Predicting Crop Yield With Machine Learning: An Extensive Analysis Of Input Modalities And Models On a Field and sub-field Level

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pathak, Deepak, Miranda, Miro, Mena, Francisco, Sanchez, Cristhian, Helber, Patrick, Bischke, Benjamin, Habelitz, Peter, Najjar, Hiba, Siddamsetty, Jayanth, Arenas, Diego, Vollmer, Michaela, Charfuelan, Marcela, Nuske, Marlon, Dengel, Andreas
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916080870490112
author Pathak, Deepak
Miranda, Miro
Mena, Francisco
Sanchez, Cristhian
Helber, Patrick
Bischke, Benjamin
Habelitz, Peter
Najjar, Hiba
Siddamsetty, Jayanth
Arenas, Diego
Vollmer, Michaela
Charfuelan, Marcela
Nuske, Marlon
Dengel, Andreas
author_facet Pathak, Deepak
Miranda, Miro
Mena, Francisco
Sanchez, Cristhian
Helber, Patrick
Bischke, Benjamin
Habelitz, Peter
Najjar, Hiba
Siddamsetty, Jayanth
Arenas, Diego
Vollmer, Michaela
Charfuelan, Marcela
Nuske, Marlon
Dengel, Andreas
contents We introduce a simple yet effective early fusion method for crop yield prediction that handles multiple input modalities with different temporal and spatial resolutions. We use high-resolution crop yield maps as ground truth data to train crop and machine learning model agnostic methods at the sub-field level. We use Sentinel-2 satellite imagery as the primary modality for input data with other complementary modalities, including weather, soil, and DEM data. The proposed method uses input modalities available with global coverage, making the framework globally scalable. We explicitly highlight the importance of input modalities for crop yield prediction and emphasize that the best-performing combination of input modalities depends on region, crop, and chosen model.
format Preprint
id arxiv_https___arxiv_org_abs_2308_08948
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Predicting Crop Yield With Machine Learning: An Extensive Analysis Of Input Modalities And Models On a Field and sub-field Level
Pathak, Deepak
Miranda, Miro
Mena, Francisco
Sanchez, Cristhian
Helber, Patrick
Bischke, Benjamin
Habelitz, Peter
Najjar, Hiba
Siddamsetty, Jayanth
Arenas, Diego
Vollmer, Michaela
Charfuelan, Marcela
Nuske, Marlon
Dengel, Andreas
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
ACM-class: J.2
We introduce a simple yet effective early fusion method for crop yield prediction that handles multiple input modalities with different temporal and spatial resolutions. We use high-resolution crop yield maps as ground truth data to train crop and machine learning model agnostic methods at the sub-field level. We use Sentinel-2 satellite imagery as the primary modality for input data with other complementary modalities, including weather, soil, and DEM data. The proposed method uses input modalities available with global coverage, making the framework globally scalable. We explicitly highlight the importance of input modalities for crop yield prediction and emphasize that the best-performing combination of input modalities depends on region, crop, and chosen model.
title Predicting Crop Yield With Machine Learning: An Extensive Analysis Of Input Modalities And Models On a Field and sub-field Level
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
ACM-class: J.2
url https://arxiv.org/abs/2308.08948