Normalized difference vegetation index prediction using reservoir computing and pretrained language models

In this study, we examined plant health prediction through the Normalized Difference Vegetation Index (NDVI) calculated from satellite image derived reflectance values in the near-infrared and red spectra. The problem is formulated as a temporal data prediction problem. Using MODIS/Terra Vegetation...

Full description

Saved in:
Bibliographic Details
Main Authors: John Olamofe, Ram Ray, Xishuang Dong, Lijun Qian
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2025-03-01
Series:Artificial Intelligence in Agriculture
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2589721724000539
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832595312232890368
author John Olamofe
Ram Ray
Xishuang Dong
Lijun Qian
author_facet John Olamofe
Ram Ray
Xishuang Dong
Lijun Qian
author_sort John Olamofe
collection DOAJ
description In this study, we examined plant health prediction through the Normalized Difference Vegetation Index (NDVI) calculated from satellite image derived reflectance values in the near-infrared and red spectra. The problem is formulated as a temporal data prediction problem. Using MODIS/Terra Vegetation Indices 16-Day L3 Global 250 m SIN Grid V061 dataset, we designed and implemented Reservoir Computing (RC) models and transformer-based models including pretrained language model, and compared the prediction performance of these models to traditional machine learning and deep learning methods such as Nonlinear Regression, Decision Tree, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) network, and DLinear. It is observed that the DLinear/LSTM model showed exceptional predictive accuracy, while the pretrained RC model significantly enhanced traditional RC model forecasts. Additionally, Frozen Pretrained Transformer (FPT), a pretrained language model, showed superior performance in predicting specific NDVI values (most often peak or lowest NDVI), suggesting its effectiveness in precise temporal predictions. Furthermore, transformer-based models, specifically PatchTST and FPT, demonstrated substantial mean squared error reductions, particularly in limited data scenarios (1 %, 5 %, 15 % and 50 % sample sizes), indicating their robustness in precise NDVI temporal predictions when data is limited. The findings in this study demonstrated the effectiveness of emerging machine learning techniques such as reservoir computing and pretrained language model for remote sensing and their contributions in precision agriculture.
format Article
id doaj-art-29152a11148b4db19ffa656ac6cf2448
institution Kabale University
issn 2589-7217
language English
publishDate 2025-03-01
publisher KeAi Communications Co., Ltd.
record_format Article
series Artificial Intelligence in Agriculture
spelling doaj-art-29152a11148b4db19ffa656ac6cf24482025-01-19T06:26:32ZengKeAi Communications Co., Ltd.Artificial Intelligence in Agriculture2589-72172025-03-01151116129Normalized difference vegetation index prediction using reservoir computing and pretrained language modelsJohn Olamofe0Ram Ray1Xishuang Dong2Lijun Qian3CREDIT Center and Electrical and Computer Engineering Department, Prairie View A&M University, Prairie View 77446, TX, USA; Corresponding author.College of Agriculture, Food, and Natural Resources, Prairie View A&M University, Prairie View 77446, TX, USACREDIT Center and Electrical and Computer Engineering Department, Prairie View A&M University, Prairie View 77446, TX, USACREDIT Center and Electrical and Computer Engineering Department, Prairie View A&M University, Prairie View 77446, TX, USAIn this study, we examined plant health prediction through the Normalized Difference Vegetation Index (NDVI) calculated from satellite image derived reflectance values in the near-infrared and red spectra. The problem is formulated as a temporal data prediction problem. Using MODIS/Terra Vegetation Indices 16-Day L3 Global 250 m SIN Grid V061 dataset, we designed and implemented Reservoir Computing (RC) models and transformer-based models including pretrained language model, and compared the prediction performance of these models to traditional machine learning and deep learning methods such as Nonlinear Regression, Decision Tree, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) network, and DLinear. It is observed that the DLinear/LSTM model showed exceptional predictive accuracy, while the pretrained RC model significantly enhanced traditional RC model forecasts. Additionally, Frozen Pretrained Transformer (FPT), a pretrained language model, showed superior performance in predicting specific NDVI values (most often peak or lowest NDVI), suggesting its effectiveness in precise temporal predictions. Furthermore, transformer-based models, specifically PatchTST and FPT, demonstrated substantial mean squared error reductions, particularly in limited data scenarios (1 %, 5 %, 15 % and 50 % sample sizes), indicating their robustness in precise NDVI temporal predictions when data is limited. The findings in this study demonstrated the effectiveness of emerging machine learning techniques such as reservoir computing and pretrained language model for remote sensing and their contributions in precision agriculture.http://www.sciencedirect.com/science/article/pii/S2589721724000539Temporal predictionNDVIDeep learning (DL)Reservoir computing (RC)Large language model (LLM)GPT2
spellingShingle John Olamofe
Ram Ray
Xishuang Dong
Lijun Qian
Normalized difference vegetation index prediction using reservoir computing and pretrained language models
Artificial Intelligence in Agriculture
Temporal prediction
NDVI
Deep learning (DL)
Reservoir computing (RC)
Large language model (LLM)
GPT2
title Normalized difference vegetation index prediction using reservoir computing and pretrained language models
title_full Normalized difference vegetation index prediction using reservoir computing and pretrained language models
title_fullStr Normalized difference vegetation index prediction using reservoir computing and pretrained language models
title_full_unstemmed Normalized difference vegetation index prediction using reservoir computing and pretrained language models
title_short Normalized difference vegetation index prediction using reservoir computing and pretrained language models
title_sort normalized difference vegetation index prediction using reservoir computing and pretrained language models
topic Temporal prediction
NDVI
Deep learning (DL)
Reservoir computing (RC)
Large language model (LLM)
GPT2
url http://www.sciencedirect.com/science/article/pii/S2589721724000539
work_keys_str_mv AT johnolamofe normalizeddifferencevegetationindexpredictionusingreservoircomputingandpretrainedlanguagemodels
AT ramray normalizeddifferencevegetationindexpredictionusingreservoircomputingandpretrainedlanguagemodels
AT xishuangdong normalizeddifferencevegetationindexpredictionusingreservoircomputingandpretrainedlanguagemodels
AT lijunqian normalizeddifferencevegetationindexpredictionusingreservoircomputingandpretrainedlanguagemodels