Study on Outdoor Spectral Inversion of Winter Jujube Based on BPDF Models

Yabei Di, Jinlong Yu, Huaping Luo (), Huaiyu Liu, Lei Kang and Yuesen Tong
Additional contact information
Yabei Di: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China
Jinlong Yu: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China
Huaping Luo: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China
Huaiyu Liu: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China
Lei Kang: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China
Yuesen Tong: College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China

Agriculture, 2025, vol. 15, issue 13, 1-16

Abstract: The outdoor spectral detection of winter jujube quality is affected by complex ambient light and surface heterogeneity, resulting in limited inversion accuracy. To address this problem, this study proposes a correction method for outdoor spectral inversion based on the bidirectional polarization reflectance distribution function (BPDF) model. It was used to enhance the detection accuracy of water content and soluble solid (SSC) content of winter jujube. Experimentally, 900–1750 nm hyperspectral data of ripe winter jujube samples were collected at non-polarization and 0°, 45°, 90°, and 135° polarization azimuths. The spectra were inverted using four semi-empirical BPDF models, Nadal–Breon, Litvinov, Maignan and Xie–Cheng, and the corrected spectra were obtained by mean fusion. The quality prediction models are subsequently combined with the competitive adaptive reweighting algorithm (CARS) and partial least squares (PLS). The results showed that the modified spectra significantly optimized the prediction performance. The prediction set correlation coefficients (Rp) of the water content and SSC models were improved by 10–30% compared with the original spectra. The percentage of models with RPIQ values greater than 2 increased from 40% to 60%. Among them, the Litvinov model performs outstandingly in the direction of no polarization and 135° polarization, with the highest Rp of 0.8829 for water content prediction and RPIQ of 2.54. The Xie–Cheng model has an RPIQ of 2.64 for SSC prediction at 90° polarization, which shows the advantage of sensitivity to the deeper constituents. The different models complemented each other in multi-polarization scenarios. The Nadal–Breon model was suitable for epidermal reflection-dominated scenarios, and the Maignan model efficiently coupled epidermal and internal moisture characteristics through the moisture sensitivity index. The study verifies the effectiveness of the spectral correction method based on the BPDF model for outdoor quality detection of winter jujube, which provides a new path for the spectral detection of agricultural products in complex environments. In the future, it is necessary to further optimize the dynamic adjustment mechanism of the model parameters and improve the ability of environmental interference correction by combining multi-source data fusion.

Keywords: BPDF models; hyperspectral imaging; spectral inversion; spectral correction; quality detection (search for similar items in EconPapers)
JEL-codes: Q1 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2077-0472/15/13/1334/pdf (application/pdf)
https://www.mdpi.com/2077-0472/15/13/1334/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jagris:v:15:y:2025:i:13:p:1334-:d:1684331

Access Statistics for this article

Agriculture is currently edited by Ms. Leda Xuan

More articles in Agriculture from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().