ASVLB ‐Net: A lightweight network for multispectral weed segmentation with NDVI ‐guided adaptive fusion
Pest Management Science · 2026
DOI: 10.1002/ps.70881 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pest Management Science, 2026. Verfasst von Zhengtong Dong und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Information Engineering, Dalian University, Dalian, China. Hauptthemen laut MEDLINE: Plant Weeds, Image Processing, Computer-Assisted, Crops, Agricultural, Neural Networks, Computer. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: No. 2025YF14GX009, No.2025JH2/102800026.
Abstract (Originalsprache)
BACKGROUND Accurate crop‐weed segmentation is critical for precision spraying but is hindered by visual similarity and occlusion. This study aims to develop a lightweight yet high‐accuracy multispectral segmentation network to address these challenges. RESULTS We propose ASVLB‐Net, a novel lightweight multiscale feature extraction network guided by normalized difference vegetation index (NDVI) priors for crop‐weed segmentation in multispectral imagery. First, we design the Adaptive Spectral‐Vegetation Fusion (ASVF) module at the input stage. It achieves adaptive allocation of spectral and spatial features, enhancing the discriminability of vegetation regions. Additionally, we adopt a U‐shaped architecture based on the Layer‐wise Concatenated Multi‐Scale Feature (LCMF) encoder. It efficiently extracts fine‐grained features and boundary information with low computational cost. Finally, we use the Bottleneck ‐SCSA (Spatial and Channel Synergistic Attention) module to capture long‐range dependencies, improving segmentation in overlapping areas. Experimental results demonstrate that ASVLB‐Net achieves a mean intersection over union (mIOU) of 86.5% and a mean precision of 91.89%, outperforming several state‐of‐the‐art (SOTA) models, while maintaining high efficiency with only 0.47 million parameters. CONCLUSION The ASVLB‐Net significantly improves segmentation accuracy and robustness in unmanned aerial vehicle (UAV) multispectral imagery, offering a robust solution for precision weeding applications. © 2026 Society of Chemical Industry.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Information Engineering, Dalian University, Dalian, China.
- Mesee Quick Technology (Dalian) Co., Ltd, Dalian, China.
ORCID-Profile: Chuanming Song
Förderung
- No. 2025YF14GX009
- No.2025JH2/102800026
Zitieren
APA
Dong, Z., Mu, Z., Ji, J., Song, C., Wang, J., & Wang, Z. (2026). ASVLB ‐Net: A lightweight network for multispectral weed segmentation with NDVI ‐guided adaptive fusion. Pest Management Science. https://doi.org/10.1002/ps.70881
Vancouver
Dong Z, Mu Z, Ji J, Song C, Wang J, Wang Z. ASVLB ‐Net: A lightweight network for multispectral weed segmentation with NDVI ‐guided adaptive fusion. Pest Management Science. 2026. doi:10.1002/ps.70881
BibTeX
@article{Dong2026,
author = {Dong, Zhengtong and Mu, Zhenhua and Ji, Jie and Song, Chuanming and Wang, Jianjun and Wang, Zumin},
title = {ASVLB ‐Net: A lightweight network for multispectral weed segmentation with NDVI ‐guided adaptive fusion},
journal = {Pest Management Science},
year = {2026},
doi = {10.1002/ps.70881}
}
EUVIMED