Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations
Leaf carotenoids (Cxc) play a crucial role in vegetation as essential pigments responsible for capturing sunlight and protecting leaf tissues. They provide vital insights into a plant physiological status and serve as sensitive indicators of plant stress. However, remote sensing of Cxc at the leaf l...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/20/4997 |
| _version_ | 1827719916464111616 |
|---|---|
| author | Weilin Hao Jia Sun Zichao Zhang Kan Zhang Feng Qiu Jin Xu |
| author_facet | Weilin Hao Jia Sun Zichao Zhang Kan Zhang Feng Qiu Jin Xu |
| author_sort | Weilin Hao |
| collection | DOAJ |
| description | Leaf carotenoids (Cxc) play a crucial role in vegetation as essential pigments responsible for capturing sunlight and protecting leaf tissues. They provide vital insights into a plant physiological status and serve as sensitive indicators of plant stress. However, remote sensing of Cxc at the leaf level has been challenging due to the low Cxc content and weaker absorption features compared to those of chlorophylls in the visible domain. Existing vegetation indices have been widely applied but often lack a solid physical foundation, which limits their applicability and robustness in characterizing Cxc. Yet, physical models can confront this ill-posed problem, though with high operational costs. To address this issue, this study presents a novel hybrid inversion method that combines the multilayer perceptron (MLP) algorithm with PROSPECT model simulations to accurately retrieve Cxc. The effectiveness of the MLP method was investigated through comparisons with the classical PROSPECT model inversion (look-up table [LUT] method), the convolutional neural network (CNN) hybrid model, and the Transformer hybrid model. In the pooled results of six experimental datasets, the MLP method exhibited its robustness and generalization capabilities for leaf Cxc content estimation, with <i>RMSE</i> of 3.12 μg/cm<sup>2</sup> and <i>R</i><sup>2</sup> of 0.52. The Transformer (<i>RMSE</i> = 3.14 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.46), CNN (<i>RMSE</i> = 3.42 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.28), and LUT (<i>RMSE</i> = 3.82 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.24) methods followed in descending order of accuracy. A comparison with previous studies using the same public datasets (ANGERS and LOPEX) also demonstrated the performance of the MLP method from another perspective. These findings underscore the potential of the proposed MLP hybrid method as a powerful tool for accurate Cxc retrieval applications, providing valuable insights into vegetation health and stress response. |
| first_indexed | 2024-03-10T20:55:21Z |
| format | Article |
| id | doaj.art-16297814e6824331a6545f95b431f05a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T20:55:21Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-16297814e6824331a6545f95b431f05a2023-11-19T17:59:22ZengMDPI AGRemote Sensing2072-42922023-10-011520499710.3390/rs15204997Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT SimulationsWeilin Hao0Jia Sun1Zichao Zhang2Kan Zhang3Feng Qiu4Jin Xu5School of Computer Science, Peking University, Beijing 100871, ChinaSchool of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, Peking University, Beijing 100871, ChinaSuzhou Enterprise Credit Service Co., Ltd., Suzhou 215000, ChinaInternational Institute for Earth System Science, Nanjing University, Nanjing 210023, ChinaSchool of Computer Science, Peking University, Beijing 100871, ChinaLeaf carotenoids (Cxc) play a crucial role in vegetation as essential pigments responsible for capturing sunlight and protecting leaf tissues. They provide vital insights into a plant physiological status and serve as sensitive indicators of plant stress. However, remote sensing of Cxc at the leaf level has been challenging due to the low Cxc content and weaker absorption features compared to those of chlorophylls in the visible domain. Existing vegetation indices have been widely applied but often lack a solid physical foundation, which limits their applicability and robustness in characterizing Cxc. Yet, physical models can confront this ill-posed problem, though with high operational costs. To address this issue, this study presents a novel hybrid inversion method that combines the multilayer perceptron (MLP) algorithm with PROSPECT model simulations to accurately retrieve Cxc. The effectiveness of the MLP method was investigated through comparisons with the classical PROSPECT model inversion (look-up table [LUT] method), the convolutional neural network (CNN) hybrid model, and the Transformer hybrid model. In the pooled results of six experimental datasets, the MLP method exhibited its robustness and generalization capabilities for leaf Cxc content estimation, with <i>RMSE</i> of 3.12 μg/cm<sup>2</sup> and <i>R</i><sup>2</sup> of 0.52. The Transformer (<i>RMSE</i> = 3.14 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.46), CNN (<i>RMSE</i> = 3.42 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.28), and LUT (<i>RMSE</i> = 3.82 μg/cm<sup>2</sup>, <i>R</i><sup>2</sup> = 0.24) methods followed in descending order of accuracy. A comparison with previous studies using the same public datasets (ANGERS and LOPEX) also demonstrated the performance of the MLP method from another perspective. These findings underscore the potential of the proposed MLP hybrid method as a powerful tool for accurate Cxc retrieval applications, providing valuable insights into vegetation health and stress response.https://www.mdpi.com/2072-4292/15/20/4997multilayer perceptronleaf carotenoid contentPROSPECThybrid inversionTransformerCNN |
| spellingShingle | Weilin Hao Jia Sun Zichao Zhang Kan Zhang Feng Qiu Jin Xu Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations Remote Sensing multilayer perceptron leaf carotenoid content PROSPECT hybrid inversion Transformer CNN |
| title | Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations |
| title_full | Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations |
| title_fullStr | Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations |
| title_full_unstemmed | Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations |
| title_short | Novel Hybrid Model to Estimate Leaf Carotenoids Using Multilayer Perceptron and PROSPECT Simulations |
| title_sort | novel hybrid model to estimate leaf carotenoids using multilayer perceptron and prospect simulations |
| topic | multilayer perceptron leaf carotenoid content PROSPECT hybrid inversion Transformer CNN |
| url | https://www.mdpi.com/2072-4292/15/20/4997 |
| work_keys_str_mv | AT weilinhao novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations AT jiasun novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations AT zichaozhang novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations AT kanzhang novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations AT fengqiu novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations AT jinxu novelhybridmodeltoestimateleafcarotenoidsusingmultilayerperceptronandprospectsimulations |