EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A review of artificial intelligence technologies for mechanized Camellia oleifera fruit harvesting

Shouxiang Jin, Lei Zhou, Hongping Zhou, Hongzhe Jiang

Industrial Crops and Products · 2026-11 · 253 · 124677

DOI: 10.1016/j.indcrop.2026.124677 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Industrial Crops and Products, 2026-11. Verfasst von Shouxiang Jin und 3 weiteren Autor:innen. Die Arbeit nennt 137 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Camellia oleifera is a major woody oil crop widely cultivated in the hilly and mountainous regions of China, where complex canopy architecture, severe branch and leaf occlusion, uneven illumination, and difficult terrain make harvesting labor-intensive. Recent advances in artificial intelligence (AI) offer opportunities to improve visual perception, environmental understanding, and robotic decision-making in mechanized harvesting. This review focuses on the application of AI to mechanized Camellia oleifera harvesting across three major harvesting paradigms: branch-oriented vibration harvesting, canopy-oriented contact harvesting, and fruit-oriented selective picking. It examines fruit detection, segmentation and three-dimensional localization, branch and canopy perception, vibration point prediction, natural frequency identification, harvesting parameter optimization, harvestability assessment, and robotic path planning. Representative studies on other fruit crops are reviewed to identify transferable technical frameworks and common challenges in unstructured orchard environments. Although these harvesting paradigms differ in their interaction targets and operational mechanisms, they share enabling technologies such as multimodal perception, point cloud processing, three-dimensional sensing, and intelligent decision-making, creating opportunities for future cross-paradigm integration. Major challenges include severe occlusion, limited high-quality datasets, constrained edge-computing capability, and the gap between visual perception and physical harvesting execution. Future priorities include domain-specific foundation models, physics-informed perception and decision-making, edge intelligence, efficient multimodal sensing, and standardized datasets and system-level evaluation frameworks. This review provides a task-oriented evidence map for developing efficient and field-ready intelligent Camellia oleifera harvesting systems.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Jin, S., Zhou, L., Zhou, H., & Jiang, H. (2026). A review of artificial intelligence technologies for mechanized Camellia oleifera fruit harvesting. Industrial Crops and Products, 253, 124677. https://doi.org/10.1016/j.indcrop.2026.124677

Vancouver

Jin S, Zhou L, Zhou H, Jiang H. A review of artificial intelligence technologies for mechanized Camellia oleifera fruit harvesting. Industrial Crops and Products. 2026;253:124677. doi:10.1016/j.indcrop.2026.124677

BibTeX
@article{Jin2026,
  author = {Jin, Shouxiang and Zhou, Lei and Zhou, Hongping and Jiang, Hongzhe},
  title = {A review of artificial intelligence technologies for mechanized Camellia oleifera fruit harvesting},
  journal = {Industrial Crops and Products},
  year = {2026},
  volume = {253},
  pages = {124677},
  doi = {10.1016/j.indcrop.2026.124677}
}
Kostenlos starten →