Hyperspectral inversion model of ginkgo leaf yield prediction based on machine learning.
Source check: Source checked. Not independent scientific confirmation.
Finding: Airborne hyperspectral measurements were used to estimate ginkgo leaf yield in field cultivation. The best reported combination, using PSAMA-selected spectral bands, vegetation indices and a BiLSTM with grid search, achieved Rp² 0.8795 and RMSEP 0.1021; the SNV-PLSR baseline achieved Rp² 0.7831 and RMSEP 0.0325. This is a field-yield prediction result, not evidence of transfer across sites or seasons.
Technology: Particle Swarm Attention Mechanism feature selection with BiLSTM grid-search regression
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Learned spectral regression models predicted ginkgo leaf yield from airborne hyperspectral features; PSAMA selected bands and the BiLSTM-GS model estimated yield.
- Test and result
- Hyperspectral reflectance and selected vegetation indices were used to predict measured ginkgo leaf yield. The paper compares preprocessing and feature-selection/model combinations; the best reported combination is PSAMA with BiLSTM-GS (Rp² 0.8795, RMSEP 0.1021).
- Limitation / still unknown
- The reviewed result is from the study’s field data; performance across other sites, seasons or cultivars is not established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC12698573/ · https://pubmed.ncbi.nlm.nih.gov/41393880/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.