Identify Crop Types Research Articles

Hierarchical classification for improving parcel-scale crop mapping using time-series Sentinel-1 data

Parcel-scale crop classification utilizing time-series satellite observations is of significant importance in precision agriculture. The prior knowledge that crop types can be organized in a hierarchical tree structure is beneficial for improving crop classification. Moreover, the crop hierarchy aligns with the coarse-to-fine cognitive process of geographic scenes. Based on the crop hierarchy, this study developed a general hierarchical classification framework for enhancing crop mapping using time-series Sentinel-1 data. Central to this method is a deep-learning-based hierarchical classification model that explores and makes use of crop hierarchical knowledge. First, preprocessed Sentinel-1 data were geometrically overlaid onto farmland parcel maps to derive parcel-scale time-series features. Second, we constructed a hierarchical crop type system for study areas based on the crop phenology of labeled crop-type samples. Third, we developed a deep-learning-based hierarchical classification model to identify crop types for each parcel, to generate final crop-type classification maps. The proposed approach was further discussed and verified through the implementation of parcel-scale time-series crop hierarchical classifications in a study area in France with farmland parcel maps and time-series Sentinel-1 data. The classification results, indicating significant improvements greater than 4.0% in overall accuracy and 5.0% in F1 score over comparative methods, demonstrated the effectiveness of the proposed method in learning multi-scale time-series features for hierarchical crop classification utilizing Sentinel-1 data sequences.

Mapping tobacco planting areas in smallholder farmlands using Phenological-Spatial-Temporal LSTM from time-series Sentinel-1 SAR images

Phenological information on crop growth aids in identifying crop types from remote sensing images, but its incorporation into classification models is insufficiently exploited, especially in deep learning frameworks. This study presents a new model, Phenological-Temporal-Spatial LSTM (PST-LSTM), for mapping tobacco planting areas in smallholder farmlands using time-series Sentinel-1 Synthetic Aperture Radar (SAR) images. The PST-LSTM model is built on a multi-modal learning framework that fuses phenological information with deep spatial–temporal features. We applied the model to extract tobacco planting areas in Ninghua, Pucheng, and Shanghang Counties, in Fujian Province, and Luoping County in Yunnan Province, China. We compared PST-LSTM with existing methods based on phenological rules and Dynamic Time Warping (DTW) methods, and analyzed its strength in feature fusion. Results showed that our model outperformed these methods, achieving an overall accuracy (OA) of 93.16% compared to 86.69% and 85.93% for the phenological rules and DTW methods, respectively, in the Ninghua area. PST-LSTM effectively integrated time-series data with phenological information derived from different strategies at the feature level and performed better than existing feature fusion methods (based upon fuzzy sets) at the decision level. It also demonstrated a better spatial transferability than other methods when applied to different study areas, achieving an OA of 90.95%, 91.41%, and 80.75% for the Pucheng, Shanghang, and Luoping areas, respectively, using training samples from Ninghua. We conclude that PST-LSTM can effectively extract tobacco planting areas in smallholder farming from time-series SAR images and has the potential for mapping other crop types.

Contrastive-Learning-Based Time-Series Feature Representation for Parcel-Based Crop Mapping Using Incomplete Sentinel-2 Image Sequences

Parcel-based crop classification using multi-temporal satellite optical images plays a vital role in precision agriculture. However, optical image sequences may be incomplete due to the occlusion of clouds and shadows. Thus, exploring inherent time-series features to identify crop types from incomplete optical image sequences is a significant challenge. This study developed a contrastive-learning-based framework for time-series feature representation to improve crop classification using incomplete Sentinel-2 image sequences. Central to this method was the combined use of inherent time-series feature representation and machine-learning-based classifications. First, preprocessed multi-temporal Sentinel-2 satellite images were overlaid onto precise farmland parcel maps to generate raw time-series spectral features (with missing values) for each parcel. Second, an enhanced contrastive learning model was established to map the raw time-series spectral features to their inherent feature representation (without missing values). Thirdly, eXtreme Gradient-Boosting-based and Long Short-Term Memory-based classifiers were applied to feature representation to produce crop classification maps. The proposed method is further discussed and validated through parcel-based time-series crop classifications in two study areas (one in Dijon of France and the other in Zhaosu of China) with multi-temporal Sentinel-2 images in comparison to the existing methods. The classification results, demonstrating significant improvements greater than 3% in overall accuracy and 0.04 in F1 scores over comparison methods, indicate the effectiveness of the proposed contrastive-learning-based time-series feature representation for parcel-based crop classification utilizing incomplete Sentinel-2 image sequences.

English

Crop mapping and its health monitoring are of key importance for a country like Pakistan where a considerable share of the GDP originates from agricultural production. Currently, extensive capital and human resource are utilized to record the crop types and monitor crop health due to variations in crop pattern and intercropping practices being observed in many parts of the country. Satellite Remote Sensing, however, can be utilized very efficiently to identify crop types and estimate acreage by using unique phenological cycles derived from NDVI thresholds. In this research, we proposed a semi-automated, efficient and cost-effective state-of-the-art technique for crop classification using satellite remote sensing datasets and GIS techniques to replace outdated conventional survey-based crop estimations procedures. This research was conducted in Sahiwal district, Pakistan to categorise several crop types using two major inputs: satellite (Sentinel-2, 10m) imagery and local crop calendar. The research adopted a semi-automatic approach to find the main crops of Rabi (winter) and Kharif (summer) using orthorectified and atmospherically corrected Level-1C images to extract phenological information based on Normalized Difference Vegetation Index (NDVI). The complete phenological life cycle of crops including sowing, growing, and harvesting stages were deeply analyzed with the help of crop calendar and NDVI profiles. Then this information passed through iterative selforganizing (ISO Data Model) of unsupervised classification technique in the form of temporal stacked images. The NDVI profiles of each crop were synchronized with the crop calendar of the study area to develop the crop categorization procedure. The results were validated with the statistics of the agriculture department Lahore and field samples gave about 85% and 88% accuracy for Rabi and Kharif cropping seasons respectively. It is concluded from the research that the tested methodology is capable of providing a fast, cost-efficient, and reliable solution for crop mapping instead of conventional survey-based crop area assessment techniques (conventional Girdawri) which requires a vast number of resources and time, and it can be replicated in different areas with minor changes

Parcel level staple crop type identification based on newly defined red-edge vegetation indices and ORNN

It is of strategic importance for early estimating planting area and crop yield to identify staple crop types on parcel level in a remotely sensed way. Limited by remote sensor imaging mechanism and current sensor technology, remote sensing images usually cannot be of high resolution in both spatial and spectral aspects. So, it has been found in many studies that supplementary data or features such as synthetic aperture radar data or temporal characteristic of time series data are usually integrated for crop type identification, which demands lots of data and overlooks the most typical spectral characteristics. Thanks to the expanded detecting bands of several moderate resolution optical satellites, the red-edge bands reflecting specific spectral features of crops are available for precise classification. Nevertheless, only a few studies have concentrated on these bands and the potential value of red-edge bands has not been sufficiently stretched in practical use. In this study, two newly defined red-edge vegetation indices (REVIs) and their time series data have been used for distinguishing different crops efficiently. Firstly, according to the reflectance characteristics in optical images and the data distribution of crops in feature space, two kinds of variants of REVIs based on green band and red-edge band, namely, ReG_RVI and ReG_NDVI have been proposed. Then, the time series data of REVIs can be generated from remote sensing images and be fed into the recurrent neural network (RNN) as training data for crop type classification. Finally, object based image analysis and RNN are combined with the idea of majority voting to achieve accurate crop type identification on cultivated land parcel level. The experimental results of two study areas using time series images captured by Gaofen 6 wide field of view sensor show an overall accuracy of above 90 % in identifying crop types, which proves the effectiveness of the newly defined REVIs and the feasibility of the proposed methodology. The predicted model based on ReG_RVI time series data performed more stable. In addition, the proposed REVIs have been transferred to other satellite sensor images captured by Sentinel 2 to testify 16 crop types and the experimental results indicate that the proposed indices also had general applicability to a certain extent. In summary, the contribution of this article is that the newly defined REVIs exploit red-edge information more sufficiently and the proposed methodology provides an effective solution for accurate identification and mapping of crop types in large areas at the cultivated land parcel scale.

Learning spectral-spatial representations from VHR images for fine-scale crop type mapping: A case study of rice-crayfish field extraction in South China

Accurate information on crop type spatial distributions is crucial for yield estimation, agroecological modeling, and agrarian policy development. In comparison with moderate resolution (10–30 m) satellite images, very high-resolution (VHR, ≤ 1 m) images are superior at describing the spatial details and complex texture features of agricultural land. However, VHR images generally suffer from insufficient spatiotemporal coverage and high spectral variation in crop type, making fine-scale crop type mapping at regional or larger scales challenging. In this study, we developed a novel data-driven residual attention U-shape network (RAUNet) to identify crop types at the regional scale based on VHR images. RAUNet was adapted from UNet by introducing a residual module and a gated attention mechanism to strengthen the multilevel representation of crop features from VHR images. We selected rice-crayfish fields (RCFs) in Hubei Province, China, as the case crop type and area to test the performance of RAUNet. Seven models, including RAUNetpan, ResUNet, AttnUNet, UNet, DeepLabV3 + and random forest at the pixel and object levels, were adopted for comparison with RAUNet. Moreover, the model transferability was evaluated over regions with different agricultural landscapes. Our results showed that RAUNet-derived RCF maps obtained an average F1-score and MCC of 0.90 and 0.85 at different cultivation stages, which significantly outperformed other methods. Additionally, RAUNet had good spatiotemporal transferability, with an average F1-score and MCC of 0.93 and 0.88 in four transfer areas. The visualization results of the attention weights indicated RAUNet’s prominent capability in combining useful and relevant information from different feature levels to produce refinement predictions for the target crop type. Furthermore, the high-level features derived from RAUNet can well separate RCFs and non-RCFs over different landscapes, which were also consistent across different cultivation periods. These encouraging results suggest that RAUNet works well for fine-scale crop type mapping in smallholder farming systems, and it is an efficient method for land cover mapping based on VHR images at the regional scale.

In-season crop type identification using optimal feature knowledge graph

Early or in-season crop type mapping using remote sensing data is important for crop managements to maximize crop yield. Existing remote sensing-based approaches in the literature rely mainly on the timing and temporal frequency of satellite data acquisition during a growing season to obtain optimal identification feature (OIF) to complete crop mapping. At present, the optimal features for mapping crop types are commonly generated by using different feature selection methods, and thereby are not generic in nature. In this paper, a novel approach based on the OIF knowledge graph is proposed for identification of crop types within the crop growing season. The concept of OIF was proposed based on the one-class classification and optimization theory, which relates a specific crop type to its OIF. The OIF knowledge graph can be created to describe the seasonal trend of OIF in a season. Eventually, the OIF knowledge graphs can be incorporated into the moment-preserving segmentation method to automatically identify crop types within the growing season. To evaluate the performance of the new method, crop mapping was conducted using 2019 Sentinel-2 data in Heilongjiang province, China, and Illinois, USA, respectively. Results demonstrate that integrating the OIF knowledge graph and moment-preserving thresholding method is effective in identifying crop types without ground-truth data. The overall accuracy higher than 90 % (Kappa coefficient greater than 0.90), and the producer’s accuracies for corn, soybean, and rice are generally greater than 93 % in Heilongjiang province. The proposed approach has the advantage of simplifying crop identification process while maintaining a satisfactory identification accuracy in shorter time than conventional methods.

ELSET: Design of an Ensemble Deep Learning Model for improving satellite image Classification Efficiency via Temporal Analysis

Satellite images with multispectral and panchromatic layers may contain a variety of information about the ROI (region of interests). Each layer displays distinct regional characteristics, allowing researchers to identify unique locations. Numerous machine learning algorithms are capable of identifying crop types, land cover types, soil types, etc. Researchers frequently use deep learning classifiers such as FCNN (Fully Connected Neural Network), RNN (Recurrent Neural Network), Q-learning, etc. as a result of their varying performance, applicability, complexity, and implementation costs. These networks are too general for classifications of application-specific satellite images. This paper proposes an improved deep learning model for real-time region-based satellite image classification in application-specific environments. ELSET captures large-scale temporal datasets using Google Earth Engine and removes outliers from these datasets using application-specific segmentation model filters. Image sets are segmented by an upgraded CNN model including VGGNet 19, GoogLeNet, and ResNet V2. VGGNet 19 outperforms other classification techniques for crop type and crop damage detection, GoogLeNet for land cover and forest cover classification, and ResNet V2 for classification of water bodies and urban areas. Combining these algorithms enables the identification of image layered regions with high precision and moderate complexity. Due to these techniques, the proposed model is capable of crop type recognition accuracy of 98.9%, land type identification accuracy of 95.4%, and water & urban cover area classification accuracy of 96.5%. The proposed model was evaluated at various locations and times with consistent precision. The precision, recall, AUC, and latency of the model were compared to state-of-the-art techniques. The proposed model demonstrated 5%, 8%, and 5.9% greater accuracy, recall, and AUC than current models. Due to increased categorization, the suggested model has significantly more latency than conventional alternatives. This paper provides researchers with strategies for context-specific satellite image classification that reduce this delay without compromising model performance levels.

Mapping of cropland, cropping patterns and crop types by combining optical remote sensing images with decision tree classifier and random forest

ABSTRACT Mapping and monitoring the distribution of croplands and crop types support policymakers and international organizations by reducing the risks to food security, notably from climate change and, for that purpose, remote sensing is routinely used. However, identifying specific crop types, cropland, and cropping patterns using space-based observations is challenging because different crop types and cropping patterns have similarity spectral signatures. This study applied a methodology to identify cropland and specific crop types, including tobacco, wheat, barley, and gram, as well as the following cropping patterns: wheat-tobacco, wheat-gram, wheat-barley, and wheat-maize, which are common in Gujranwala District, Pakistan, the study region. The methodology consists of combining optical remote sensing images from Sentinel-2 and Landsat-8 with Machine Learning (ML) methods, namely a Decision Tree Classifier (DTC) and a Random Forest (RF) algorithm. The best time-periods for differentiating cropland from other land cover types were identified, and then Sentinel-2 and Landsat 8 NDVI-based time-series were linked to phenological parameters to determine the different crop types and cropping patterns over the study region using their temporal indices and ML algorithms. The methodology was subsequently evaluated using Landsat images, crop statistical data for 2020 and 2021, and field data on cropping patterns. The results highlight the high level of accuracy of the methodological approach presented using Sentinel-2 and Landsat-8 images, together with ML techniques, for mapping not only the distribution of cropland, but also crop types and cropping patterns when validated at the county level. These results reveal that this methodology has benefits for monitoring and evaluating food security in Pakistan, adding to the evidence base of other studies on the use of remote sensing to identify crop types and cropping patterns in other countries.

Assessing evapotranspiration observed from ECOSTRESS using flux measurements in agroecosystems

Crop water use (also know as evapotranspiraton, ET) monitoring is vital for agricultural management under the threat of water scarcity and global warming, as well as the pressure of increasing food demand. The ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) has provided valuable high-resolution ET estimates since its launch in 2018. Yet, evaluation of its capability in monitoring ET in the agricultural field is lacking, as well as the intercomparison of the two ECOSTRESS ET products calculated through different algorithms (PT-JPL vs. DisALEXI). This study focused on the agroecosystems and compared the ECOSTRESS ET to eddy covariance observations at 15 cropland flux sites in North America from July 2018 to December 2020. It is found that PT-JPL captured the diurnal pattern but tended to overestimate instantaneous ET in the morning. Besides, intercomparison of the magnitude, seasonal pattern and the dependency on crop phenology of ET from PT-JPL and DisALEXI revealed the higher fidelity of DisALEXI. The site-specific validations and visual comparisons with MODIS ET retrievals demonstrated the capability of PT-JPL and DisALEXI in identifying crop types and agricultural management practices to some extent. The heterogeneity of ET values in the small-scale agroecosystems highlights the importance of accurate representation of local conditions in the ET model.

Exploring the potential of Chinese GF-6 images for crop mapping in regions with complex agricultural landscapes

Accurate and timely crop mapping is crucial for environment assessment, food security and agricultural production. However, for the areas with high landscape heterogeneity and frequent cloudy and rainy weather, the insufficient high-quality satellite images limit the accuracy of crop classification. The recently launched Chinese GF-6 wide field-of-view camera (WFV) with a revisit cycle of 4-day and spatial resolution of 16-meter shows great potential for agricultural monitoring. In this study, Qianjiang City characterized by complex agricultural landscapes was selected as the research area to assess the potential of GF-6 data in identifying crop types. Firstly, the pairwise and global separability were calculated to analyze the effect of different spectral-temporal features of GF-6 images on crop classification. A total of 255 spectral-temporal features derived from 15 GF-6 tiles were then used to perform random forest classification. Furthermore, the classification results were evaluated based on 671 field samples and then compared the accuracy between GF-6 data and Sentinel-2 or Landsat-8 data. In addition, the earliest identifiable time of crop types was also determined by iteratively using all available GF-6 data during each time period. The results suggested that the overall accuracy (OA) of all available GF-6 images was 91.55%, which was significantly higher than that of Landsat-8 data (OA = 85.97%) and was slightly lower than that of Sentinel-2 data (OA = 93.10%). The newly added red-edge bands (0.69 ∼ 0.73 μm, 0.73 ∼ 0.77 μm) and their derivative vegetation indices were important spectral features, and the period from mid-March to early-April was the best temporal window for crop identification in our research area. Moreover, late July was the earliest crop identifiable time with overall accuracy of 90% for the first time of the year. These results indicated the great potential of GF-6 images for classifying crop types in the areas with complex cropping system and fragmented agricultural landscapes, particularly when integrating other satellite data with comparable spatial resolution (e.g. Chinese GF-1 data and Sentinel-2 data).

Adding simulated NDVI images can be effective for identifying crop types from Sentinel-1 C-SAR data

Adding simulated NDVI images can be effective for identifying crop types from Sentinel-1 C-SAR data

Crop Mapping with Combined Use of European and Chinese Satellite Data

Agricultural landscapes are characterized by diversity and complexity, which makes crop mapping at a regional scale a top priority for different purposes such as administrative decisions and farming management. Project 32194 of the Dragon 4 Program was implemented to meet the requirements of crop mapping, with the specific objective to develop suitable approaches for precise crop mapping with combined uses of European and Chinese high- and medium-resolution satellite images. Two sub-projects were involved in the project. The first was to focus on the use of time series high-resolution satellite data, including Sentinel-2 (S2, European satellite data) and Gaofen-1 (GF-1, Chinese satellite data), due to their similar spectral bands for Earth observation, while the second was to focus on medium-resolution data sources, i.e., the European Project for On-Board Autonomy–Vegetation (PROBA-V) and Chinese Fengyun-3 Medium Resolution Spectral Imager (FY-3 MERSI) satellite data, also due to their similar spectral channels. The approach of the European Space Agency (ESA) Sent2Agri project for crop mapping was adapted in the first sub-project and applied to the Yellow River irrigated district (YERID) of Ningxia in northwest China in order to assess its ability to accurately identify crop types in China. The goal of the second sub-project was to explore the potential of both European and Chinese medium-resolution satellite data for crop assessment in a large area. Methods to handle the data and retrieve the required information for the precise crop mapping were developed in the study, including the adaptation of the ESA approach to GF-1 data and the application of algorithms for classification. A scheme for the validation of the crop mapping was developed in the study. The results of implementing the scheme to the YERID in Ningxia indicated that the overall accuracies of crop mapping with S2 and GF-1 can be high, up to 94–97%, and the mapping had an accuracy of 88% with the PROBA-V and FY3B-MERSI data. The very high accuracy suggests the possibility of precise crop mapping with the combined use of time series high- and medium-resolution satellite data when suitable approaches are chosen to handle the data for the classification of crop types.

Towards interpreting multi-temporal deep learning models in crop mapping

Multi-temporal deep learning approaches have exhibited excellent classification performance in large-scale crop mapping. These approaches efficiently and automatically transform remote sensing time series into high-dimensional feature representations to identify crop types. The lack of interpretation, however, is regarded as a major drawback of these high-performance approaches. Interpreting deep learning approaches in multi-temporal crop mapping is critical for verifying their reliability. This study aims to quantify the impact of multi-temporal information in input time series on classification performance and develop a multi-perspective interpretation pipeline for deep learning models. The pipeline involves three interpretation approaches: evaluating input feature importance, analyzing hidden features, and monitoring temporal changes in model's soft output. An experiment is conducted to classify corn and soybean in the U.S corn belt in 2018. The study area consists of three sites each encompassing millions of pixel-level samples at 30 m resolution. The Landsat Analysis Ready Data are used as the input remote sensing time series and Cropland Data Layer is used as the ground reference. Attention-based Long Short-Term Memory (AtLSTM) and Transformer models are built as multi-temporal deep learning models, and compared to Random Forest (RF). Complete time series input in the correct order achieves a higher overall accuracy of 97.8% than using single-window or out-of-order inputs, indicating multi-temporal information facilitates crop classification. An assessment of the input feature importance demonstrates that the AtLSTM, Transformer, and RF models all consider the period from weeks 11 to 20 (early-July to late-August) as a key growth period and the shortwave infrared band as the critical band for corn and soybean discrimination. Hidden feature analysis suggests that the AtLSTM model accumulates the useful information over the growth period, while the Transformer model extracts the temporal dependencies that contribute important information to high-level feature learning. The learned features contain more effective and refined information than the raw input features and thus are better suited for crop classification. The soft output analysis in the in-season classification scenario demonstrates that increased length of input time series improves the model's confidence in the classification results. The further comparison of input feature importance in different sites and years demonstrates the applicability of the interpretation approach at larger spatiotemporal extents with heterogeneous landscapes and interannual variability. This study provides a multi-perspective evaluation to identify key features in multi-spectral and multi-temporal remote sensing data, and yields a practical approach to integrate agronomy knowledge in deep learning-based crop mapping.

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

Accurate crop type maps play an important role in food security due to their widespread applicability. Optical time series data (TSD) have proven to be significant for crop type mapping. However, filling in missing information due to clouds in optical imagery is always needed, which will increase the workload and the risk of error transmission, especially for imagery with high spatial resolution. The development of optical imagery with high temporal and spatial resolution and the emergence of deep learning algorithms provide solutions to this problem. Although the one-dimensional convolutional neural network (1D CNN), long short-term memory (LSTM), and gate recurrent unit (GRU) models have been used to classify crop types in previous studies, their ability to identify crop types using optical TSD with missing information needs to be further explored due to their different mechanisms for handling invalid values in TSD. In this research, we designed two groups of experiments to explore the performances and characteristics of the 1D CNN, LSTM, GRU, LSTM-CNN, and GRU-CNN models for crop type mapping using unfilled Sentinel-2 (Sentinel-2) TSD and to discover the differences between unfilled and filled Sentinel-2 TSD based on the same algorithm. A case study was conducted in Hengshui City, China, of which 70.3% is farmland. The results showed that the 1D CNN, LSTM-CNN, and GRU-CNN models achieved acceptable classification accuracies (above 85%) using unfilled TSD, even though the total missing rate of the sample values was 43.5%; these accuracies were higher and more stable than those obtained using filled TSD. Furthermore, the models recalled more samples on crop types with small parcels when using unfilled TSD. Although LSTM and GRU models did not attain accuracies as high as the other three models using unfilled TSD, their results were almost close to those with filled TSD. This research showed that crop types could be identified by deep learning features in Sentinel-2 dense time series images with missing information due to clouds or cloud shadows randomly, which avoided spending a lot of time on missing information reconstruction.

UAV image crop classification based on deep learning with spatial and spectral features

Unmanned Aerial Vehicle image analysis technology has become an effective means to classify crops. However, the UAV image classification mostly adopts shallow learning algorithm, with few computing units. These methods have low recognition accuracy and poor generalization ability. Deep learning has powerful function expression and excellent feature learning abilities. The learned features have more essential character for data and have achieved remarkable results in image classification. Therefore, the study proposes a crop classification method based on Unmanned Aerial Vehicle image with deep learning and spatial and spectral feature fusion. The method uses deep convolutional neural network to classify Unmanned Aerial Vehicle images. Simplified AlexNet network structure, which reduces the number of network layers, was used to accelerate the convergence speed of the model while ensuring the accuracy of crop classification in practical applications. Then, the vegetation index and height features of the Unmanned Aerial Vehicle image were extracted. Feature combination and comparative analyses were carried out to find the most effective feature combination and improve the accuracy of crop classification through spatial and spectral feature fusion. In addition, a Sample Expansion Strategy was used to optimize the classification model and further improve the classification result to achieve a perfect performance in the crop classification of drone images. The experimental results showed that the deep learning method can effectively identify crop types in Unmanned Aerial Vehicle images, and the overall classification accuracy is raised from 86.07% to 92.76% when combining spatial and spectral feature fusion with Sample Expansion Strategy.

Classification and Recognition of Crops Based on GIS Image

Today, due to the rapid development of science and technology, new science and technology constantly change the status of agriculture, precision agriculture is developing rapidly. Precision agriculture is the combination of three new technologies: Global Positioning System (GPS), Remote Sensing Technology (RS) and Geographic Information System (GIS). In the 3S technology, GIS technology is the basis of realizing the other two technologies. It is a platform to achieve precision agriculture, supporting the input, analysis and output of agricultural resources data and other functions of precision agriculture. Classification and recognition of crop types is one of the most basic applications of precision agriculture. It can extract information from remote sensing images, accurately identify crop types and determine crop growth. In this paper, 7200 remote sensing image samples are divided into two groups: learning group and testing group. Combined with GIS image and RS technology, the characteristics of these four crops were summarized and counted through feature image extraction and feature data analysis of 5200 study group samples in GIS system. 2000 samples in the test group were used to calculate the recognition rate. Through experiments, we can know the recognition rate of peanut, spring maize, sweet potato and summer maize. The highest recognition rate is 98.60%, the lowest recognition rate is 90.20%, and the total recognition rate is 94.10%. It shows that in this experiment, the crop recognition rate of remote sensing image is very high in the application of classification and recognition based on GIS crop images. During the period when the spectral characteristics are very obvious, the recognition rate of crops in remote sensing images is higher, and the total recognition rate is increased by 3.98%. The NDVI values of crops in remote sensing images were analyzed and calculated, and the NDVI values of four crops in each growing period were obtained. By studying the NDVI value, we can determine the types of four crops and their growth period, and the accuracy of determining the types of crops is higher than that of determining the growth period of crops. Through the classification and identification of crop species and crop growth period, two experiments verify that the combination of GIS image and remote sensing image can improve the recognition accuracy of crop classification and identification, and the use of geospatial data in GIS can identify the differences between crops. By combining GIS images with remote sensing images, the development of precision agriculture can be promoted, and the reference for agricultural production and agricultural structure adjustment in designated areas can be provided.

Fusion of Sentinel-2 and PlanetScope time-series data into daily 3 m surface reflectance and wheat LAI monitoring

The dynamics of Leaf Area Index (LAI) from space is key to identify crop types and their phenology over large areas, and to characterize spatial variations within growers’ fields. However, for years remote-sensing applications have been constrained by a trade-off between the spatial and temporal resolutions. This study resolves this limitation. Over the past decade, the number of companies and organizations developing CubeSat constellations has increased. These new satellites make it possible to acquire large image collections at high spatial and temporal resolutions at a relatively low cost. However, the images obtained from CubeSat constellations frequently suffer from inconsistency in the data calibration between the different satellites within the constellation. To overcome these inconsistencies, a new method to fuse a time series of images sourced from two different satellite constellations is proposed, combining the advantages of both (i.e., the temporal, spatial and spectral resolution). This new technique was applied to fuse PlanetScope images with Sentinel-2 images, to create spectrally-consistent daily images of wheat LAI at a 3 m resolution. The daily 3 m LAI estimations were compared with 57 in-situ wheat LAI measurements taken in Australia and Israel. This approach was demonstrated to successfully estimate Green LAI (LAI before the major on-set of leaf senescence) with an R2 of 0.94 and 86% relative accuracy (RMSE of 1.37) throughout the growing season without using any ground calibration. However, both the Sentinel-2 based estimates and the fused Green LAI were underestimated at high LAI values (LAI > 3). To account for this, regression models were developed, improving the relative accuracy of the Green LAI estimations by up to a further 47% (RMSE of 0.35–0.63) in comparison with field measured LAI. The new time series fusion method is an effective tool for continuous daily monitoring of crops at high-resolution over large scales, which opens up a range of new precision agriculture applications. These high spatio-temporal resolution time-series are valuable for monitoring crop growth and health, and can improve the effectiveness of farming practices and enhance yield forecasts at the field and sub-field scales.

Crop Classification Method Based on Optimal Feature Selection and Hybrid CNN-RF Networks for Multi-Temporal Remote Sensing Imagery

Although efforts and progress have been made in crop classification using optical remote sensing images, it is still necessary to make full use of the high spatial, temporal, and spectral resolutions of remote sensing images. However, with the increasing volume of remote sensing data, a key emerging issue in the field of crop classification is how to find useful information from massive data to balance classification accuracy and processing time. To address this challenge, we developed a novel crop classification method, combining optimal feature selection (OFSM) with hybrid convolutional neural network-random forest (CNN-RF) networks for multi-temporal optical remote sensing images. This research used 234 features including spectral, segmentation, color, and texture features from three scenes of Sentinel-2 images to identify crop types in the Jilin province of northeast China. To effectively extract the effective features of remote sensing data with lower time requirements, the use of OFSM was proposed with the results compared with two traditional feature selection methods (TFSM): random forest feature importance selection (RF-FI) and random forest recursive feature elimination (RF-RFE). Although the time required for OFSM was 26.05 s, which was between RF-FI with 1.97 s and RF-RFE with 132.54 s, OFSM outperformed RF-FI and RF-RFE in terms of the overall accuracy (OA) of crop classification by 4% and 0.3%, respectively. On the basis of obtaining effective feature information, to further improve the accuracy of crop classification we designed two hybrid CNN-RF networks to leverage the advantages of one-dimensional convolution (Conv1D) and Visual Geometry Group (VGG) with random forest (RF), respectively. Based on the selected optimal features using OFSM, four networks were tested for comparison: Conv1D-RF, VGG-RF, Conv1D, and VGG. Conv1D-RF achieved the highest OA at 94.27% as compared with VGG-RF (93.23%), Conv1D (92.59%), and VGG (91.89%), indicating that the Conv1D-RF method with optimal feature input provides an effective and efficient method of time series representation for multi-temporal crop-type classification.

Early-season crop type mapping using 30-m reference time series

Early-season crop type mapping could provide important information for crop growth monitoring and yield prediction, but the lack of ground-surveyed training samples is the main challenge for crop type identification. Although reference time series based method (RBM) has been proposed to identify crop types without the use of ground-surveyed training samples, the methods are not suitable for study regions with small field size because the reference time series are mainly generated using data set with low spatial resolution. As the combination of Landsat data and Sentinel-2 data could increase the temporal resolution of 30-m image time series, we improved the RBM by generating reference normalized difference vegetation index (NDVI)/enhanced vegetation index (EVI) time series at 30-m resolution (30-m RBM) using both Landsat and Sentinel-2 data, then tried to estimate the potential of the reference NDVI/EVI time series for crop identification at early season. As a test case, we tried to use the 30-m RBM to identify major crop types in Hengshui, China at early season of 2018, the results showed that when the time series of the entire growing season were used for classification, overall classification accuracies of the 30-m RBM were higher than 95%, which were similar to the accuracies acquired using the ground-surveyed training samples. In addition, cotton, spring maize and summer maize distribution could be accurately generated 8, 6 and 8 weeks before their harvest using the 30-m RBM; but winter wheat can only be accurately identified around the harvest time phase. Finally, NDVI outperformed EVI for crop type classification as NDVI had better separability for distinguishing crops at the green-up time phases. Comparing with the previous RBM, advantage of 30-m RBM is that the method could use the samples of the small fields to generate reference time series and process image time series with missing value for early-season crop classification; while, samples collected from multiple years should be further used so that the reference time series could contain more crop growth conditions.