Beyond Weak Labels: Prompt-Guided Local Refinement for Weakly Supervised Water Segmentation in High-Resolution Multispectral Imagery
本文针对高分辨率多光谱图像中水体分割标签难以获取的问题，提出了一种两阶段框架，通过伪标签学习和局部细化提升分割精度。
0 citationsRead paper
Selected work
Representative Papers
Multispectral airborne laser scanning dataset for tree species classification: MS-ALS-SPECIES
This study addresses the scarcity of publicly available multispectral airborne laser scanning (MS-ALS) datasets with high-quality field validation, which has hindered individual tree species classification research. We present the first open MS-ALS dataset, comprising 6,326 individually delineated trees across nine species in southern Finland, acquired using the HeliALS and Optech Titan dual-system sensors to capture three-wavelength point clouds. Ground truth was collected via an efficient and scalable field protocol. Leveraging deep learning models—including point cloud segmentation and Point Transformer architectures—we achieve high-accuracy species classification, demonstrating particularly strong performance for small-sized and rare tree species. Our results validate the efficacy of multispectral ALS data for fine-grained species discrimination and establish a benchmark platform to advance future research in this domain.
0 citationsRead paper
Deep Learning-based Robust Autonomous Navigation of Aerial Robots in Dense Forests
To address autonomous UAV navigation challenges in dense forests—characterized by GNSS denial, low visibility, slender and irregular obstacles, and degraded perception—this paper proposes a semantic-enhanced end-to-end navigation framework. Methodologically, it integrates stereo visual–inertial tightly coupled odometry, semantics-guided deep feature encoding, neural motion primitive evaluation, and introduces two novelties: a lateral maneuver control module and a temporal consistency-aware planning suppression mechanism; additionally, a real-time safety action filtering layer ensures flight stability. Evaluated across three real-world northern forest sites, the framework achieves 100% task completion in medium- and high-density scenes and 80% in extremely dense shrubland. Compared to baseline methods, it improves success rate by 23%, reduces trajectory jitter by 41%, and lowers collision rate by 67%, significantly enhancing robustness and safety in complex forest environments.
0 citationsRead paper
Field evaluation and optimization of a lightweight lidar-based UAV navigation system for dense boreal forest environments
To address localization failure and unreliable path planning caused by dense canopy occlusion in boreal forests, this paper proposes a lightweight, solid-state LiDAR–driven autonomous navigation system for quadcopters. Methodologically, we integrate LTA-OM SLAM with the IPC path planner to establish a real-time mapping and obstacle-avoidance framework, and introduce a standardized under-canopy evaluation protocol with quantitative metrics. Our key contribution is the co-optimization of SLAM and path planning, significantly enhancing system reproducibility and robustness. Experimental validation in medium-density (12/15 successful missions) and high-density forests (15/15) demonstrates substantial improvements in mission success rate. Leveraging 93 flight trials, we release the first publicly available dense-forest benchmark dataset. The proposed system reduces average mission completion time by 37% and markedly improves operational reliability.
0 citationsRead paper
Towards autonomous photogrammetric forest inventory using a lightweight under-canopy robotic drone
In GNSS-denied dense forest environments, autonomous UAV navigation remains challenging, and inversion accuracy of individual tree parameters—particularly diameter at breast height (DBH)—is typically low. To address these issues, this study proposes a lightweight under-canopy autonomous UAV system. It integrates an open-source robotics framework with a miniature binocular photogrammetry system, incorporating visual-inertial odometry (VIO), real-time SLAM, stereo matching, 3D point cloud reconstruction, and a deep learning-based stem detection algorithm. The system enables fully autonomous obstacle-avoidance flight and high-precision retrieval of individual tree structural parameters. Field validation in boreal coniferous forests achieved a stem detection rate of 79.31%; the overall DBH estimation RMSE was 3.33 cm (12.79%), decreasing to 1.16 cm (5.74%) for small-diameter trees (<30 cm). These results significantly enhance automation capability for forest resource inventory in GNSS-denied under-canopy scenarios.
0 citationsRead paper
Recent publications
Latest Papers
Beyond Weak Labels: Prompt-Guided Local Refinement for Weakly Supervised Water Segmentation in High-Resolution Multispectral Imagery
本文针对高分辨率多光谱图像中水体分割标签难以获取的问题，提出了一种两阶段框架，通过伪标签学习和局部细化提升分割精度。
0 citationsRead paper
Multispectral airborne laser scanning dataset for tree species classification: MS-ALS-SPECIES
This study addresses the scarcity of publicly available multispectral airborne laser scanning (MS-ALS) datasets with high-quality field validation, which has hindered individual tree species classification research. We present the first open MS-ALS dataset, comprising 6,326 individually delineated trees across nine species in southern Finland, acquired using the HeliALS and Optech Titan dual-system sensors to capture three-wavelength point clouds. Ground truth was collected via an efficient and scalable field protocol. Leveraging deep learning models—including point cloud segmentation and Point Transformer architectures—we achieve high-accuracy species classification, demonstrating particularly strong performance for small-sized and rare tree species. Our results validate the efficacy of multispectral ALS data for fine-grained species discrimination and establish a benchmark platform to advance future research in this domain.
0 citationsRead paper
Deep Learning-based Robust Autonomous Navigation of Aerial Robots in Dense Forests
To address autonomous UAV navigation challenges in dense forests—characterized by GNSS denial, low visibility, slender and irregular obstacles, and degraded perception—this paper proposes a semantic-enhanced end-to-end navigation framework. Methodologically, it integrates stereo visual–inertial tightly coupled odometry, semantics-guided deep feature encoding, neural motion primitive evaluation, and introduces two novelties: a lateral maneuver control module and a temporal consistency-aware planning suppression mechanism; additionally, a real-time safety action filtering layer ensures flight stability. Evaluated across three real-world northern forest sites, the framework achieves 100% task completion in medium- and high-density scenes and 80% in extremely dense shrubland. Compared to baseline methods, it improves success rate by 23%, reduces trajectory jitter by 41%, and lowers collision rate by 67%, significantly enhancing robustness and safety in complex forest environments.
0 citationsRead paper
Field evaluation and optimization of a lightweight lidar-based UAV navigation system for dense boreal forest environments
To address localization failure and unreliable path planning caused by dense canopy occlusion in boreal forests, this paper proposes a lightweight, solid-state LiDAR–driven autonomous navigation system for quadcopters. Methodologically, we integrate LTA-OM SLAM with the IPC path planner to establish a real-time mapping and obstacle-avoidance framework, and introduce a standardized under-canopy evaluation protocol with quantitative metrics. Our key contribution is the co-optimization of SLAM and path planning, significantly enhancing system reproducibility and robustness. Experimental validation in medium-density (12/15 successful missions) and high-density forests (15/15) demonstrates substantial improvements in mission success rate. Leveraging 93 flight trials, we release the first publicly available dense-forest benchmark dataset. The proposed system reduces average mission completion time by 37% and markedly improves operational reliability.
0 citationsRead paper
Towards autonomous photogrammetric forest inventory using a lightweight under-canopy robotic drone
In GNSS-denied dense forest environments, autonomous UAV navigation remains challenging, and inversion accuracy of individual tree parameters—particularly diameter at breast height (DBH)—is typically low. To address these issues, this study proposes a lightweight under-canopy autonomous UAV system. It integrates an open-source robotics framework with a miniature binocular photogrammetry system, incorporating visual-inertial odometry (VIO), real-time SLAM, stereo matching, 3D point cloud reconstruction, and a deep learning-based stem detection algorithm. The system enables fully autonomous obstacle-avoidance flight and high-precision retrieval of individual tree structural parameters. Field validation in boreal coniferous forests achieved a stem detection rate of 79.31%; the overall DBH estimation RMSE was 3.33 cm (12.79%), decreasing to 1.16 cm (5.74%) for small-diameter trees (<30 cm). These results significantly enhance automation capability for forest resource inventory in GNSS-denied under-canopy scenarios.
0 citationsRead paper