Explainability of Sub-Field Level Crop Yield Prediction using Remote Sensing

• URL: http://arxiv.org/abs/2407.08274v1
• Date: Thu, 11 Jul 2024 08:23:46 GMT
• Title: Explainability of Sub-Field Level Crop Yield Prediction using Remote Sensing
• Authors: Hiba Najjar, Miro Miranda, Marlon Nuske, Ribana Roscher, Andreas Dengel,
• Abstract summary: We focus on the task of crop yield prediction, specifically for soybean, wheat, and rapeseed crops in Argentina, Uruguay, and Germany. Our goal is to develop and explain predictive models for these crops, using a large dataset of satellite images, additional data modalities, and crop yield maps. For model explainability, we utilize feature attribution methods to quantify input feature contributions, identify critical growth stages, analyze yield variability at the field level, and explain less accurate predictions.
• Score: 6.65506917941232
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Crop yield forecasting plays a significant role in addressing growing concerns about food security and guiding decision-making for policymakers and farmers. When deep learning is employed, understanding the learning and decision-making processes of the models, as well as their interaction with the input data, is crucial for establishing trust in the models and gaining insight into their reliability. In this study, we focus on the task of crop yield prediction, specifically for soybean, wheat, and rapeseed crops in Argentina, Uruguay, and Germany. Our goal is to develop and explain predictive models for these crops, using a large dataset of satellite images, additional data modalities, and crop yield maps. We employ a long short-term memory network and investigate the impact of using different temporal samplings of the satellite data and the benefit of adding more relevant modalities. For model explainability, we utilize feature attribution methods to quantify input feature contributions, identify critical growth stages, analyze yield variability at the field level, and explain less accurate predictions. The modeling results show an improvement when adding more modalities or using all available instances of satellite data. The explainability results reveal distinct feature importance patterns for each crop and region. We further found that the most influential growth stages on the prediction are dependent on the temporal sampling of the input data. We demonstrated how these critical growth stages, which hold significant agronomic value, closely align with the existing literature in agronomy and crop development biology.

Related papers

• Anticipatory Understanding of Resilient Agriculture to Climate [66.008020515555]
  We present a framework to better identify food security hotspots using a combination of remote sensing, deep learning, crop yield modeling, and causal modeling of the food distribution system. We focus our analysis on the wheat breadbasket of northern India, which supplies a large percentage of the world's population.
  arXiv  Detail & Related papers  (2024-11-07T22:29:05Z)
• A Novel Fusion of Optical and Radar Satellite Data for Crop Phenology Estimation using Machine Learning and Cloud Computing [0.0]
  In the era of big Earth observation data ubiquity, attempts have been made to accurately predict crop phenology based on Remote Sensing data. Here, we estimate phenological developments for eight major crops and 13 phenological stages across Germany at 30m scale using a novel framework.
  arXiv  Detail & Related papers  (2024-08-16T13:44:35Z)
• VegeDiff: Latent Diffusion Model for Geospatial Vegetation Forecasting [58.12667617617306]
  We propose VegeDiff for the geospatial vegetation forecasting task. VegeDiff is the first to employ a diffusion model to probabilistically capture the uncertainties in vegetation change processes. By capturing the uncertainties in vegetation changes and modeling the complex influence of relevant variables, VegeDiff outperforms existing deterministic methods.
  arXiv  Detail & Related papers  (2024-07-17T14:15:52Z)
• Naïve Bayes and Random Forest for Crop Yield Prediction [0.0]
  This study analyzes crop yield prediction in India from 1997 to 2020, focusing on various crops and key environmental factors. It aims to predict agricultural yields by utilizing advanced machine learning techniques like Linear Regression, Decision Tree, KNN, Na"ive Bayes, K-Mean Clustering, and Random Forest.
  arXiv  Detail & Related papers  (2024-04-23T16:55:45Z)
• Machine Learning-based Nutrient Application's Timeline Recommendation for Smart Agriculture: A Large-Scale Data Mining Approach [0.0]
  Inaccurate fertiliser application decisions can lead to costly consequences, hinder food production, and cause environmental harm. We propose a solution to predict nutrient application by determining required fertiliser quantities for an entire season. The proposed solution recommends adjusting fertiliser amounts based on weather conditions and soil characteristics to promote cost-effective and environmentally friendly agriculture.
  arXiv  Detail & Related papers  (2023-10-18T15:37:19Z)
• Agave crop segmentation and maturity classification with deep learning data-centric strategies using very high-resolution satellite imagery [101.18253437732933]
  We present an Agave tequilana Weber azul crop segmentation and maturity classification using very high resolution satellite imagery. We solve real-world deep learning problems in the very specific context of agave crop segmentation. With the resulting accurate models, agave production forecasting can be made available for large regions.
  arXiv  Detail & Related papers  (2023-03-21T03:15:29Z)
• Exploring Self-Attention for Crop-type Classification Explainability [15.822486263693355]
  We introduce a novel explainability framework that aims to shed a light on the essential crop disambiguation patterns learned by a state-of-the-art transformer encoder model. We also present a sensitivity analysis approach to understand better the attention capability for revealing crop-specific phenological events.
  arXiv  Detail & Related papers  (2022-10-24T12:36:40Z)
• Extreme Gradient Boosting for Yield Estimation compared with Deep Learning Approaches [0.0]
  We propose a pipeline to process remote sensing images into feature-based representations that allow the employment of Extreme Gradient Boosting (XGBoost) for yield prediction. A comparative evaluation of soybean yield prediction within the United States shows promising prediction accuracies compared to state-of-the-art yield prediction systems based on Deep Learning.
  arXiv  Detail & Related papers  (2022-08-26T12:48:18Z)
• End-to-end deep learning for directly estimating grape yield from ground-based imagery [53.086864957064876]
  This study demonstrates the application of proximal imaging combined with deep learning for yield estimation in vineyards. Three model architectures were tested: object detection, CNN regression, and transformer models. The study showed the applicability of proximal imaging and deep learning for prediction of grapevine yield on a large scale.
  arXiv  Detail & Related papers  (2022-08-04T01:34:46Z)
• Using satellite imagery to understand and promote sustainable development [87.72561825617062]
  We synthesize the growing literature that uses satellite imagery to understand sustainable development outcomes. We quantify the paucity of ground data on key human-related outcomes and the growing abundance and resolution of satellite imagery. We review recent machine learning approaches to model-building in the context of scarce and noisy training data.
  arXiv  Detail & Related papers  (2020-09-23T05:20:00Z)
• Learning from Data to Optimize Control in Precision Farming [77.34726150561087]
  Special issue presents the latest development in statistical inference, machine learning and optimum control for precision farming. Satellite positioning and navigation followed by Internet-of-Things generate vast information that can be used to optimize farming processes in real-time.
  arXiv  Detail & Related papers  (2020-07-07T12:44:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.