4.7 Article

Improvement in crop mapping from satellite image time series by effectively supervising deep neural networks

Related references

Note: Only part of the references are listed.
Article Geography, Physical

Multi-modal temporal attention models for crop mapping from satellite time series

Vivien Sainte Fare Garnot et al.

Summary: Optical and radar satellite time series can work synergistically, with optical images providing rich spectral information and C band radar capturing useful geometric information while being immune to cloud cover. This study investigates the adaptation of temporal attention-based models to operate on multiple modalities, proposing and evaluating various fusion schemes. The results show that multimodal temporal attention-based models, leveraging both optical and radar time series, outperform single modality models in terms of performance and resilience to cloud cover.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2022)

Article Environmental Sciences

Early- and in-season crop type mapping without current-year ground truth: Generating labels from historical information via a topology-based approach

Chenxi Lin et al.

Summary: This study proposes a new approach for land cover classification using remote sensing, which utilizes historical knowledge about crop types to generate labels in different years, resulting in high-quality crop maps with high accuracy. The approach successfully transfers topology information to support crop classification, regardless of inter-annual weather and management variations.

REMOTE SENSING OF ENVIRONMENT (2022)

Article Geography, Physical

Mapping corn dynamics using limited but representative samples with adaptive strategies

Yanan Wen et al.

Summary: Mapping corn dynamics at a large scale is crucial for global food security. This study developed an efficient approach to map corn dynamics in the main corn districts of the United States using adaptive strategies for collecting high-quality training samples. The proposed approach demonstrated robust classification models with good performance in mapping large-scale corn dynamics.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2022)

Article Geography, Physical

UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery

Libo Wang et al.

Summary: This paper discusses the importance of semantic segmentation of remotely sensed urban scene images in practical applications, and highlights the advantages of using Transformer and the proposed UNetFormer model for real-time segmentation. Experimental results demonstrate that UNetFormer achieves faster inference speed and higher accuracy compared to state-of-the-art lightweight models.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2022)

Article Environmental Sciences

Cloud Removal for Optical Remote Sensing Imagery Using Distortion Coding Network Combined with Compound Loss Functions

Jianjun Zhou et al.

Summary: This paper proposes a cloud removal framework based on generative adversarial networks (GAN), which removes the interference of clouds on optical remote sensing images through distortion coding and feature refinement mechanisms. By considering coherent semantics and local adaptive reconstruction factors in the loss functions, the consistency of features and pixels is achieved. The experimental results show that the proposed method has good performance in terms of peak signal-to-noise ratio (PSNR) and visual perception.

REMOTE SENSING (2022)

Article Multidisciplinary Sciences

Evidence of sweet corn yield losses from rising temperatures

Daljeet S. Dhaliwal et al.

Summary: This study analyzed a large amount of data collected over a 27-year period and found that sweet corn yield is highly sensitive to changes in temperature and precipitation. Particularly, high temperatures have a significant negative impact on crop yield. The study provides evidence for yield losses in sweet corn due to climate change and emphasizes the urgency to accelerate crop adaptation strategies.

SCIENTIFIC REPORTS (2022)

Article Remote Sensing

Crop-specific phenomapping by fusing Landsat and Sentinel data with MODIS time series

Jonas Schreier et al.

Summary: Agricultural production and food security rely heavily on crop growth and condition throughout the growing season. Remote sensing technology provides timely and accurate crop phenology information for informed decision making and agricultural management. By integrating remote sensing data with diverse spatiotemporal resolution, high-precision crop phenometrics can be obtained for intra-field level assessment of crops.

EUROPEAN JOURNAL OF REMOTE SENSING (2021)

Article Geography, Physical

SceneNet: Remote sensing scene classification deep learning network using multi-objective neural evolution architecture search

Ailong Ma et al.

Summary: This study introduces a scene classification network architecture search method based on multi-objective neural evolution, which allows for more flexible extraction of remote sensing image scene information and achieves effective results in balancing computational complexity and performance error.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Article Geography, Physical

PBNet: Part-based convolutional neural network for complex composite object detection in remote sensing imagery

Xian Sun et al.

Summary: This paper validates the limitations of existing methods in detecting composite objects in remote sensing images and proposes a unified part-based convolutional neural network (PBNet) specifically designed for this task. Experiments on two datasets show that PBNet outperforms existing detection algorithms and achieves state-of-the-art accuracy.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Article Environmental Sciences

Crop mapping from image time series: Deep learning with multi-scale label hierarchies

Mehmet Ozgur Turkoglu et al.

Summary: This paper introduces a method for agricultural crop classification based on expert knowledge, using a hierarchical neural network model. Validated on a new large dataset, the hierarchical model shows at least 9.9 percentage points superiority in F1-score compared to baseline methods.

REMOTE SENSING OF ENVIRONMENT (2021)

Article Environmental Sciences

Towards interpreting multi-temporal deep learning models in crop mapping

Jinfan Xu et al.

Summary: Multi-temporal deep learning methods show promising performance in large-scale crop mapping by transforming remote sensing data into high-dimensional features for crop identification. The study demonstrates the importance of complete time series input, key growth periods, and critical bands for crop discrimination. By analyzing input feature importance and hidden features, deep learning models extract refined information for effective crop classification.

REMOTE SENSING OF ENVIRONMENT (2021)

Article Environmental Sciences

Nutritional quality of crops in a high CO2 world: an agenda for research and technology development

Kristie L. Ebi et al.

Summary: Essential nutrients are crucial for human health and development, with inadequate intake having negative impacts. Rising CO2 concentrations can affect the nutritional quality of plants, potentially impacting human health. Bridging data and knowledge gaps is key to addressing critical challenges in nutrition and food security.

ENVIRONMENTAL RESEARCH LETTERS (2021)

Article Geography, Physical

Mapping crop types in complex farming areas using SAR imagery with dynamic time warping

Getachew Workineh Gella et al.

Summary: This study successfully mapped crop types in fragmented agricultural landscapes using a combination of Synthetic Aperture Radar (SAR) and crop phenological information with different Dynamic Time Warping implementation strategies, demonstrating the potential for crop type mapping in smallholder farming areas.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Article Geography, Physical

Video object detection with a convolutional regression tracker

Ye Lyu et al.

Summary: Video object detection is less researched compared to object detection in images due to shortage of labelled video datasets. Frames in a video clip are highly correlated, requiring more video labels for good data variation. Propose to improve the performance of an image object detector by augmenting it with a class-agnostic convolutional regression tracker.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Article Environmental Sciences

Two shifts for crop mapping: Leveraging aggregate crop statistics to improve satellite-based maps in new regions

Dan M. Kluger et al.

Summary: Mapping crop types at the field level using satellite imagery is crucial for agricultural monitoring, but challenges arise from shifts in crop distribution and feature transformations between regions. This study presents a methodology that corrects classifiers by reweighting posterior probabilities and removing linear shifts in mean feature vectors, leading to substantial improvements in classification accuracy for different training regions in France and Kenya. This method, originally designed for Linear Discriminant Analysis, can be applied to other classifiers such as Random Forest with successful outcomes.

REMOTE SENSING OF ENVIRONMENT (2021)

Article Environmental Sciences

Evaluation of Five Deep Learning Models for Crop Type Mapping Using Sentinel-2 Time Series Images with Missing Information

Hongwei Zhao et al.

Summary: Accurate crop type maps are essential for food security. Optical time series data is important for crop type mapping, but the need to fill in missing information due to clouds in imagery can increase workload and error risk. The development of high resolution optical imagery and deep learning algorithms provides solutions to this issue. Different models such as 1D CNN, LSTM, and GRU have been used in previous studies for crop type classification, but their ability to handle missing information in TSD needs further exploration. Some models performed well with unfilled TSD, achieving higher and more stable classification accuracies compared to filled TSD. This research showed that deep learning models can accurately identify crop types in Sentinel-2 images with missing information.

REMOTE SENSING (2021)

Article Green & Sustainable Science & Technology

Massive soybean expansion in South America since 2000 and implications for conservation

Xiao-Peng Song et al.

Summary: Policies aiming to reduce climate change and biodiversity loss often target achieving zero deforestation in global supply chains of key commodities like palm oil and soybean. However, a study on soybean expansion in South America between 2000 and 2019 found that most of the expansion occurred on pastures originally converted for cattle production, particularly in the Brazilian Amazon. The fastest soybean expansion was observed in Brazil, with the majority of soy-driven deforestation concentrated in the Brazilian Cerrado region.

NATURE SUSTAINABILITY (2021)

Article Engineering, Electrical & Electronic

TimeSen2Crop: A Million Labeled Samples Dataset of Sentinel 2 Image Time Series for Crop-Type Classification

Giulio Weikmann et al.

Summary: The article introduces the TimeSen2Crop dataset, consisting of over 1 million samples of Sentinel 2 time series for research in crop type mapping. The dataset includes atmospherically corrected images and information on snow, shadows, and clouds.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2021)

Article Remote Sensing

Learning discriminative spatiotemporal features for precise crop classification from multi-temporal satellite images

Shunping Ji et al.

INTERNATIONAL JOURNAL OF REMOTE SENSING (2020)

Article Geography, Physical

Self-attention for raw optical Satellite Time Series Classification

Marc Russwurm et al.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2020)

Article Engineering, Electrical & Electronic

A CNN-Transformer Hybrid Approach for Crop Classification Using Multitemporal Multisensor Images

Zhengtao Li et al.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2020)

Article Computer Science, Artificial Intelligence

Visual object tracking based on adaptive Siamese and motion estimation network

Hossein Kashiani et al.

IMAGE AND VISION COMPUTING (2019)

Article Environmental Sciences

Crop type mapping without field-level labels: Random forest transfer and unsupervised clustering techniques

Sherrie Wang et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Article Environmental Sciences

Deep learning based multi-temporal crop classification

Liheng Zhong et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Article Environmental Sciences

Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series

Charlotte Pelletier et al.

REMOTE SENSING (2019)

Article Computer Science, Artificial Intelligence

Weakly Supervised Adversarial Domain Adaptation for Semantic Segmentation in Urban Scenes

Qi Wang et al.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2019)

Article Environmental Sciences

Key issues in rigorous accuracy assessment of land cover products

Stephen V. Stehman et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Article Environmental Sciences

Needle in a haystack: Mapping rare and infrequent crops using satellite imagery and data balancing methods

Francois Waldner et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Proceedings Paper Computer Science, Artificial Intelligence

BASNet: Boundary-Aware Salient Object Detection

Xuebin Qin et al.

2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019) (2019)

Article Agricultural Economics & Policy

Agricultural and Applied Economics Priorities for Solving Societal Challenges

Jill J. McCluskey et al.

APPLIED ECONOMIC PERSPECTIVES AND POLICY (2018)

Article Computer Science, Information Systems

Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders

Marc Russwurm et al.

ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION (2018)

Article Geochemistry & Geophysics

Deep Learning Classification of Land Cover and Crop Types Using Remote Sensing Data

Nataliia Kussul et al.

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS (2017)

Article Environmental Sciences

MODIS phenology-derived, multi-year distribution of conterminous US crop types

Richard Massey et al.

REMOTE SENSING OF ENVIRONMENT (2017)

Article Environmental Sciences

A multi-resolution approach to national-scale cultivated area estimation of soybean

LeeAnn King et al.

REMOTE SENSING OF ENVIRONMENT (2017)

Article Environmental Sciences

Toward mapping crop progress at field scales through fusion of Landsat and MODIS imagery

Feng Gao et al.

REMOTE SENSING OF ENVIRONMENT (2017)

Proceedings Paper Computer Science, Artificial Intelligence

Temporal Vegetation Modelling using Long Short-Term Memory Networks for Crop Identification from Medium-Resolution Multi-Spectral Satellite Images

Marc Russwurm et al.

2017 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW) (2017)

Proceedings Paper Computer Science, Artificial Intelligence

Deeply Supervised Salient Object Detection with Short Connections

Qibin Hou et al.

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) (2017)

Review Geography, Physical

Random forest in remote sensing: A review of applications and future directions

Mariana Belgiu et al.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2016)

Article Geography, Physical

An Assessment of Algorithmic Parameters Affecting Image Classification Accuracy by Random Forests

Di Shi et al.

PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING (2016)

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

Article Environmental Sciences

Land surface phenology along urban to rural gradients in the US Great Plains

J. J. Walker et al.

REMOTE SENSING OF ENVIRONMENT (2015)

Article Environmental Sciences

Efficient corn and soybean mapping with temporal extendability: A multi-year experiment using Landsat imagery

Liheng Zhong et al.

REMOTE SENSING OF ENVIRONMENT (2014)

Article Environmental Sciences

Near real-time prediction of US corn yields based on time-series MODIS data

Toshihiro Sakamoto et al.

REMOTE SENSING OF ENVIRONMENT (2014)

Article Geography, Physical

Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using Support Vector Machines

F. Loew et al.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2013)

Article Environmental Sciences

The response of African land surface phenology to large scale climate oscillations

Molly E. Brown et al.

REMOTE SENSING OF ENVIRONMENT (2010)

Article Meteorology & Atmospheric Sciences

Satellite-based near-real-time estimation of irrigated crop water consumption

Qiuhong Tang et al.

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES (2009)

Article Environmental Sciences

Phenologically-tuned MODIS NDVI-based production anomaly estimates for Zimbabwe

Chris Funk et al.

REMOTE SENSING OF ENVIRONMENT (2009)

Article Environmental Sciences

A crop phenology detection method using time-series MODIS data

T Sakamoto et al.

REMOTE SENSING OF ENVIRONMENT (2005)

Article Environmental Sciences

Mapping paddy rice agriculture in southern China using multi-temporal MODIS images

XM Xiao et al.

REMOTE SENSING OF ENVIRONMENT (2005)