Near real-time monitoring of cassava cultivation area

Remote sensing technologies and deep learning/machine learning approaches play valuable roles in crop inventory, yield estimation, cultivated area estimation, and crop status monitoring. Satellite-based remote sensing has led to increased spatial and temporal resolution, leading to a better quality...

Full description

Bibliographic Details
Main Authors: Phan, Trong Van, Reymondin, Louis, Vantalon, Thibaud, Delaquis, Erik, Nguyen, Thuy Thanh, Mienmany, Bandit
Format: Conference Paper
Language:Inglés
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10568/127367
_version_ 1855525526295281664
author Phan, Trong Van
Reymondin, Louis
Vantalon, Thibaud
Delaquis, Erik
Nguyen, Thuy Thanh
Mienmany, Bandit
author_browse Delaquis, Erik
Mienmany, Bandit
Nguyen, Thuy Thanh
Phan, Trong Van
Reymondin, Louis
Vantalon, Thibaud
author_facet Phan, Trong Van
Reymondin, Louis
Vantalon, Thibaud
Delaquis, Erik
Nguyen, Thuy Thanh
Mienmany, Bandit
author_sort Phan, Trong Van
collection Repository of Agricultural Research Outputs (CGSpace)
description Remote sensing technologies and deep learning/machine learning approaches play valuable roles in crop inventory, yield estimation, cultivated area estimation, and crop status monitoring. Satellite-based remote sensing has led to increased spatial and temporal resolution, leading to a better quality of land-cover mapping (greater precision, and detail in the number of land cover classes). In this work, we propose to use a long short-term memory neural network (LSTM), an advanced technical model adapted from artificial neural networks (ANN) to estimate cassava cultivation area in southern Laos. LSTM is a modified version of a Recurrent Neural Network (RNN) that uses internal memory to store the information received prior to a given time. This property of LSTMs makes them advantageous for time series regression. We employ Landsat-7/8 and Sentinel-2 time-series datasets and crop phenology information to identify and classify cassava fields using multi-sources remote sensing time-series in a highly fragmented landscape. The results indicate an overall accuracy of > 89% for cassava and > 84% for all-class (barren, bush/grassland, cassava, coffee, forest, seasonal, and water) validating the feasibility of the proposed method. This study demonstrates the potential of LSTM approaches for crop classification using multi-temporal, multi-sources remote sensing time series.
format Conference Paper
id CGSpace127367
institution CGIAR Consortium
language Inglés
publishDate 2022
publishDateRange 2022
publishDateSort 2022
record_format dspace
spelling CGSpace1273672025-11-05T11:41:59Z Near real-time monitoring of cassava cultivation area Phan, Trong Van Reymondin, Louis Vantalon, Thibaud Delaquis, Erik Nguyen, Thuy Thanh Mienmany, Bandit cassava forest cover remote sensing machine learning conservación de la naturaleza satélites de observación terrestre mandioca-yuca Remote sensing technologies and deep learning/machine learning approaches play valuable roles in crop inventory, yield estimation, cultivated area estimation, and crop status monitoring. Satellite-based remote sensing has led to increased spatial and temporal resolution, leading to a better quality of land-cover mapping (greater precision, and detail in the number of land cover classes). In this work, we propose to use a long short-term memory neural network (LSTM), an advanced technical model adapted from artificial neural networks (ANN) to estimate cassava cultivation area in southern Laos. LSTM is a modified version of a Recurrent Neural Network (RNN) that uses internal memory to store the information received prior to a given time. This property of LSTMs makes them advantageous for time series regression. We employ Landsat-7/8 and Sentinel-2 time-series datasets and crop phenology information to identify and classify cassava fields using multi-sources remote sensing time-series in a highly fragmented landscape. The results indicate an overall accuracy of > 89% for cassava and > 84% for all-class (barren, bush/grassland, cassava, coffee, forest, seasonal, and water) validating the feasibility of the proposed method. This study demonstrates the potential of LSTM approaches for crop classification using multi-temporal, multi-sources remote sensing time series. 2022-11-26 2023-01-18T09:45:28Z 2023-01-18T09:45:28Z Conference Paper https://hdl.handle.net/10568/127367 en Open Access application/pdf Phan, T.V.; Reymondin, L.; Vantalon, T.; Delaquis, E.; Nguyen, T.T.; Mienmany, B. (2022) Near real-time monitoring of cassava cultivation area. Asian Federation for Information Technology in Agriculture (AFITA) conference 2022, 6th edition: Promoting Smart Technologies for Sustainable Agriculture. 9 p.
spellingShingle cassava
forest cover
remote sensing
machine learning
conservación de la naturaleza
satélites de observación terrestre
mandioca-yuca
Phan, Trong Van
Reymondin, Louis
Vantalon, Thibaud
Delaquis, Erik
Nguyen, Thuy Thanh
Mienmany, Bandit
Near real-time monitoring of cassava cultivation area
title Near real-time monitoring of cassava cultivation area
title_full Near real-time monitoring of cassava cultivation area
title_fullStr Near real-time monitoring of cassava cultivation area
title_full_unstemmed Near real-time monitoring of cassava cultivation area
title_short Near real-time monitoring of cassava cultivation area
title_sort near real time monitoring of cassava cultivation area
topic cassava
forest cover
remote sensing
machine learning
conservación de la naturaleza
satélites de observación terrestre
mandioca-yuca
url https://hdl.handle.net/10568/127367
work_keys_str_mv AT phantrongvan nearrealtimemonitoringofcassavacultivationarea
AT reymondinlouis nearrealtimemonitoringofcassavacultivationarea
AT vantalonthibaud nearrealtimemonitoringofcassavacultivationarea
AT delaquiserik nearrealtimemonitoringofcassavacultivationarea
AT nguyenthuythanh nearrealtimemonitoringofcassavacultivationarea
AT mienmanybandit nearrealtimemonitoringofcassavacultivationarea