Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (881)

Search Parameters:
Keywords = space target images

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
18 pages, 2839 KiB  
Article
Detection of Maize Pathogenic Fungal Spores Based on Deep Learning
by Yijie Ren, Ying Xu, Huilin Tian, Qian Zhang, Mingxiu Yang, Rongsheng Zhu, Dawei Xin, Qingshan Chen, Qiaorong Wei and Shuang Song
Agriculture 2025, 15(15), 1689; https://doi.org/10.3390/agriculture15151689 - 5 Aug 2025
Abstract
Timely detection of pathogen spores is fundamental to ensuring early intervention and reducing the spread of corn diseases, like northern corn leaf blight, corn head smut, and corn rust. Traditional spore detection methods struggle to identify spore-level targets within complex backgrounds. To improve [...] Read more.
Timely detection of pathogen spores is fundamental to ensuring early intervention and reducing the spread of corn diseases, like northern corn leaf blight, corn head smut, and corn rust. Traditional spore detection methods struggle to identify spore-level targets within complex backgrounds. To improve the recognition accuracy of various maize disease spores, this study introduced the YOLOv8s-SPM model by incorporating the space-to-depth and convolution (SPD-Conv) layers, the Partial Self-Attention (PSA) mechanism, and Minimum Point Distance Intersection over Union (MPDIoU) loss function. First, we combined SPD-Conv layers into the Backbone of the YOLOv8s to enhance recognition performance on small targets and low-resolution images. To improve computational efficiency, the PSA mechanism was incorporated within the Neck layer of the network. Finally, MPDIoU loss function was applied to refine the localization performance of bounding boxes. The results revealed that the YOLOv8s-SPM model achieved 98.9% accuracy on the mixed spore dataset. Relative to the baseline YOLOv8s, the YOLOv8s-SPM model yielded a 1.4% gain in accuracy. The improved model significantly improved spore detection accuracy and demonstrated superior performance in recognizing diverse spore types under complex background conditions. It met the demands for high-precision spore detection and filled a gap in intelligent spore recognition for maize, offering an effective starting point and practical path for future research in this field. Full article
(This article belongs to the Section Crop Protection, Diseases, Pests and Weeds)
Show Figures

Figure 1

18 pages, 5280 KiB  
Article
A Drilling Debris Tracking and Velocity Measurement Method Based on Fine Target Feature Fusion Optimization
by Jinteng Yang, Yu Bao, Zumao Xie, Haojie Zhang, Zhongnian Li and Yonggang Li
Appl. Sci. 2025, 15(15), 8662; https://doi.org/10.3390/app15158662 (registering DOI) - 5 Aug 2025
Abstract
During unmanned drilling operations, the velocity of drill cuttings serves as an important indicator of drilling conditions, which necessitates real-time and accurate measurements. To address challenges such as the small size of cuttings, weak feature representations, and complex motion trajectories, we propose a [...] Read more.
During unmanned drilling operations, the velocity of drill cuttings serves as an important indicator of drilling conditions, which necessitates real-time and accurate measurements. To address challenges such as the small size of cuttings, weak feature representations, and complex motion trajectories, we propose a novel velocity measurement method integrating small-object detection and tracking. Specifically, we enhance the multi-scale feature fusion capability of the YOLOv11 detection head by incorporating a lightweight feature extraction module, Ghost Conv, and a feature-aligned fusion module, FA-Concat, resulting in an improved model named YOLOv11-Dd (drilling debris). Furthermore, considering the robustness of the ByteTrack algorithm in retaining low-confidence targets and handling occlusions, we integrate ByteTrack into the tracking phase to enhance tracking stability. A velocity estimation module is introduced to achieve high-precision measurement by mapping the pixel displacement of detection box centers across consecutive frames to physical space. To facilitate model training and performance evaluation, we establish a drill-cutting splash simulation dataset comprising 3787 images, covering a diverse range of ejection angles, velocities, and material types. The experimental results show that the YOLOv11-Dd model achieves a 4.65% improvement in mAP@80 over YOLOv11, reaching 76.04%. For mAP@75–95, it improves by 0.79%, reaching 41.73%. The proposed velocity estimation method achieves an average accuracy of 92.12% in speed measurement tasks, representing a 0.42% improvement compared to the original YOLOv11. Full article
(This article belongs to the Special Issue AI from Industry 4.0 to Industry 5.0: Engineering for Social Change)
Show Figures

Figure 1

23 pages, 23638 KiB  
Article
Enhanced YOLO and Scanning Portal System for Vehicle Component Detection
by Feng Ye, Mingzhe Yuan, Chen Luo, Shuo Li, Duotao Pan, Wenhong Wang, Feidao Cao and Diwen Chen
Sensors 2025, 25(15), 4809; https://doi.org/10.3390/s25154809 - 5 Aug 2025
Abstract
In this paper, a novel online detection system is designed to enhance accuracy and operational efficiency in the outbound logistics of automotive components after production. The system consists of a scanning portal system and an improved YOLOv12-based detection algorithm which captures images of [...] Read more.
In this paper, a novel online detection system is designed to enhance accuracy and operational efficiency in the outbound logistics of automotive components after production. The system consists of a scanning portal system and an improved YOLOv12-based detection algorithm which captures images of automotive parts passing through the scanning portal in real time. By integrating deep learning, the system enables real-time monitoring and identification, thereby preventing misdetections and missed detections of automotive parts, in this way promoting intelligent automotive part recognition and detection. Our system introduces the A2C2f-SA module, which achieves an efficient feature attention mechanism while maintaining a lightweight design. Additionally, Dynamic Space-to-Depth (Dynamic S2D) is employed to improve convolution and replace the stride convolution and pooling layers in the baseline network, helping to mitigate the loss of fine-grained information and enhancing the network’s feature extraction capability. To improve real-time performance, a GFL-MBConv lightweight detection head is proposed. Furthermore, adaptive frequency-aware feature fusion (Adpfreqfusion) is hybridized at the end of the neck network to effectively enhance high-frequency information lost during downsampling, thereby improving the model’s detection accuracy for target objects in complex backgrounds. On-site tests demonstrate that the system achieves a comprehensive accuracy of 97.3% and an average vehicle detection time of 7.59 s, exhibiting not only high precision but also high detection efficiency. These results can make the proposed system highly valuable for applications in the automotive industry. Full article
(This article belongs to the Topic Smart Production in Terms of Industry 4.0 and 5.0)
Show Figures

Figure 1

16 pages, 1618 KiB  
Article
Multimodal Temporal Knowledge Graph Embedding Method Based on Mixture of Experts for Recommendation
by Bingchen Liu, Guangyuan Dong, Zihao Li, Yuanyuan Fang, Jingchen Li, Wenqi Sun, Bohan Zhang, Changzhi Li and Xin Li
Mathematics 2025, 13(15), 2496; https://doi.org/10.3390/math13152496 - 3 Aug 2025
Viewed by 148
Abstract
Knowledge-graph-based recommendation aims to provide personalized recommendation services to users based on their historical interaction information, which is of great significance for shopping transaction rates and other aspects. With the rapid growth of online shopping, the knowledge graph constructed from users’ historical interaction [...] Read more.
Knowledge-graph-based recommendation aims to provide personalized recommendation services to users based on their historical interaction information, which is of great significance for shopping transaction rates and other aspects. With the rapid growth of online shopping, the knowledge graph constructed from users’ historical interaction data now incorporates multiattribute information, including timestamps, images, and textual content. The information of multiple modalities is difficult to effectively utilize due to their different representation structures and spaces. The existing methods attempt to utilize the above information through simple embedding representation and aggregation, but ignore targeted representation learning for information with different attributes and learning effective weights for aggregation. In addition, existing methods are not sufficient for effectively modeling temporal information. In this article, we propose MTR, a knowledge graph recommendation framework based on mixture of experts network. To achieve this goal, we use a mixture-of-experts network to learn targeted representations and weights of different product attributes for effective modeling and utilization. In addition, we effectively model the temporal information during the user shopping process. A thorough experimental study on popular benchmarks validates that MTR can achieve competitive results. Full article
(This article belongs to the Special Issue Data-Driven Decentralized Learning for Future Communication Networks)
Show Figures

Figure 1

17 pages, 4148 KiB  
Article
Contribution of the Gravity Component and Surface Type During the Initial Stages of Biofilm Formation at Solid–Liquid Interfaces
by Elisavet Malea, Maria Petala, Margaritis Kostoglou and Theodoros Karapantsios
Water 2025, 17(15), 2277; https://doi.org/10.3390/w17152277 - 31 Jul 2025
Viewed by 255
Abstract
Water systems are highly vulnerable to biofilm formation, which can compromise water quality, operational efficiency, and public health. Factors such as surface material properties and gravitational orientation of the surface play critical roles in the early stages of microbial attachment and biofilm development. [...] Read more.
Water systems are highly vulnerable to biofilm formation, which can compromise water quality, operational efficiency, and public health. Factors such as surface material properties and gravitational orientation of the surface play critical roles in the early stages of microbial attachment and biofilm development. This study examines the impact of gravity and surface composition on the initial adhesion of Pseudomonas fluorescens AR11—a model organism for biofilm research. Focusing on stainless steel (SS) and polycarbonate (PC), two materials commonly used in water and wastewater infrastructure, bacterial adhesion was evaluated at surface inclinations of 0°, 45°, 90°, and 180° to assess gravitational impact. After three hours of contact, fluorescence microscopy and image analysis were used to quantify surface coverage and cluster size distribution. The results showed that both material type and orientation significantly affected early biofilm formation. PC surfaces consistently exhibited higher bacterial adhesion at all angles, with modest variations, suggesting that material properties are a dominant factor in initial colonization. In contrast, SS showed angle-dependent variation, indicating a combined effect of gravitational convection and surface characteristics. These insights contribute to a deeper understanding of biofilm dynamics under realistic environmental conditions, including those encountered in space systems, and support the development of targeted strategies for biofilm control in water systems and spaceflight-related infrastructure. Full article
Show Figures

Figure 1

13 pages, 9867 KiB  
Article
Recurrence Patterns After Resection of Sacral Chordoma: Toward an Optimized Postoperative Target Volume Definition
by Hanna Waldsperger, Burkhard Lehner, Andreas Geisbuesch, Felix Jotzo, Eva Meixner, Laila König, Sebastian Regnery, Katharina Kozyra, Lars Wessel, Sandro Krieg, Klaus Herfarth, Jürgen Debus and Katharina Seidensaal
Cancers 2025, 17(15), 2521; https://doi.org/10.3390/cancers17152521 - 30 Jul 2025
Viewed by 119
Abstract
Background: Postoperative recurrence of sacrococcygeal chordomas presents significant clinical challenges due to unusual recurrence patterns. This study aimed to characterize these patterns of recurrence to inform improved adjuvant radiotherapy planning. Methods: We retrospectively analyzed 31 patients with recurrent sacrococcygeal chordoma following surgery, assessing [...] Read more.
Background: Postoperative recurrence of sacrococcygeal chordomas presents significant clinical challenges due to unusual recurrence patterns. This study aimed to characterize these patterns of recurrence to inform improved adjuvant radiotherapy planning. Methods: We retrospectively analyzed 31 patients with recurrent sacrococcygeal chordoma following surgery, assessing recurrence locations considering initial tumor extent, resection levels, and postoperative anatomical changes on MRI. In 18 patients, pre- and postoperative imaging enabled the spatial mapping of early recurrence origins relative to the initial tumor volume using isotropic expansions. The median initial gross tumor volume was 113 mL. Results: Recurrences were mostly multifocal and predominantly involved soft tissues (e.g., mesorectal/perirectal space (80.6%), piriformis and gluteal muscles (80.6% and 67.7%, respectively) and osseous structures, particularly the sacrum (87.1%)). The median time to recurrence was 15 months. The initial surgery was R0 in 17 patients (55%). The highest infiltrated sacral vertebra was S1 in 3%, S2 in 10%, S3 in 35%, S4 in 23%, S5 in 10%, and coccygeal in 19%. Anatomical changes post-resection, including rectal herniation into gluteal and subcutaneous tissues, significantly affected radiotherapy planning. Expansion of the initial tumor volume by 2 cm failed to encompass all recurrence origins in 72% of cases. A 5 cm expansion was required to achieve full coverage in 56% of patients, though 22% of recurrences still lay beyond this margin and the remaining were covered only partially. Conclusions: Recurrent sacrococcygeal chordomas exhibit complex, soft-tissue-dominant patterns and are influenced by significant anatomical displacement post-surgery. Standard target volume expansions are often insufficient to cover the predominantly multifocal recurrences. Full article
(This article belongs to the Special Issue Advanced Research on Spine Tumor)
Show Figures

Figure 1

25 pages, 6401 KiB  
Article
Efficient Sampling Schemes for 3D Imaging of Radar Target Scattering Based on Synchronized Linear Scanning and Rotational Motion
by Changyu Lou, Jingcheng Zhao, Xingli Wu, Yuchen Zhang, Zongkai Yang, Jiahui Li and Jungang Miao
Remote Sens. 2025, 17(15), 2636; https://doi.org/10.3390/rs17152636 - 29 Jul 2025
Viewed by 246
Abstract
Three-dimensional (3D) radar imaging is essential for target detection and measurement of scattering characteristics. Cylindrical scanning, a prevalent spatial sampling technique, provides benefits in engineering applications and has been extensively utilized for assessing the radar stealth capabilities of large aircraft. Traditional cylindrical scanning [...] Read more.
Three-dimensional (3D) radar imaging is essential for target detection and measurement of scattering characteristics. Cylindrical scanning, a prevalent spatial sampling technique, provides benefits in engineering applications and has been extensively utilized for assessing the radar stealth capabilities of large aircraft. Traditional cylindrical scanning generally utilizes highly sampled full-coverage techniques, leading to an excessive quantity of sampling points and diminished image efficiency, constraining its use for quick detection applications. This work presents an efficient 3D sampling strategy that integrates vertical linear scanning with horizontal rotating motion to overcome these restrictions. A joint angle–space sampling model is developed, and geometric constraints are implemented to enhance the scanning trajectory. The experimental results demonstrate that, compared to conventional techniques, the proposed method achieves a 94% reduction in the scanning duration while maintaining a peak sidelobe level ratio (PSLR) of 12 dB. Furthermore, this study demonstrates that 3D imaging may be accomplished solely by a “V”-shaped trajectory, efficiently determining the minimal possible sampling aperture. This approach offers novel insights and theoretical backing for the advancement of high-efficiency, low-redundancy 3D radar imaging systems. Full article
(This article belongs to the Special Issue Recent Advances in SAR: Signal Processing and Target Recognition)
Show Figures

Figure 1

20 pages, 1273 KiB  
Article
Safety and Anatomical Accuracy of Dry Needling of the Quadratus Femoris Muscle: A Cadaveric Study
by Marta Sánchez-Montoya, Jaime Almazán-Polo, Néstor Vallecillo Hernández, Charles Cotteret, Fabien Guerineau, Domingo de Guzman Monreal-Redondo and Ángel González-de-la-Flor
Healthcare 2025, 13(15), 1828; https://doi.org/10.3390/healthcare13151828 - 26 Jul 2025
Viewed by 263
Abstract
Introduction: Deep dry needling (DDN) is commonly applied in physiotherapy to treat musculoskeletal pain. The quadratus femoris (QF) muscle, located in the ischiofemoral space (IFS), represents a clinically relevant yet anatomically complex target. However, limited evidence exists on the safety, accuracy, and reliability [...] Read more.
Introduction: Deep dry needling (DDN) is commonly applied in physiotherapy to treat musculoskeletal pain. The quadratus femoris (QF) muscle, located in the ischiofemoral space (IFS), represents a clinically relevant yet anatomically complex target. However, limited evidence exists on the safety, accuracy, and reliability of non-ultrasound-guided DDN in this region. Aims: To assess the safety and accuracy of a standardized, non-ultrasound-guided DDN approach to the QF muscle, and to evaluate the intra- and inter-rater reliability of key procedural outcomes. Additionally, to determine the agreement between ultrasound imaging and anatomical dissection as validation methods for needle placement. Methods: An experimental cross-sectional study was conducted on five fresh cadavers (n = 24 approaches) by two physiotherapists with different DN experience. A standardized dry needling protocol was executed without ultrasound guidance, and anatomical and procedural variables were documented. Reliability (intra/inter-rater) was assessed for needle size, sciatic nerve (SN) puncture, IFS targeting, and overall success. In a subset, needle placement was validated through ultrasound and subsequent dissection. Results: The IFS was reached in 70.8% of procedures, and the SN was punctured in 16.7%. Inter-rater reliability for needle size was poor (κ = 0.04). Agreement between ultrasound and dissection was excellent for the ischiofemoral location and success (100%) and moderate for non SN puncture (90%; κ = 0.62). Conclusions: The standardized protocol demonstrated moderate accuracy and revealed a relevant clinical risk when targeting the quadratus femoris muscle. While inter-rater reliability was limited, agreement between ultrasound and dissection methods was high, supporting their complementary use for validating needle placement in anatomically complex procedures. Full article
Show Figures

Figure 1

58 pages, 1238 KiB  
Review
The Collapse of Brain Clearance: Glymphatic-Venous Failure, Aquaporin-4 Breakdown, and AI-Empowered Precision Neurotherapeutics in Intracranial Hypertension
by Matei Șerban, Corneliu Toader and Răzvan-Adrian Covache-Busuioc
Int. J. Mol. Sci. 2025, 26(15), 7223; https://doi.org/10.3390/ijms26157223 - 25 Jul 2025
Viewed by 328
Abstract
Although intracranial hypertension (ICH) has traditionally been framed as simply a numerical escalation of intracranial pressure (ICP) and usually dealt with in its clinical form and not in terms of its complex underlying pathophysiology, an emerging body of evidence indicates that ICH is [...] Read more.
Although intracranial hypertension (ICH) has traditionally been framed as simply a numerical escalation of intracranial pressure (ICP) and usually dealt with in its clinical form and not in terms of its complex underlying pathophysiology, an emerging body of evidence indicates that ICH is not simply an elevated ICP process but a complex process of molecular dysregulation, glymphatic dysfunction, and neurovascular insufficiency. Our aim in this paper is to provide a complete synthesis of all the new thinking that is occurring in this space, primarily on the intersection of glymphatic dysfunction and cerebral vein physiology. The aspiration is to review how glymphatic dysfunction, largely secondary to aquaporin-4 (AQP4) dysfunction, can lead to delayed cerebrospinal fluid (CSF) clearance and thus the accumulation of extravascular fluid resulting in elevated ICP. A range of other factors such as oxidative stress, endothelin-1, and neuroinflammation seem to significantly impair cerebral autoregulation, making ICH challenging to manage. Combining recent studies, we intend to provide a revised conceptualization of ICH that recognizes the nuance and complexity of ICH that is understated by previous models. We wish to also address novel diagnostics aimed at better capturing the dynamic nature of ICH. Recent advances in non-invasive imaging (i.e., 4D flow MRI and dynamic contrast-enhanced MRI; DCE-MRI) allow for better visualization of dynamic changes to the glymphatic and cerebral blood flow (CBF) system. Finally, wearable ICP monitors and AI-assisted diagnostics will create opportunities for these continuous and real-time assessments, especially in limited resource settings. Our goal is to provide examples of opportunities that exist that might augment early recognition and improve personalized care while ensuring we realize practical challenges and limitations. We also consider what may be therapeutically possible now and in the future. Therapeutic opportunities discussed include CRISPR-based gene editing aimed at restoring AQP4 function, nano-robotics aimed at drug targeting, and bioelectronic devices purposed for ICP modulation. Certainly, these proposals are innovative in nature but will require ethically responsible confirmation of long-term safety and availability, particularly to low- and middle-income countries (LMICs), where the burdens of secondary ICH remain preeminent. Throughout the review, we will be restrained to a balanced pursuit of innovative ideas and ethical considerations to attain global health equity. It is not our intent to provide unequivocal answers, but instead to encourage informed discussions at the intersections of research, clinical practice, and the public health field. We hope this review may stimulate further discussion about ICH and highlight research opportunities to conduct translational research in modern neuroscience with real, approachable, and patient-centered care. Full article
(This article belongs to the Special Issue Latest Review Papers in Molecular Neurobiology 2025)
Show Figures

Figure 1

23 pages, 7457 KiB  
Article
An Efficient Ship Target Integrated Imaging and Detection Framework (ST-IIDF) for Space-Borne SAR Echo Data
by Can Su, Wei Yang, Yongchen Pan, Hongcheng Zeng, Yamin Wang, Jie Chen, Zhixiang Huang, Wei Xiong, Jie Chen and Chunsheng Li
Remote Sens. 2025, 17(15), 2545; https://doi.org/10.3390/rs17152545 - 22 Jul 2025
Viewed by 317
Abstract
Due to the sparse distribution of ship targets in wide-area offshore scenarios, the typical cascade mode of imaging and detection for space-borne Synthetic Aperture Radar (SAR) echo data would consume substantial computational time and resources, severely affecting the timeliness of ship target information [...] Read more.
Due to the sparse distribution of ship targets in wide-area offshore scenarios, the typical cascade mode of imaging and detection for space-borne Synthetic Aperture Radar (SAR) echo data would consume substantial computational time and resources, severely affecting the timeliness of ship target information acquisition tasks. Therefore, we propose a ship target integrated imaging and detection framework (ST-IIDF) for SAR oceanic region data. A two-step filtering structure is added in the SAR imaging process to extract the potential areas of ship targets, which can accelerate the whole process. First, an improved peak-valley detection method based on one-dimensional scattering characteristics is used to locate the range gate units for ship targets. Second, a dynamic quantization method is applied to the imaged range gate units to further determine the azimuth region. Finally, a lightweight YOLO neural network is used to eliminate false alarm areas and obtain accurate positions of the ship targets. Through experiments on Hisea-1 and Pujiang-2 data, within sparse target scenes, the framework maintains over 90% accuracy in ship target detection, with an average processing speed increase of 35.95 times. The framework can be applied to ship target detection tasks with high timeliness requirements and provides an effective solution for real-time onboard processing. Full article
(This article belongs to the Special Issue Efficient Object Detection Based on Remote Sensing Images)
Show Figures

Figure 1

35 pages, 7685 KiB  
Article
Spatial and Spectral Structure-Aware Mamba Network for Hyperspectral Image Classification
by Jie Zhang, Ming Sun and Sheng Chang
Remote Sens. 2025, 17(14), 2489; https://doi.org/10.3390/rs17142489 - 17 Jul 2025
Viewed by 284
Abstract
Recently, a network based on selective state space models (SSMs), Mamba, has emerged as a research focus in hyperspectral image (HSI) classification due to its linear computational complexity and strong long-range dependency modeling capability. Originally designed for 1D causal sequence modeling, Mamba is [...] Read more.
Recently, a network based on selective state space models (SSMs), Mamba, has emerged as a research focus in hyperspectral image (HSI) classification due to its linear computational complexity and strong long-range dependency modeling capability. Originally designed for 1D causal sequence modeling, Mamba is challenging for HSI tasks that require simultaneous awareness of spatial and spectral structures. Current Mamba-based HSI classification methods typically convert spatial structures into 1D sequences and employ various scanning patterns to capture spatial dependencies. However, these approaches inevitably disrupt spatial structures, leading to ineffective modeling of complex spatial relationships and increased computational costs due to elongated scanning paths. Moreover, the lack of neighborhood spectral information utilization fails to mitigate the impact of spatial variability on classification performance. To address these limitations, we propose a novel model, Dual-Aware Discriminative Fusion Mamba (DADFMamba), which is simultaneously aware of spatial-spectral structures and adaptively integrates discriminative features. Specifically, we design a Spatial-Structure-Aware Fusion Module (SSAFM) to directly establish spatial neighborhood connectivity in the state space, preserving structural integrity. Then, we introduce a Spectral-Neighbor-Group Fusion Module (SNGFM). It enhances target spectral features by leveraging neighborhood spectral information before partitioning them into multiple spectral groups to explore relations across these groups. Finally, we introduce a Feature Fusion Discriminator (FFD) to discriminate the importance of spatial and spectral features, enabling adaptive feature fusion. Extensive experiments on four benchmark HSI datasets demonstrate that DADFMamba outperforms state-of-the-art deep learning models in classification accuracy while maintaining low computational costs and parameter efficiency. Notably, it achieves superior performance with only 30 training samples per class, highlighting its data efficiency. Our study reveals the great potential of Mamba in HSI classification and provides valuable insights for future research. Full article
(This article belongs to the Section Remote Sensing Image Processing)
Show Figures

Graphical abstract

28 pages, 7404 KiB  
Article
SR-YOLO: Spatial-to-Depth Enhanced Multi-Scale Attention Network for Small Target Detection in UAV Aerial Imagery
by Shasha Zhao, He Chen, Di Zhang, Yiyao Tao, Xiangnan Feng and Dengyin Zhang
Remote Sens. 2025, 17(14), 2441; https://doi.org/10.3390/rs17142441 - 14 Jul 2025
Viewed by 379
Abstract
The detection of aerial imagery captured by Unmanned Aerial Vehicles (UAVs) is widely employed across various domains, including engineering construction, traffic regulation, and precision agriculture. However, aerial images are typically characterized by numerous small targets, significant occlusion issues, and densely clustered targets, rendering [...] Read more.
The detection of aerial imagery captured by Unmanned Aerial Vehicles (UAVs) is widely employed across various domains, including engineering construction, traffic regulation, and precision agriculture. However, aerial images are typically characterized by numerous small targets, significant occlusion issues, and densely clustered targets, rendering traditional detection algorithms largely ineffective for such imagery. This work proposes a small target detection algorithm, SR-YOLO. It is specifically tailored to address these challenges in UAV-captured aerial images. First, the Space-to-Depth layer and Receptive Field Attention Convolution are combined, and the SR-Conv module is designed to replace the Conv module within the original backbone network. This hybrid module extracts more fine-grained information about small target features by converting image spatial information into depth information and the attention of the network to targets of different scales. Second, a small target detection layer and a bidirectional feature pyramid network mechanism are introduced to enhance the neck network, thereby strengthening the feature extraction and fusion capabilities for small targets. Finally, the model’s detection performance for small targets is improved by utilizing the Normalized Wasserstein Distance loss function to optimize the Complete Intersection over Union loss function. Empirical results demonstrate that the SR-YOLO algorithm significantly enhances the precision of small target detection in UAV aerial images. Ablation experiments and comparative experiments are conducted on the VisDrone2019 and RSOD datasets. Compared to the baseline algorithm YOLOv8s, our SR-YOLO algorithm has improved mAP@0.5 by 6.3% and 3.5% and mAP@0.5:0.95 by 3.8% and 2.3% on the datasets VisDrone2019 and RSOD, respectively. It also achieves superior detection results compared to other mainstream target detection methods. Full article
Show Figures

Figure 1

24 pages, 3798 KiB  
Article
A Robust Tracking Method for Aerial Extended Targets with Space-Based Wideband Radar
by Linlin Fang, Yuxin Hu, Lihua Zhong and Lijia Huang
Remote Sens. 2025, 17(14), 2360; https://doi.org/10.3390/rs17142360 - 9 Jul 2025
Viewed by 207
Abstract
Space-based radar systems offer significant advantages for air surveillance, including wide-area coverage and extended early-warning capabilities. The integrated design of detection and imaging in space-based wideband radar further enhances its accuracy. However, in the wideband tracking mode, large aircraft targets exhibit extended characteristics. [...] Read more.
Space-based radar systems offer significant advantages for air surveillance, including wide-area coverage and extended early-warning capabilities. The integrated design of detection and imaging in space-based wideband radar further enhances its accuracy. However, in the wideband tracking mode, large aircraft targets exhibit extended characteristics. Measurements from the same target cross multiple range resolution cells. Additionally, the nonlinear observation model and uncertain measurement noise characteristics under space-based long-distance observation substantially increase the tracking complexity. To address these challenges, we propose a robust aerial target tracking method for space-based wideband radar applications. First, we extend the observation model of the gamma Gaussian inverse Wishart probability hypothesis density filter to three-dimensional space by incorporating a spherical–radial cubature rule for improved nonlinear filtering. Second, variational Bayesian processing is integrated to enable the joint estimation of the target state and measurement noise parameters, and a recursive process is derived for both Gaussian and Student’s t-distributed measurement noise, enhancing the method’s robustness against noise uncertainty. Comprehensive simulations evaluating varying target extension parameters and noise conditions demonstrate that the proposed method achieves superior tracking accuracy and robustness. Full article
Show Figures

Graphical abstract

16 pages, 2134 KiB  
Article
Research on Field-of-View Reconstruction Technology of Specific Bands for Spatial Integral Field Spectrographs
by Jie Song, Yuyu Tang, Jun Wei and Xiaoxian Huang
Photonics 2025, 12(7), 682; https://doi.org/10.3390/photonics12070682 - 7 Jul 2025
Viewed by 228
Abstract
Integral field technology, as an advanced spectroscopic imaging technique, can be used to acquire the spatial and spectral information of the target area simultaneously. In this paper, we propose a method for the field reconstruction of characteristic wavelength bands of a space integral [...] Read more.
Integral field technology, as an advanced spectroscopic imaging technique, can be used to acquire the spatial and spectral information of the target area simultaneously. In this paper, we propose a method for the field reconstruction of characteristic wavelength bands of a space integral field spectrograph. The precise positioning of the image slicer is crucial to ensure that the spectrograph can accurately capture the position of each slicer in space. Firstly, the line spread function information and the characteristic location coordinates are obtained. Next, the positioning points of each group of image slicers under a specific spectral band are determined by quintic spline interpolation and a double-closed-loop optimization framework, thus establishing connection points for the responses of different image slicers. Then, the accuracy and reliability of the data are further improved by fitting the signal intensity of pixel points. Finally, the data of all image slicers are aligned to complete the field reconstruction of the characteristic wavelength bands of the space integral field spectrograph. This provides new ideas for the two-dimensional spatial reconstruction of spectrographs using image slicers as integral field units in specific spectral bands and accurately restores the two-dimensional spatial field observations of spatial integral field spectrographs. Full article
Show Figures

Figure 1

37 pages, 8636 KiB  
Article
Attitude Estimation of Spinning Space Targets Utilizing Multistatic ISAR Joint Observation
by Jishun Li, Canbin Yin, Can Xu, Jun He, Pengju Li and Yasheng Zhang
Remote Sens. 2025, 17(13), 2263; https://doi.org/10.3390/rs17132263 - 1 Jul 2025
Viewed by 257
Abstract
When a space target malfunctions and is no longer controlled by its attitude control system, it usually tumbles in orbit and exhibits a slow spinning state. Accurately estimating the on-orbit attitude of spinning space targets is of vital importance for ensuring the operation [...] Read more.
When a space target malfunctions and is no longer controlled by its attitude control system, it usually tumbles in orbit and exhibits a slow spinning state. Accurately estimating the on-orbit attitude of spinning space targets is of vital importance for ensuring the operation of space assets. Moreover, it plays a significant role in tasks such as reentry observation and collision avoidance. Currently, most existing methods estimate the attitude of space targets by using a single inverse synthetic aperture radar (ISAR) for long-term observation. However, this approach not only requires a long observation time but also fails to estimate the attitude of spinning targets. To address these limitations, this paper proposes a novel approach for estimating the attitude of spinning space targets, which utilizes the joint observations of a multiple-station ISAR. Specifically, the proposed method fully exploits the projection principle of ISAR imaging and uses an ISAR high-resolution network (ISAR-HRNet) to automatically extract the projection features of typical components of the target. Then, the analytical expressions for the target’s instantaneous attitude and spin vector under the multi-station observation imaging projection model are derived. Based on the extracted features of the typical components, the lengths, orientations, and spin vectors of the space target are determined. Importantly, the proposed method can achieve the attitude estimation of the spinning space targets within a single observation period, without the need for manual intervention or prior information about the target’s three-dimensional (3D) model. Additionally, the analytical method for solving the spin vector offers high efficiency and accuracy. Finally, the effectiveness of the proposed attitude estimation algorithm is verified by experiments on simulated data, and the performance of the ISAR-HRNet is also tested in the key point extraction experiments using measured data. Full article
(This article belongs to the Section Remote Sensing Image Processing)
Show Figures

Figure 1

Back to TopTop