EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PlantEFRSegnet: A Plant Point Cloud Segmentation Network Based on Edge Point Preservation and Feature Feedback Repair

Bin Li iD, Peng Liu, Yonghan Zhang

Sensors · 14.05.2026 · 26 (10) · 3104

DOI: 10.3390/s26103104 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 14.05.2026. Verfasst von Bin Li und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, Northeast Electric Power University, Jilin 132012, China. Hauptthemen laut MEDLINE: Plants, Image Processing, Computer-Assisted, Imaging, Three-Dimensional. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Education of Jilin Province, Jilin Provincial Department of Education of Jilin Province, the Key Laboratory of Intelligent Rehabilitation and Barrier-free for the Disabled (Changchun University), Ministry of Education.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The segmentation of 3D point clouds of plant organs, such as leaves and stems, helps to monitor plant growth and is a key step in plant growth phenotype analysis. Compared to point cloud segmentation tasks in other fields, plant point cloud segmentation is more challenging due to the interwoven distribution of various parts such as stems, leaves, and flowers. In this paper, we propose a universal point cloud segmentation network PlantEFRSegnet that can be used for multi-species of plants. The proposed PlantEFRSegnet utilizes a newly designed edge point preservation downsampling module to identify and preserve the points at the edges of plant organs during the downsampling process, in order to assist the segmentation network in learning the contours of various plant organs. PlantEFRSegnet performs supervised feature repair on the point cloud features obtained through downsampling to mitigate the impact of feature loss on segmentation performance during feature embedding. The encoder of the segmentation network is composed of four local feature extraction modules. These four modules can not only extract features but also enhance the features corresponding to points with high contributions in local regions based on point attention mechanism. We evaluated the proposed PlantEFRSegnet on a laser-scanned plant point cloud dataset. Compared with the state-of-the-art approaches, the proposed PlantEFRSegnet achieved better segmentation results.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PlantsImage Processing, Computer-AssistedImaging, Three-DimensionalAlgorithmsPlant Leaves

Stichwörter der Autor:innen

plant phenotypingplant point cloudpoint cloud downsamplingpoint cloud segmentation

Autor:innen und Institutionen

  • School of Computer Science, Northeast Electric Power University, Jilin 132012, China.
  • School of Mathematical and Computer Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK.

ORCID-Profile: Bin Li

Förderung

  • Department of Education of Jilin Province
  • Jilin Provincial Department of Education of Jilin Province
  • the Key Laboratory of Intelligent Rehabilitation and Barrier-free for the Disabled (Changchun University), Ministry of Education

Zitieren

APA

Li, B., Liu, P., & Zhang, Y. (2026). PlantEFRSegnet: A Plant Point Cloud Segmentation Network Based on Edge Point Preservation and Feature Feedback Repair. Sensors, 26(10), 3104. https://doi.org/10.3390/s26103104

Vancouver

Li B, Liu P, Zhang Y. PlantEFRSegnet: A Plant Point Cloud Segmentation Network Based on Edge Point Preservation and Feature Feedback Repair. Sensors. 2026;26(10):3104. doi:10.3390/s26103104

BibTeX
@article{Li2026,
  author = {Li, Bin and Liu, Peng and Zhang, Yonghan},
  title = {PlantEFRSegnet: A Plant Point Cloud Segmentation Network Based on Edge Point Preservation and Feature Feedback Repair},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {3104},
  doi = {10.3390/s26103104}
}
Kostenlos starten →