#树高测量

提出3D实例分割与骨架化框架，自动提取苹果树枝条表型性状。

构建非洲四国3493棵树的冠幅-胸径异速生长数据集，用于遥感估算森林外树木生物量。

Eucalyptus species are widely cultivated globally, with over 22.5 million hafv planted, owing to their rapid growth, adaptability, versatility for cellulose, wood, and energy production, and their role in carbon fixation. Other species, including pine, acacia, teak, poplar, rubber tree, and birch, are also widely cultivated for various uses. Measuring tree height in plantations is fundamental for estimating volume and biomass and for assessing the plantation condition, a key variable in forest inventories. Although challenging, various techniques, including trigonometric methods and technologies such as Light Detection and Ranging (LiDAR) sensors, photogrammetry, and Mobile Laser Scanner (MLS), have been employed to obtain accurate height measurem​ents. This study presents a new algorithm for estimating the total height ( H ) of trees in forest plantations using MLS data. It incorporates characteristics of organized plantations to improve trunk search and uses the theory of natural tree taper to obtain more accurate height estimates. Validated in two study areas, it demonstrated reliability, especially for short trees, and high accuracy, achieving Root Mean Squared Error (RMSE) (%) values of 3.32% for area 1 and 3.26% for area 2. In contrast, traditional algorithms such as, TreeLS yielded RMSE (%) values of 14.70% for area 1 and 14.40% for area 2, whereas the 3DFin algorithm obtained RMSE (%) values of 12.49% for area 1 and 21.43% for area 2. Although it requires more processing time than the 3DFin algorithm, the proposed method provides more accurate results for forest inventories.

调查加拿大城市树木清单的现状、发展与应用，并分析其与市政特征的关系。

整合土壤与冠层传感技术，绘制酸樱桃园地块内变异图，评估其用于管理决策的可靠性。

利用永久地面激光扫描量化北方森林枯枝的吸湿运动，发现其与木材平衡含水率强相关。

本文系统综述双目立体视觉在农业应用中的最新进展,涵盖系统架构、传统与基于深度学习的立体匹配方法、点云重建技术及新兴的3D高斯溅射等监督策略。综述总结了关键应用,包括高通量表型分析、果实定位与机器人采摘、杂草检测与精准喷施、自主导航以及畜禽体况评估,突出其在多任务农业感知系统中的作用。论文还讨论了低纹理匹配困难、复杂环境中的遮挡、跨域泛化、实时轻量化部署及有限数据集可用性等主要挑战,并展望基于基础模型的视觉感知、自监督与弱监督学习、多模态融合及边缘高效模型设计的未来方向。原文标题:Advances in Binocular Stereo Vision-Driven 3D Perception and Intelligent Analysis Methods for Agriculture,作者:江苏大学Ye R、Wang J、Kong Z、Ou M,发表于MDPI Applied Sciences 2026年第16卷第16期7957。

研究视频多目标跟踪系统在不同扫描条件下估算苹果产量的性能。