模型约束与机器学习下的植物类胡萝卜素和叶绿素含量反演方法

doi:10.3964/j.issn.1000-0593(2024)08-2174-09

Abstract
References
Related Citation (15)

Download: PDF (6344 KB)
Export: BibTeX | EndNote (RIS)

Abstract The chlorophyll and carotenoid content is an important indicator for evaluating the health status of plants. The PROSPECT model, coupled with machine learning, has been widely used to retrieve the biochemical properties of vegetation. However, the application of the coupled model is limited due to the differences between the leaf-directional hemispherical reflectance factor (DHRF) spectra and the bidirectional reflectance factor (BRF) spectra. This paper utilizes the leaf spectral database of North American plant (EcoSIS) as the experimental dataset and introduces the PROSPECT model as an additional constraint for machine learning. This approach creates a hybrid dataset by employing wavelet continuous wavelet transform (CWT) to generate the wavelet coefficient spectrum and the derivative spectrum generated by the first-order derivative (FD). Three kinds of feature extraction algorithms, namely competitive adaptive reweighting algorithm (CARS), successive projection algorithm (SPA), and principal component analysis (PCA) were applied to extract spectral features for chlorophylls and carotenoids in the full-spectral domains and the subdomain of VNIR spectroscopy. Based on the above 12 combinations of different methods, artificial neural network (ANN) prediction models for chlorophyll and carotenoids were separately established. The results show that the simulated data under the constraint of the PROSPECT model enhanced the quality of the training set for machine learning to a certain extent. Additionally, the spectra processed by the first-order derivatives and wavelet transforms were able to reduce better the bias between the simulated spectra of the DHRF and the measured spectra of the BRF. The best inversion of leaf chlorophyll is achieved with the FD+CARS combination in the whole spectral domain, yielding a test set R² of 0.806 4 and RMSE of 2.911 4. Meanwhile, the CWT+CARS combination in the VNIR spectral sub-domain offers the best results for leaf carotenoids, with a test set R² of 0.797 2 and RMSE of 0.414 1. The proposed method can provide researchers with a reference to extract biochemical characteristics of plant leaves more accurately and efficiently from BRF spectra and other near-end reflectance images.

Key words：Spectral data; Model constraints; Wavelet coefficient spectrum; First-order derivative spectrum; Chlorophyll; Carotenoid; Precise retrieval

Received: 2023-04-18 Accepted: 2023-10-31

ZTFLH:

O433.4

Corresponding Authors: XU Yuan-yuan E-mail: yuanyuanxulark@126.com

Cite this article:

TANG Fu-rui,XU Yuan-yuan,GENG Yan, et al. Retrieval of Plant Carotenoids and Chlorophyll Contents With Model Constraints and Machine Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(08): 2174-2182.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2024)08-2174-09 OR https://www.gpxygpfx.com/EN/Y2024/V44/I08/2174

[1] Féret J B，Gitelson A A，Noble S D，et al. Remote Sensing of Environment，2017，193：204.
[2] Du L，Yang J，Sun J，et al. Frontiers in Plant Science，2020，11：533.
[3] Du L，Shi S，Yang J，et al. Remote Sensing，2016，8(6)：526.
[4] LEI Xiang-xiang，ZHAO Jing，LIU Hou-cheng，et al(雷祥祥，赵静，刘厚诚，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2019，39(10)：3256.
[5] Boukabara S A，Krasnopolsky V，Penny S G，et al. Bulletin of the American Meteorological Society，2021，102(5)：E1016.
[6] Berger K，Verrelst J，Féret J B，et al. International Journal of Applied Earth Observation and Geoinformation，2020，92：102174.
[7] YANG Qian-qian，JIN Cai-yi，LI Tong-wen，et al(杨倩倩，靳才溢，李同文，等). National Remote Sensing Bulletin(遥感学报)，2022，26(2)：268.
[8] Li D, Tian L, Wan Z F，et al. Remote Sensing of Environment，2019，231：111240.
[9] Jay S, Bendoula R, Hadoux X，et al. Remote Sensing of Environment，2016，177：220.
[10] Li D, Cheng T, Jia M，et al. Remote Sensing of Environment，2018，206：1.
[11] Wan L, Zhang J F, Xu Y，et al. Remote Sensing of Environment，2021，267：112761.
[12] Kothari S, Beauchamp-Rioux R, Blanchard F, et al. New Phytologist, 2023, 238: 549.
[13] Wang Z H, Townsend P A, Kruger E L. New Phytologist, 2022, 235: 923.
[14] Asner G P. New Phytologist, 2014, 204(1): 127.
[15] Bousquet L, Lachérade S, Jacquemoud S，et al. Remote Sensing of Environment，2005，98：201.
[16] Cheng T, Rivard B, Sánchez-Azofeifa A G，et al. ISPRS Journal of Photogrammetry and Remote Sensing，2014，87：28.
[17] ZHANG Li-xin, YANG Cui-fang, CHEN Jie, et al(张立欣，杨翠芳，陈杰，等). Food and Fermentation Industries(食品与发酵工业)，2022，48(20)：36.
[18] ZHANG Ran-ran, YING Lu-na, ZHOU Wei-dong(张冉冉，应璐娜，周卫东). Chinese Journal of Quantum Electronics(量子电子学报)，2023，40(3)：376.
[19] Féret J B, Berger K, Boissieu F，et al. Remote Sensing of Environment，2021，252：112173.
[20] Spafford L, Maire G, MacDougall A，et al. Remote Sensing of Environment，2021，252：112176.
[21] Rajewicz P A，Atherton J，Alonso L，et al. Remote Sensing，2019，11(5)：532.
[22] Li X, Sun Z, Lu S Q，et al. Remote Sensing of Environment，2023，297：113754.