Deep learning based multi-temporal crop classification

Zhong, Liheng; Hu, Ling; Zhou, Hang

doi:10.1016/j.rse.2018.11.032

Cited by 693 publications

(433 citation statements)

References 85 publications

Supporting

Mentioning

334

Contrasting

Unclassified

Order By: Relevance

“…As a result, these models are widely used in crop yield forecasts with the help of monthly weather observations and VIs (Becker-Reshef et al 2010, Bolton and Friedl 2013, Sakamoto et al 2014, Peng et al 2018, Li et al 2019b. In recent years, machine learning algorithms, especially deep neural networks, have received increased attention given their ability to describe complex relationships (Yang et al 2019, Zhong et al 2019. Compared to statistical models, machine learning algorithms require no prior assumption about the relationships between response and predictor variables and allow for higher-order interactions, resulting in improved predictive power , Zhang et al 2019.…”

Section: Introductionmentioning

confidence: 99%

Comparative assessment of environmental variables and machine learning algorithms for maize yield prediction in the US Midwest

Kang

Özdoğan

Zhu

et al. 2020

Environ. Res. Lett.

128

View full text Add to dashboard Cite

Crop yield estimates over large areas are conventionally made using weather observations, but a comprehensive understanding of the effects of various environmental indicators, observation frequency, and the choice of prediction algorithm remains elusive. Here we present a thorough assessment of county-level maize yield prediction in U.S. Midwest using six statistical/machine learning algorithms (Lasso, Support Vector Regressor, Random Forest, XGBoost, Long-short term memory (LSTM), and Convolutional Neural Network (CNN)) and an extensive set of environmental variables derived from satellite observations, weather data, land surface model results, soil maps, and crop progress reports. Results show that seasonal crop yield forecasting benefits from both more advanced algorithms and a large composite of information associated with crop canopy, environmental stress, phenology, and soil properties (i.e. hundreds of features). The XGBoost algorithm outperforms other algorithms both in accuracy and stability, while deep neural networks such as LSTM and CNN are not advantageous. The compositing interval (8-day, 16-day or monthly) of time series variable does not have significant effects on the prediction. Combining the best algorithm and inputs improves the prediction accuracy by 5% when compared to a baseline statistical model (Lasso) using only basic climatic and satellite observations. Reasonable county-level yield foresting is achievable from early June, almost four months prior to harvest. At the national level, early-season (June and July) prediction from the best model outperforms that of the United States Department of Agriculture (USDA) World Agricultural Supply and Demand Estimates (WASDE). This study provides insights into practical crop yield forecasting and the understanding of yield response to climatic and environmental conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparative assessment of environmental variables and machine learning algorithms for maize yield prediction in the US Midwest

Kang

Özdoğan

Zhu

et al. 2020

Environ. Res. Lett.

128

View full text Add to dashboard Cite

show abstract

“…2 illustrates the processed Sentinel-2 image of central Munich, Germany, and the reference data. There are two approaches for remote sensing image classification via deep learning: working with either patch-based CNNs designed for image classification [24,26,30,31,32,49,50,51] or encoder-decoder-like neural networks designed for semantic segmentation [25,27,28,29]. The former works under the assumption of just a single label for each image patch, and applies the trained model to the image of a study area via a sliding window approach, with the target GSD as the stride of the sliding window.…”

Section: Sentinel-2 Image Pre-processing and Reference Ground Truth Pmentioning

confidence: 99%

A framework for large-scale mapping of human settlement extent from Sentinel-2 images via fully convolutional neural networks

Qiu

Schmitt

Geiß

et al. 2020

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

This article was submitted to ISPRS Journal of Photogrammetry and Remote Sensing.Human settlement extent (HSE) information is a valuable indicator of worldwide urbanization as well as the resulting human pressure on the natural environment. Therefore, mapping HSE is critical for various environmental issues at local, regional, and even global scales. This paper presents a deep-learning-based framework to automatically map HSE from multi-spectral Sentinel-2 data using regionally available geo-products as training labels. A straightforward, simple, yet effective fully convolutional network-based architecture, Sen2HSE, is implemented as an example for semantic segmentation within the framework. The framework is validated against both manually labelled checking points distributed evenly over the test areas, and * the OpenStreetMap building layer. The HSE mapping results were extensively compared to several baseline products in order to thoroughly evaluate the effectiveness of the proposed HSE mapping framework. The HSE mapping power is consistently demonstrated over 10 representative areas across the world. We also present one regional-scale and one country-wide HSE mapping example from our framework to show the potential for upscaling. The results of this study contribute to the generalization of the applicability of CNN-based approaches for large-scale urban mapping to cases where no up-to-date and accurate ground truth is available, as well as the subsequent monitor of global urbanization.

show abstract

“…Scarce examples on the application of deep learning techniques on remotely sensed time-series can be found out of the GeoAI term scope. Zhong, Liheng et al [42] have exploited the intrinsic characteristics of time-series data to describe seasonal patterns and sequential relationships for classifying summer crops. They developed different deep neural network architectures and used Enhanced Vegetation Index (EVI) calculated from Landsat Level 2 product imagery bands and ground in-situ data from California Department of Water Resources (see Table 1).…”

Section: Geoai Models For Sustainable Agriculturementioning

confidence: 99%

“…Pelletier et al [43] proposed a temporal convolutional neural network constructed with three convolutional layers, a dense layer and finally, a Softmax layer. Different to [42], authors of this study used three spectral bands of the available satellite imagery. Results show that the proposed architecture outperformed Random Forest algorithm by 2 to 3 % and based on the evidence gathered they point out the importance of using both spectral and temporal dimensions when computing the convolutions.…”

Section: Geoai Models For Sustainable Agriculturementioning

confidence: 99%

Data Acquisition and Processing for GeoAI Models to Support Sustainable Agricultural Practices

Ojo¹,

Curry²,

Porwol³

2020

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

There are growing opportunities to leverage new technologies and data sources to address global problems related to sustainability, climate change, and biodiversity loss. The emerging discipline of GeoAI resulting from the convergence of AI and Geospatial science (Geo-AI) is enabling the possibility to harness the increasingly available open Earth Observation data collected from different constellations of satellites and sensors with high spatial, spectral and temporal resolutions. However, transforming these raw data into high-quality datasets that could be used for training AI and specifically deep learning models are technically challenging. This paper describes the process and results of synthesizing labelled-datasets that could be used for training AI (specifically Convolutional Neural Networks) models for determining agricultural land use pattern to support decisions for sustainable farming. In our opinion, this work is a significant step forward in addressing the paucity of usable datasets for developing scalable GeoAI models for sustainable agriculture.

show abstract

Deep learning based multi-temporal crop classification

Cited by 693 publications

References 85 publications

Comparative assessment of environmental variables and machine learning algorithms for maize yield prediction in the US Midwest

Comparative assessment of environmental variables and machine learning algorithms for maize yield prediction in the US Midwest

A framework for large-scale mapping of human settlement extent from Sentinel-2 images via fully convolutional neural networks

Data Acquisition and Processing for GeoAI Models to Support Sustainable Agricultural Practices

Contact Info

Product

Resources

About