-
Prediction of Sentinel-2 multi-band imagery with attention BiLSTM for continuous earth surface monitoring
Authors:
Weiying Zhao,
Natalia Efremova
Abstract:
Continuous monitoring of crops and forecasting crop conditions through time series analysis is crucial for effective agricultural management. This study proposes a framework based on an attention Bidirectional Long Short-Term Memory (BiLSTM) network for predicting multiband images. Our model can forecast target images on user-defined dates, including future dates and periods characterized by persi…
▽ More
Continuous monitoring of crops and forecasting crop conditions through time series analysis is crucial for effective agricultural management. This study proposes a framework based on an attention Bidirectional Long Short-Term Memory (BiLSTM) network for predicting multiband images. Our model can forecast target images on user-defined dates, including future dates and periods characterized by persistent cloud cover. By focusing on short sequences within a sequence-to-one forecasting framework, the model leverages advanced attention mechanisms to enhance prediction accuracy. Our experimental results demonstrate the model's superior performance in predicting NDVI, multiple vegetation indices, and all Sentinel-2 bands, highlighting its potential for improving remote sensing data continuity and reliability.
△ Less
Submitted 30 June, 2024;
originally announced July 2024.
-
Soil nitrogen forecasting from environmental variables provided by multisensor remote sensing images
Authors:
Weiying Zhao,
Ganzorig Chuluunbat,
Aleksei Unagaev,
Natalia Efremova
Abstract:
This study introduces a framework for forecasting soil nitrogen content, leveraging multi-modal data, including multi-sensor remote sensing images and advanced machine learning methods. We integrate the Land Use/Land Cover Area Frame Survey (LUCAS) database, which covers European and UK territory, with environmental variables from satellite sensors to create a dataset of novel features. We further…
▽ More
This study introduces a framework for forecasting soil nitrogen content, leveraging multi-modal data, including multi-sensor remote sensing images and advanced machine learning methods. We integrate the Land Use/Land Cover Area Frame Survey (LUCAS) database, which covers European and UK territory, with environmental variables from satellite sensors to create a dataset of novel features. We further test a broad range of machine learning algorithms, focusing on tree-based models such as CatBoost, LightGBM, and XGBoost. We test the proposed methods with a variety of land cover classes, including croplands and grasslands to ensure the robustness of this approach. Our results demonstrate that the CatBoost model surpasses other methods in accuracy. This research advances the field of agricultural management and environmental monitoring and demonstrates the significant potential of integrating multisensor remote sensing data with machine learning for environmental analysis.
△ Less
Submitted 14 June, 2024;
originally announced June 2024.
-
Grapevine Disease Prediction Using Climate Variables from Multi-Sensor Remote Sensing Imagery via a Transformer Model
Authors:
Weiying Zhao,
Natalia Efremova
Abstract:
Early detection and management of grapevine diseases are important in pursuing sustainable viticulture. This paper introduces a novel framework leveraging the TabPFN model to forecast blockwise grapevine diseases using climate variables from multi-sensor remote sensing imagery. By integrating advanced machine learning techniques with detailed environmental data, our approach significantly enhances…
▽ More
Early detection and management of grapevine diseases are important in pursuing sustainable viticulture. This paper introduces a novel framework leveraging the TabPFN model to forecast blockwise grapevine diseases using climate variables from multi-sensor remote sensing imagery. By integrating advanced machine learning techniques with detailed environmental data, our approach significantly enhances the accuracy and efficiency of disease prediction in vineyards. The TabPFN model's experimental evaluations showcase comparable performance to traditional gradient-boosted decision trees, such as XGBoost, CatBoost, and LightGBM. The model's capability to process complex data and provide per-pixel disease-affecting probabilities enables precise, targeted interventions, contributing to more sustainable disease management practices. Our findings underscore the transformative potential of combining Transformer models with remote sensing data in precision agriculture, offering a scalable solution for improving crop health and productivity while reducing environmental impact.
△ Less
Submitted 11 June, 2024;
originally announced June 2024.
-
Precision Agriculture: Crop Map** using Machine Learning and Sentinel-2 Satellite Imagery
Authors:
Kui Zhao,
Siyang Wu,
Chang Liu,
Yue Wu,
Natalia Efremova
Abstract:
Food security has grown in significance due to the changing climate and its warming effects. To support the rising demand for agricultural products and to minimize the negative impact of climate change and mass cultivation, precision agriculture has become increasingly important for crop cultivation. This study employs deep learning and pixel-based machine learning methods to accurately segment la…
▽ More
Food security has grown in significance due to the changing climate and its warming effects. To support the rising demand for agricultural products and to minimize the negative impact of climate change and mass cultivation, precision agriculture has become increasingly important for crop cultivation. This study employs deep learning and pixel-based machine learning methods to accurately segment lavender fields for precision agriculture, utilizing various spectral band combinations extracted from Sentinel-2 satellite imagery. Our fine-tuned final model, a U-Net architecture, can achieve a Dice coefficient of 0.8324. Additionally, our investigation highlights the unexpected efficacy of the pixel-based method and the RGB spectral band combination in this task.
△ Less
Submitted 25 November, 2023;
originally announced March 2024.
-
Soil Organic Carbon Estimation from Climate-related Features with Graph Neural Network
Authors:
Weiying Zhao,
Natalia Efremova
Abstract:
Soil organic carbon (SOC) plays a pivotal role in the global carbon cycle, impacting climate dynamics and necessitating accurate estimation for sustainable land and agricultural management. While traditional methods of SOC estimation face resolution and accuracy challenges, recent technological solutions harness remote sensing, machine learning, and high-resolution satellite map**. Graph Neural…
▽ More
Soil organic carbon (SOC) plays a pivotal role in the global carbon cycle, impacting climate dynamics and necessitating accurate estimation for sustainable land and agricultural management. While traditional methods of SOC estimation face resolution and accuracy challenges, recent technological solutions harness remote sensing, machine learning, and high-resolution satellite map**. Graph Neural Networks (GNNs), especially when integrated with positional encoders, can capture complex relationships between soil and climate. Using the LUCAS database, this study compared four GNN operators in the positional encoder framework. Results revealed that the PESAGE and PETransformer models outperformed others in SOC estimation, indicating their potential in capturing the complex relationship between SOC and climate features. Our findings confirm the feasibility of applications of GNN architectures in SOC prediction, establishing a framework for future explorations of this topic with more advanced GNN models.
△ Less
Submitted 27 November, 2023;
originally announced November 2023.
-
AI for Agriculture: the Comparison of Semantic Segmentation Methods for Crop Map** with Sentinel-2 Imagery
Authors:
Irina Korotkova,
Natalia Efremova
Abstract:
Crop map** is one of the most common tasks in artificial intelligence for agriculture due to higher food demands from a growing population and increased awareness of climate change. In case of vineyards, the texture is very important for crop segmentation: with higher resolution satellite imagery the texture is easily detected by majority of state-of-the-art algorithms. However, this task become…
▽ More
Crop map** is one of the most common tasks in artificial intelligence for agriculture due to higher food demands from a growing population and increased awareness of climate change. In case of vineyards, the texture is very important for crop segmentation: with higher resolution satellite imagery the texture is easily detected by majority of state-of-the-art algorithms. However, this task becomes increasingly more difficult as the resolution of satellite imagery decreases and the information about the texture becomes unavailable. In this paper we aim to explore the main machine learning methods that can be used with freely available satellite imagery and discuss how and when they can be applied for vineyard segmentation problem. We assess the effectiveness of various widely-used machine learning techniques and offer guidance on selecting the most suitable model for specific scenarios.
△ Less
Submitted 21 November, 2023;
originally announced November 2023.
-
Optimal Use of Multi-spectral Satellite Data with Convolutional Neural Networks
Authors:
Sagar Vaze,
James Foley,
Mohamed Seddiq,
Alexey Unagaev,
Natalia Efremova
Abstract:
The analysis of satellite imagery will prove a crucial tool in the pursuit of sustainable development. While Convolutional Neural Networks (CNNs) have made large gains in natural image analysis, their application to multi-spectral satellite images (wherein input images have a large number of channels) remains relatively unexplored. In this paper, we compare different methods of leveraging multi-ba…
▽ More
The analysis of satellite imagery will prove a crucial tool in the pursuit of sustainable development. While Convolutional Neural Networks (CNNs) have made large gains in natural image analysis, their application to multi-spectral satellite images (wherein input images have a large number of channels) remains relatively unexplored. In this paper, we compare different methods of leveraging multi-band information with CNNs, demonstrating the performance of all compared methods on the task of semantic segmentation of agricultural vegetation (vineyards). We show that standard industry practice of using bands selected by a domain expert leads to a significantly worse test accuracy than the other methods compared. Specifically, we compare: using bands specified by an expert; using all available bands; learning attention maps over the input bands; and leveraging Bayesian optimisation to dictate band choice. We show that simply using all available band information already increases test time performance, and show that the Bayesian optimisation, first applied to band selection in this work, can be used to further boost accuracy.
△ Less
Submitted 15 September, 2020;
originally announced September 2020.
-
SMArtCast: Predicting soil moisture interpolations into the future using Earth observation data in a deep learning framework
Authors:
Conrad James Foley,
Sagar Vaze,
Mohamed El Amine Seddiq,
Alexey Unagaev,
Natalia Efremova
Abstract:
Soil moisture is critical component of crop health and monitoring it can enable further actions for increasing yield or preventing catastrophic die off. As climate change increases the likelihood of extreme weather events and reduces the predictability of weather, and non-optimal soil moistures for crops may become more likely. In this work, we a series of LSTM architectures to analyze measurement…
▽ More
Soil moisture is critical component of crop health and monitoring it can enable further actions for increasing yield or preventing catastrophic die off. As climate change increases the likelihood of extreme weather events and reduces the predictability of weather, and non-optimal soil moistures for crops may become more likely. In this work, we a series of LSTM architectures to analyze measurements of soil moisture and vegetation indiced derived from satellite imagery. The system learns to predict the future values of these measurements. These spatially sparse values and indices are used as input features to an interpolation method that infer spatially dense moisture map for a future time point. This has the potential to provide advance warning for soil moistures that may be inhospitable to crops across an area with limited monitoring capacity.
△ Less
Submitted 24 April, 2020; v1 submitted 16 March, 2020;
originally announced March 2020.
-
Prediction of Soil Moisture Content Based On Satellite Data and Sequence-to-Sequence Networks
Authors:
Natalia Efremova,
Dmitry Zausaev,
Gleb Antipov
Abstract:
The main objective of this study is to combine remote sensing and machine learning to detect soil moisture content. Growing population and food consumption has led to the need to improve agricultural yield and to reduce wastage of natural resources. In this paper, we propose a neural network architecture, based on recent work by the research community, that can make a strong social impact and aid…
▽ More
The main objective of this study is to combine remote sensing and machine learning to detect soil moisture content. Growing population and food consumption has led to the need to improve agricultural yield and to reduce wastage of natural resources. In this paper, we propose a neural network architecture, based on recent work by the research community, that can make a strong social impact and aid United Nations Sustainable Development Goal of Zero Hunger. The main aims here are to: improve efficiency of water usage; reduce dependence on irrigation; increase overall crop yield; minimise risk of crop loss due to drought and extreme weather conditions. We achieve this by applying satellite imagery, crop segmentation, soil classification and NDVI and soil moisture prediction on satellite data, ground truth and climate data records. By applying machine learning to sensor data and ground data, farm management systems can evolve into a real time AI enabled platform that can provide actionable recommendations and decision support tools to the farmers.
△ Less
Submitted 5 June, 2019;
originally announced July 2019.
-
AI-based evaluation of the SDGs: The case of crop detection with earth observation data
Authors:
Natalia Efremova,
Dennis West,
Dmitry Zausaev
Abstract:
The framework of the seventeen sustainable development goals is a challenge for developers and researchers applying artificial intelligence (AI). AI and earth observations (EO) can provide reliable and disaggregated data for better monitoring of the sustainable development goals (SDGs). In this paper, we present an overview of SDG targets, which can be effectively measured with AI tools. We identi…
▽ More
The framework of the seventeen sustainable development goals is a challenge for developers and researchers applying artificial intelligence (AI). AI and earth observations (EO) can provide reliable and disaggregated data for better monitoring of the sustainable development goals (SDGs). In this paper, we present an overview of SDG targets, which can be effectively measured with AI tools. We identify indicators with the most significant contribution from the AI and EO and describe an application of state-of-the-art machine learning models to one of the indicators. We describe an application of U-net with SE blocks for efficient segmentation of satellite imagery for crop detection. Finally, we demonstrate how AI can be more effectively applied in solutions directly contributing towards specific SDGs and propose further research on an AI-based evaluative infrastructure for SDGs.
△ Less
Submitted 5 July, 2019;
originally announced July 2019.
-
Convolutional neural networks pretrained on large face recognition datasets for emotion classification from video
Authors:
Boris Knyazev,
Roman Shvetsov,
Natalia Efremova,
Artem Kuharenko
Abstract:
In this paper we describe a solution to our entry for the emotion recognition challenge EmotiW 2017. We propose an ensemble of several models, which capture spatial and audio features from videos. Spatial features are captured by convolutional neural networks, pretrained on large face recognition datasets. We show that usage of strong industry-level face recognition networks increases the accuracy…
▽ More
In this paper we describe a solution to our entry for the emotion recognition challenge EmotiW 2017. We propose an ensemble of several models, which capture spatial and audio features from videos. Spatial features are captured by convolutional neural networks, pretrained on large face recognition datasets. We show that usage of strong industry-level face recognition networks increases the accuracy of emotion recognition. Using our ensemble we improve on the previous best result on the test set by about 1 %, achieving a 60.03 % classification accuracy without any use of visual temporal information.
△ Less
Submitted 13 November, 2017;
originally announced November 2017.
-
Unsupervised Neural Architecture for Saliency Detection: Extended Version
Authors:
Natalia Efremova,
Sergey Tarasenko
Abstract:
We propose a novel neural network architecture for visual saliency detections, which utilizes neurophysiologically plausible mechanisms for extraction of salient regions. The model has been significantly inspired by recent findings from neurophysiology and aimed to simulate the bottom-up processes of human selective attention. Two types of features were analyzed: color and direction of maximum var…
▽ More
We propose a novel neural network architecture for visual saliency detections, which utilizes neurophysiologically plausible mechanisms for extraction of salient regions. The model has been significantly inspired by recent findings from neurophysiology and aimed to simulate the bottom-up processes of human selective attention. Two types of features were analyzed: color and direction of maximum variance. The mechanism we employ for processing those features is PCA, implemented by means of normalized Hebbian learning and the waves of spikes. To evaluate performance of our model we have conducted psychological experiment. Comparison of simulation results with those of experiment indicates good performance of our model.
△ Less
Submitted 10 April, 2015; v1 submitted 18 November, 2014;
originally announced December 2014.
