial vehicle(UAV) detection scenarios, small target detection is extremely challenging due to the low pixel proportion, sparse features, and complex backgrounds of targets. The reliability of low-altitude security, in particular, is directly dependent on the accuracy of small target detection. However, current methods still face three major limitations: insufficient detection accuracy for targets smaller than 20 pixels; artifacts and false textures introduced by Generative Adversarial Network-based enhancement, which lead to increased false detection rates; and the reliance of existing approaches on specialized architectures, resulting in weak generalization capability and difficulty in adapting to multi-scenario deployment requirements. To address these issues, this paper proposes a plug-and-play dual-mechanism collaborative enhancement framework named HD-BSNet. Firstly, a High-Frequency Differential Perception mechanism is designed to enhance the detailed feature representation of small targets. Secondly, a Background Semantic Modeling mechanism is introduced to learn key features that distinguish targets from the background. Additionally, a Parallel Multi-Scale Focus Module is constructed to further reinforce target features. Extensive experiments on three small target datasets demonstrate that the proposed method effectively improves the accuracy and generalization ability of small target detection.
- 28 January 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (3) , 423
Open AccessHD-BSNet: A Plug-and-Play Dual-Mechanism Synergistic Enhancement Framework for Small Object Detection
In remote sensing and low-altitude unmanned aerial vehicle(UAV) detection scenarios, small target detection is extremely challenging due to the low pixel proportion, sparse features, and complex backgrounds of targets. The reliability of low-altitude security, in particular, is directly dependent on the accuracy of small target detection. However, current methods still face three major limitations: insufficient detection accuracy for targets smaller than 20 pixels; artifacts and false textures introduced by Generative Adversarial Network-based enhancement, which lead to increased false detection rates; and the reliance of existing approaches on specialized architectures, resulting in weak generalization capability and difficulty in adapting to multi-scenario deployment requirements. To address these issues, this paper proposes a plug-and-play dual-mechanism collaborative enhancement framework named HD-BSNet. Firstly, a High-Frequency Differential Perception mechanism is designed to enhance the detailed feature representation of small targets. Secondly, a Background Semantic Modeling mechanism is introduced to learn key features that distinguish targets from the background. Additionally, a Parallel Multi-Scale Focus Module is constructed to further reinforce target features. Extensive experiments on three small target datasets demonstrate that the proposed method effectively improves the accuracy and generalization ability of small target detection.
- Open Access
- 18 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2502
MEFA-Net: Multilevel Feature Extraction and Fusion Attention Network for Infrared Small-Target Detection
Infrared small-target detection encounters significant challenges due to a low image signal-to-noise ratio, limited target size, and complex background noise. To address the issues of sparse feature loss for small targets during the down-sampling phase of the traditional U-Net network and the semantic gap in the feature fusion process, a multilevel feature extraction and fusion attention network (MEFA-Net) is designed. Specifically, the dilated direction-sensitive convolution block (DDCB) is devised to collaboratively extract local detail features, contextual features, and Gaussian salient features via ordinary convolution, dilated convolution and parallel strip convolution. Furthermore, the encoder attention fusion module (EAF) is employed, where spatial and channel attention weights are generated using dual-path pooling to achieve the adaptive fusion of deep and shallow layer features. Lastly, an efficient up-sampling block (EUB) is constructed, integrating a hybrid up-sampling strategy with multi-scale dilated convolution to refine the localization of small targets. The experimental results confirm that the proposed algorithm model surpasses most existing recent methods. Compared with the baseline, the intersection over union (IoU) and probability of detection of MEFA-Net on the IRSTD-1k dataset are increased by 2.25% and 3.05%, respectively, achieving better detection performance and a lower false alarm rate in complex scenarios.- 18 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2502
Open AccessMEFA-Net: Multilevel Feature Extraction and Fusion Attention Network for Infrared Small-Target Detection
Infrared small-target detection encounters significant challenges due to a low image signal-to-noise ratio, limited target size, and complex background noise. To address the issues of sparse feature loss for small targets during the down-sampling phase of the traditional U-Net network and the semantic gap in the feature fusion process, a multilevel feature extraction and fusion attention network (MEFA-Net) is designed. Specifically, the dilated direction-sensitive convolution block (DDCB) is devised to collaboratively extract local detail features, contextual features, and Gaussian salient features via ordinary convolution, dilated convolution and parallel strip convolution. Furthermore, the encoder attention fusion module (EAF) is employed, where spatial and channel attention weights are generated using dual-path pooling to achieve the adaptive fusion of deep and shallow layer features. Lastly, an efficient up-sampling block (EUB) is constructed, integrating a hybrid up-sampling strategy with multi-scale dilated convolution to refine the localization of small targets. The experimental results confirm that the proposed algorithm model surpasses most existing recent methods. Compared with the baseline, the intersection over union (IoU) and probability of detection of MEFA-Net on the IRSTD-1k dataset are increased by 2.25% and 3.05%, respectively, achieving better detection performance and a lower false alarm rate in complex scenarios. - Open Access
- 8 November 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (22) , 4160
SFFNet: Shallow Feature Fusion Network Based on Detection Framework for Infrared Small Target Detection
Infrared small target detection (IRSTD) is the process of recognizing and distinguishing small targets from infrared images that are obstructed by crowded backgrounds. This technique is used in various areas, including ground monitoring, flight navigation, and so on. However, due to complex backgrounds and the loss of information in deep networks, infrared small target detection remains a difficult undertaking. To solve the above problems, we present a shallow feature fusion network (SFFNet) based on detection framework. Specifically, we design the shallow-layer-guided feature enhancement (SLGFE) module, which guides multi-scale feature fusion with shallow layer information, effectively mitigating the loss of information in deep networks. Then, we design the visual-Mamba-based global information extension (VMamba-GIE) module, which leverages a multi-branch structure combining the capability of convolutional layers to extract features in local space with the advantages of state space models in the exploration of long-distance information. The design significantly extends the network’s capacity to acquire global contextual information, enhancing its capability to handle complex backgrounds. And through the effective fusion of the SLGFE and VMamba-GIE modules, the exorbitant computation brought by the SLGFE module is substantially reduced. The experimental results on two publicly available infrared small target datasets demonstrate that the SFFNet surpasses other state-of-the-art algorithms.
- 8 November 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (22) , 4160
Open AccessSFFNet: Shallow Feature Fusion Network Based on Detection Framework for Infrared Small Target Detection
Infrared small target detection (IRSTD) is the process of recognizing and distinguishing small targets from infrared images that are obstructed by crowded backgrounds. This technique is used in various areas, including ground monitoring, flight navigation, and so on. However, due to complex backgrounds and the loss of information in deep networks, infrared small target detection remains a difficult undertaking. To solve the above problems, we present a shallow feature fusion network (SFFNet) based on detection framework. Specifically, we design the shallow-layer-guided feature enhancement (SLGFE) module, which guides multi-scale feature fusion with shallow layer information, effectively mitigating the loss of information in deep networks. Then, we design the visual-Mamba-based global information extension (VMamba-GIE) module, which leverages a multi-branch structure combining the capability of convolutional layers to extract features in local space with the advantages of state space models in the exploration of long-distance information. The design significantly extends the network’s capacity to acquire global contextual information, enhancing its capability to handle complex backgrounds. And through the effective fusion of the SLGFE and VMamba-GIE modules, the exorbitant computation brought by the SLGFE module is substantially reduced. The experimental results on two publicly available infrared small target datasets demonstrate that the SFFNet surpasses other state-of-the-art algorithms.
- Open Access
- 18 September 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-11
Two-plus-two fringe projection profilometry based on phase-shifted coding
In fringe projection profilometry based on temporal phase unwrapping, determining a fringe order map commonly requires a large number of fringes. To reduce the fringe number, this paper proposes a concise absolute phase retrieval algorithm just by projecting four fringes. The first two orthogonal fringes with relatively large frequency can collect reliable height information. The second two fringes are designed the same as the first two, but the only difference is that each 2π-phase of them is shifted by a unique amount, which can robustly label a large number of fringe orders. For decoding the fringes, we develop an average intensity one-time extraction algorithm, which allows for the rapid acquisition of the two pairs of alternating current components. From this, the wrapped phase containing height information and the stair-coded phase providing fringe orders can be directly extracted by arctangent operation in a point-to-point manner. Furthermore, we also develop a universal fringe order correction algorithm that can simultaneously correct the common errors and the misalignment between the wrapped phase and fringe orders. Experiment results demonstrate that this method achieves comparable accuracy and adaptability to the phase-coding method, while utilizing two fewer fringes.
- 18 September 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-11
Open AccessTwo-plus-two fringe projection profilometry based on phase-shifted coding
In fringe projection profilometry based on temporal phase unwrapping, determining a fringe order map commonly requires a large number of fringes. To reduce the fringe number, this paper proposes a concise absolute phase retrieval algorithm just by projecting four fringes. The first two orthogonal fringes with relatively large frequency can collect reliable height information. The second two fringes are designed the same as the first two, but the only difference is that each 2π-phase of them is shifted by a unique amount, which can robustly label a large number of fringe orders. For decoding the fringes, we develop an average intensity one-time extraction algorithm, which allows for the rapid acquisition of the two pairs of alternating current components. From this, the wrapped phase containing height information and the stair-coded phase providing fringe orders can be directly extracted by arctangent operation in a point-to-point manner. Furthermore, we also develop a universal fringe order correction algorithm that can simultaneously correct the common errors and the misalignment between the wrapped phase and fringe orders. Experiment results demonstrate that this method achieves comparable accuracy and adaptability to the phase-coding method, while utilizing two fewer fringes.
- Open Access
Three-Shot Dual-Frequency Fringe Scheme Based on Spatial Computer-Generated Moiré Fringe
A highly robust dual-frequency hierarchical temporal phase unwrapping (DHTPU) based on the novel spatial computer-generated Moiré profilometry (SCGMP) is proposed. The method requires only three patterns: a high-frequency fringe to provide robust surface information, a multi-period low-frequency fringe to eliminate the 2π-phase ambiguities, and a flat pattern to remove the average intensity of the two fringes. In decoding, different from traditional Moiré profilometries that rely on spectrum filters, SCGMP only employs spatial-domain calculations to extract the wrapped phase, thereby preserving more detailed information. Furthermore, we fully explore SCGMP’s capability to significantly alleviate phase ambiguity and provide an algorithm to determine the maximum measurable height range for a fixed system, enabling the direct extraction of the continuous basic phase from the multi-period low-frequency fringe. Consequently, the proposed basic phase exhibits an enhanced signal-to-noise ratio, compared to the traditional basic phase derived from the single-period fringes, effectively releasing the high-frequency restriction in the traditional DHTPU. The experimental results verify that the proposed DHTPU method has considerable accuracy and great potential for high-speed measurements, due to there being only three shots required.
Open AccessThree-Shot Dual-Frequency Fringe Scheme Based on Spatial Computer-Generated Moiré Fringe
A highly robust dual-frequency hierarchical temporal phase unwrapping (DHTPU) based on the novel spatial computer-generated Moiré profilometry (SCGMP) is proposed. The method requires only three patterns: a high-frequency fringe to provide robust surface information, a multi-period low-frequency fringe to eliminate the 2π-phase ambiguities, and a flat pattern to remove the average intensity of the two fringes. In decoding, different from traditional Moiré profilometries that rely on spectrum filters, SCGMP only employs spatial-domain calculations to extract the wrapped phase, thereby preserving more detailed information. Furthermore, we fully explore SCGMP’s capability to significantly alleviate phase ambiguity and provide an algorithm to determine the maximum measurable height range for a fixed system, enabling the direct extraction of the continuous basic phase from the multi-period low-frequency fringe. Consequently, the proposed basic phase exhibits an enhanced signal-to-noise ratio, compared to the traditional basic phase derived from the single-period fringes, effectively releasing the high-frequency restriction in the traditional DHTPU. The experimental results verify that the proposed DHTPU method has considerable accuracy and great potential for high-speed measurements, due to there being only three shots required.