Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models
Unmanned aerial vehicle (UAV) based remote sensing is gaining momentum worldwide in a variety of agricultural and environmental monitoring and modelling applications. At the same time, the increasing availability of yield monitoring devices in harvesters enables input-target mapping of in-season RGB...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/23/4000 |
| _version_ | 1797545541277581312 |
|---|---|
| author | Petteri Nevavuori Nathaniel Narra Petri Linna Tarmo Lipping |
| author_facet | Petteri Nevavuori Nathaniel Narra Petri Linna Tarmo Lipping |
| author_sort | Petteri Nevavuori |
| collection | DOAJ |
| description | Unmanned aerial vehicle (UAV) based remote sensing is gaining momentum worldwide in a variety of agricultural and environmental monitoring and modelling applications. At the same time, the increasing availability of yield monitoring devices in harvesters enables input-target mapping of in-season RGB and crop yield data in a resolution otherwise unattainable by openly availabe satellite sensor systems. Using time series UAV RGB and weather data collected from nine crop fields in Pori, Finland, we evaluated the feasibility of spatio-temporal deep learning architectures in crop yield time series modelling and prediction with RGB time series data. Using Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM) networks as spatial and temporal base architectures, we developed and trained CNN-LSTM, convolutional LSTM and 3D-CNN architectures with full 15 week image frame sequences from the whole growing season of 2018. The best performing architecture, the 3D-CNN, was then evaluated with several shorter frame sequence configurations from the beginning of the season. With 3D-CNN, we were able to achieve 218.9 kg/ha mean absolute error (MAE) and 5.51% mean absolute percentage error (MAPE) performance with full length sequences. The best shorter length sequence performance with the same model was 292.8 kg/ha MAE and 7.17% MAPE with four weekly frames from the beginning of the season. |
| first_indexed | 2024-03-10T14:16:55Z |
| format | Article |
| id | doaj.art-c3507527a5e8449e9af5400836ab3a2c |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T14:16:55Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c3507527a5e8449e9af5400836ab3a2c2023-11-20T23:44:27ZengMDPI AGRemote Sensing2072-42922020-12-011223400010.3390/rs12234000Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning ModelsPetteri Nevavuori0Nathaniel Narra1Petri Linna2Tarmo Lipping3Mtech Digital Solutions Oy, 01301 Vantaa, FinlandFaculty of Information Technology and Communication Sciences, Tampere University, 33014 Tampere, FinlandFaculty of Information Technology and Communication Sciences, Tampere University, 33014 Tampere, FinlandFaculty of Information Technology and Communication Sciences, Tampere University, 33014 Tampere, FinlandUnmanned aerial vehicle (UAV) based remote sensing is gaining momentum worldwide in a variety of agricultural and environmental monitoring and modelling applications. At the same time, the increasing availability of yield monitoring devices in harvesters enables input-target mapping of in-season RGB and crop yield data in a resolution otherwise unattainable by openly availabe satellite sensor systems. Using time series UAV RGB and weather data collected from nine crop fields in Pori, Finland, we evaluated the feasibility of spatio-temporal deep learning architectures in crop yield time series modelling and prediction with RGB time series data. Using Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM) networks as spatial and temporal base architectures, we developed and trained CNN-LSTM, convolutional LSTM and 3D-CNN architectures with full 15 week image frame sequences from the whole growing season of 2018. The best performing architecture, the 3D-CNN, was then evaluated with several shorter frame sequence configurations from the beginning of the season. With 3D-CNN, we were able to achieve 218.9 kg/ha mean absolute error (MAE) and 5.51% mean absolute percentage error (MAPE) performance with full length sequences. The best shorter length sequence performance with the same model was 292.8 kg/ha MAE and 7.17% MAPE with four weekly frames from the beginning of the season.https://www.mdpi.com/2072-4292/12/23/4000crop yield predictionUAVspatio-temporal modellingtime seriesdeep learningcnn-lstm |
| spellingShingle | Petteri Nevavuori Nathaniel Narra Petri Linna Tarmo Lipping Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models Remote Sensing crop yield prediction UAV spatio-temporal modelling time series deep learning cnn-lstm |
| title | Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models |
| title_full | Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models |
| title_fullStr | Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models |
| title_full_unstemmed | Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models |
| title_short | Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models |
| title_sort | crop yield prediction using multitemporal uav data and spatio temporal deep learning models |
| topic | crop yield prediction UAV spatio-temporal modelling time series deep learning cnn-lstm |
| url | https://www.mdpi.com/2072-4292/12/23/4000 |
| work_keys_str_mv | AT petterinevavuori cropyieldpredictionusingmultitemporaluavdataandspatiotemporaldeeplearningmodels AT nathanielnarra cropyieldpredictionusingmultitemporaluavdataandspatiotemporaldeeplearningmodels AT petrilinna cropyieldpredictionusingmultitemporaluavdataandspatiotemporaldeeplearningmodels AT tarmolipping cropyieldpredictionusingmultitemporaluavdataandspatiotemporaldeeplearningmodels |