Using Time Series Sentinel-1 Images for Object-Oriented Crop Classification in Google Earth Engine

Mangrove Species Classification from Unmanned Aerial Vehicle Hyperspectral Images Using Object-Oriented Methods Based on Feature Combination and Optimization

Accurate and timely acquisition of the spatial distribution of mangrove species is essential for conserving ecological diversity. Hyperspectral imaging sensors are recognized as effective tools for monitoring mangroves. However, the spatial complexity of mangrove forests and the spectral redundancy of hyperspectral images pose challenges to fine classification. Moreover, finely classifying mangrove species using only spectral information is difficult due to spectral similarities among species. To address these issues, this study proposes an object-oriented multi-feature combination method for fine classification. Specifically, hyperspectral images were segmented using multi-scale segmentation techniques to obtain different species of objects. Then, a variety of features were extracted, including spectral, vegetation indices, fractional order differential, texture, and geometric features, and a genetic algorithm was used for feature selection. Additionally, ten feature combination schemes were designed to compare the effects on mangrove species classification. In terms of classification algorithms, the classification capabilities of four machine learning classifiers were evaluated, including K-nearest neighbor (KNN), support vector machines (SVM), random forests (RF), and artificial neural networks (ANN) methods. The results indicate that SVM based on texture features achieved the highest classification accuracy among single-feature variables, with an overall accuracy of 97.04%. Among feature combination variables, ANN based on raw spectra, first-order differential spectra, texture features, vegetation indices, and geometric features achieved the highest classification accuracy, with an overall accuracy of 98.03%. Texture features and fractional order differentiation are identified as important variables, while vegetation index and geometric features can further improve classification accuracy. Object-based classification, compared to pixel-based classification, can avoid the salt-and-pepper phenomenon and significantly enhance the accuracy and efficiency of mangrove species classification. Overall, the multi-feature combination method and object-based classification strategy proposed in this study provide strong technical support for the fine classification of mangrove species and are expected to play an important role in mangrove restoration and management.

Extraction of Lilium davidii var. unicolor Planting Information Based on Deep Learning and Multi-Source Data

Accurate extraction of crop acreage is an important element of digital agriculture. This study uses Sentinel-2A, Sentinel-1, and DEM as data sources to construct a multidimensional feature dataset encompassing spectral features, vegetation index, texture features, terrain features, and radar features. The Relief-F algorithm is applied for feature selection to identify the optimal feature dataset. And the combination of deep learning and the random forest (RF) classification method is utilized to identify lilies in Qilihe District and Yuzhong County of Lanzhou City, obtain their planting structure, and analyze their spatial distribution characteristics in Gansu Province. The findings indicate that terrain features significantly contribute to ground object classification, with the highest classification accuracy when the number of features in the feature dataset is 36. The precision of the deep learning classification method exceeds that of RF, with an overall classification accuracy and kappa coefficient of 95.9% and 0.934, respectively. The Lanzhou lily planting area is 137.24 km2, and it primarily presents a concentrated and contiguous distribution feature. The study's findings can serve as a solid scientific foundation for Lanzhou City's lily planting structure adjustment and optimization and a basis of data for local lily yield forecasting, development, and application.

Automated in-season rice crop mapping using Sentinel time-series data and Google Earth Engine: A case study in climate-risk prone Bangladesh

High-resolution mapping of rice fields is crucial for understanding and managing rice cultivation in countries like Bangladesh, particularly in the face of climate change. Rice is a vital crop, cultivated in small scale farms that contributes significantly to the economy and food security in Bangladesh. Accurate mapping can facilitate improved rice production, the development of sustainable agricultural management policies, and formulation of strategies for adapting to climatic risks. To address the need for timely and accurate rice mapping, we developed a framework specifically designed for the diverse environmental conditions in Bangladesh. We utilized Sentinel-1 and Sentinel-2 time-series data to identify transplantation and peak seasons and employed the multi-Otsu automatic thresholding approach to map rice during the peak season (April-May). We also compared the performance of a random forest (RF) classifier with the multi-Otsu approach using two different data combinations: D1, which utilizes data from the transplantation and peak seasons (D1 RF) and D2, which utilizes data from the transplantation to the harvest seasons (D2 RF). Our results demonstrated that the multi-Otsu approach achieved an overall classification accuracy (OCA) ranging from 61.18% to 94.43% across all crop zones. The D2 RF showed the highest mean OCA (92.15%) among the fourteen crop zones, followed by D1 RF (89.47%) and multi-Otsu (85.27%). Although the multi-Otsu approach had relatively lower OCA, it proved effective in accurately mapping rice areas prior to harvest, eliminating the need for training samples that can be challenging to obtain during the growing season. In-season rice area maps generated through this framework are crucial for timely decision-making regarding adaptive management in response to climatic stresses and forecasting area-wide productivity. The scalability of our framework across space and time makes it particularly suitable for addressing field data scarcity challenges in countries like Bangladesh and offers the potential for future operationalization.

The 10-m cotton maps in Xinjiang, China during 2018-2021

Cotton maps (10 m) of Xinjiang (XJ_COTTON10), which is the largest cotton production region of China, were produced from 2018 to 2021 through supervised classification. A two-step mapping strategy, i.e., cropland mapping followed by cotton extraction, was employed to improve the accuracy and efficiency of cotton mapping for a large region of about 1.66 million km2 with high heterogeneity. Additionally, the time-series satellite data related to spectral, textural, structural, and phenological features were combined and used in a supervised random forest classifier. The cotton/non-cotton classification model achieved overall accuracies of about 95% and 90% on the test samples of the same and adjacent years, respectively. The proposed two-step cotton mapping strategy proved promising and effective in producing multi-year and consistent cotton maps. XJ_COTTON10 agreed well with the statistical areas of cotton at the county level (R2 = 0.84-0.94). This is the first cotton mapping for the entire Xinjiang at 10-meter resolution, which can provide a basis for high-precision cotton monitoring and policymaking in China.

Quantifying spatio-temporal variation in aquaculture production areas in Satkhira, Bangladesh using geospatial and social survey

Despite Bangladesh being one of the leading countries in aquaculture food production worldwide, there is a considerable lack of updated scientific information about aquaculture activities in remote sites, making it difficult to manage sustainably. This study explored the use of geospatial and field data to monitor spatio-temporal changes in aquaculture production sites in the Satkhira district from 2017-2019. We used Shuttle Radar Topographic Mission digital elevation model (SRTM DEM) to locate aquaculture ponds based on the terrain elevation and slope. Radar backscatter information from the Sentinel-1 satellite, and different water indices derived from Sentinel-2 were used to assess the spatio-temporal extents of aquaculture areas. An image segmentation algorithm was applied to detect aquaculture ponds based on backscattering intensity, size and shape characteristics. Our results show that the highest number of aquaculture ponds were observed in January, with a size of more than 30,000 ha. Object-based image classification of Sentinel-1 data showed an overall accuracy above 80%. The key factors responsible for the variation in aquaculture were investigated using field surveys. We noticed that despite a significant number of aquaculture ponds in the study area, shrimp production and export are decreasing because of a lack of infrastructure, poor governance, and lack of awareness in the local communities. The result of this study can provide in-depth information about aquaculture areas, which is vital for policymakers and environmental administrators for successful aquaculture management in Satkhira, Bangladesh and other countries with similar issues.

Rapid and Automated Approach for Early Crop Mapping Using Sentinel-1 and Sentinel-2 on Google Earth Engine; A Case of a Highly Heterogeneous and Fragmented Agricultural Region

Accurate and rapid crop type mapping is critical for agricultural sustainability. The growing trend of cloud-based geospatial platforms provides rapid processing tools and cloud storage for remote sensing data. In particular, a variety of remote sensing applications have made use of publicly accessible data from the Sentinel missions of the European Space Agency (ESA). However, few studies have employed these data to evaluate the effectiveness of Sentinel-1, and Sentinel-2 spectral bands and Machine Learning (ML) techniques in challenging highly heterogeneous and fragmented agricultural landscapes using the Google Earth Engine (GEE) cloud computing platform. This work aims to map, accurately and early, the crop types in a highly heterogeneous and fragmented agricultural region of the Tadla Irrigated Perimeter (TIP) as a case study using the high spatiotemporal resolution of Sentinel-1, Sentinel-2, and a Random Forest (RF) classifier implemented on GEE. More specifically, five experiments were performed to assess the optical band reflectance values, vegetation indices, and SAR backscattering coefficients on the accuracy of crop classification. Besides, two scenarios were used to assess the monthly temporal windows on classification accuracy. The findings of this study show that the fusion of Sentinel-1 and Sentinel-2 data can accurately produce the early crop mapping of the studied area with an Overall Accuracy (OA) reaching 95.02%. The scenarios prove that the monthly time series perform better in terms of classification accuracy than single monthly windows images. Red-edge and shortwave infrared bands can improve the accuracy of crop classification by 1.72% when compared to only using traditional bands (i.e., visible and near-infrared bands). The inclusion of two common vegetation indices (The Normalized Vegetation Index (NDVI), the Enhanced Vegetation Index (EVI)) and Sentinel-1 backscattering coefficients to the crop classification enhanced the overall classification accuracy by 0.02% and 2.94%, respectively, compared to using the Sentinel-2 reflectance bands alone. The monthly windows analysis indicated that the improvement in the accuracy of crop classification is the greatest when the March images are accessible, with an OA higher than 80%.

Object-Based Automatic Mapping of Winter Wheat Based on Temporal Phenology Patterns Derived from Multitemporal Sentinel-1 and Sentinel-2 Imagery

Although winter wheat has been mapped by remote sensing in several studies, such mapping efforts did not sufficiently utilize contextual information to reduce the noise and still depended heavily on optical imagery and exhausting classification approaches. Furthermore, the influence of similarity measures on winter wheat identification remains unclear. To overcome these limitations, this study developed an object-based automatic approach to map winter wheat using multitemporal Sentinel-1 (S1) and Sentinel-2 (S2) imagery. First, after S1 and S2 images were preprocessed, the Simple Non-Iterative Clustering (SNIC) algorithm was used to conduct image segmentation to obtain homogeneous spatial objects with a fusion of S1 and S2 bands. Second, the temporal phenology patterns (TPP) of winter wheat and other typical land covers were derived from object-level S1 and S2 imagery based on the collected ground truth samples, and two improved distance measures (i.e., a composite of Euclidean distance and Spectral Angle Distance, (ESD) and the difference–similarity factor distance (DSF)) were built to evaluate the similarity between two TPPs. Third, winter wheat objects were automatically identified from the segmented spatial objects by the maximum between-class variance method (OTSU) with distance measures based on the unique TPP of winter wheat. According to ground truth data, the DSF measure was superior to other distance measures in winter wheat mapping, since it achieved the best overall accuracy (OA), best kappa coefficient (Kappa) and more spatial details for each feasible band (i.e., NDVI, VV, and VH/VV), or it obtained results comparable to those for the best one (e.g., NDVI + VV). The resultant winter wheat maps derived from the NDVI band with the DSF measure achieved the best accuracy and more details, and had an average OA and Kappa of 92% and 84%, respectively. The VV polarization with the DSF measure produced the second best winter wheat maps with an average OA and Kappa of 91% and 80%, respectively. The results indicate the great potential of the proposed object-based approach for automatic winter wheat mapping for both optical and Synthetic Aperture Radar (SAR) imagery.

Google Earth Engine and Artificial Intelligence (AI): A Comprehensive Review

Remote sensing (RS) plays an important role gathering data in many critical domains (e.g., global climate change, risk assessment and vulnerability reduction of natural hazards, resilience of ecosystems, and urban planning). Retrieving, managing, and analyzing large amounts of RS imagery poses substantial challenges. Google Earth Engine (GEE) provides a scalable, cloud-based, geospatial retrieval and processing platform. GEE also provides access to the vast majority of freely available, public, multi-temporal RS data and offers free cloud-based computational power for geospatial data analysis. Artificial intelligence (AI) methods are a critical enabling technology to automating the interpretation of RS imagery, particularly on object-based domains, so the integration of AI methods into GEE represents a promising path towards operationalizing automated RS-based monitoring programs. In this article, we provide a systematic review of relevant literature to identify recent research that incorporates AI methods in GEE. We then discuss some of the major challenges of integrating GEE and AI and identify several priorities for future research. We developed an interactive web application designed to allow readers to intuitively and dynamically review the publications included in this literature review.

The Classification Method Study of Crops Remote Sensing with Deep Learning, Machine Learning, and Google Earth Engine

The extraction and classification of crops is the core issue of agricultural remote sensing. The precise classification of crop types is of great significance to the monitoring and evaluation of crops planting area, growth, and yield. Based on the Google Earth Engine and Google Colab cloud platform, this study takes the typical agricultural oasis area of Xiangride Town, Qinghai Province, as an example. It compares traditional machine learning (random forest, RF), object-oriented classification (object-oriented, OO), and deep neural networks (DNN), which proposes a random forest combined with deep neural network (RF+DNN) classification framework. In this study, the spatial characteristics of band information, vegetation index, and polarization of main crops in the study area were constructed using Sentinel-1 and Sentinel-2 data. The temporal characteristics of crops phenology and growth state were analyzed using the curve curvature method, and the data were screened in time and space. By comparing and analyzing the accuracy of the four classification methods, the advantages of RF+DNN model and its application value in crops classification were illustrated. The results showed that for the crops in the study area during the period of good growth and development, a better crop classification result could be obtained using RF+DNN classification method, whose model accuracy, training, and predict time spent were better than that of using DNN alone. The overall accuracy and Kappa coefficient of classification were 0.98 and 0.97, respectively. It is also higher than the classification accuracy of random forest (OA = 0.87, Kappa = 0.82), object oriented (OA = 0.78, Kappa = 0.70) and deep neural network (OA = 0.93, Kappa = 0.90). The scalable and simple classification method proposed in this paper gives full play to the advantages of cloud platform in data and operation, and the traditional machine learning combined with deep learning can effectively improve the classification accuracy. Timely and accurate extraction of crop types at different spatial and temporal scales is of great significance for crops pattern change, crops yield estimation, and crops safety warning.

PlanetScope, Sentinel-2, and Sentinel-1 Data Integration for Object-Based Land Cover Classification in Google Earth Engine

PlanetScope (PL) high-resolution composite base maps have recently become available within Google Earth Engine (GEE) for the tropical regions thanks to the partnership between Google and the Norway’s International Climate and Forest Initiative (NICFI). Object-based (OB) image classification in the GEE environment has increased rapidly due to the broadly recognized advantages of applying these approaches to medium- and high-resolution images. This work aimed to assess the advantages for land cover classification of (a) adopting an OB approach with PL data; and (b) integrating the PL datasets with Sentinel 2 and Sentinel 1 data both in Pixel-based (PB) or OB approaches. For this purpose, in this research, we compared ten LULC classification approaches (PB and OB, all based on the Random Forest (RF) algorithm), where the three satellite datasets were used according to different levels of integration and combination. The study area, which is 69,272 km2 wide and located in central Brazil, was selected within the tropical region, considering a preliminary availability of sample points and its complex landscape mosaic composed of heterogeneous agri-natural spaces, including scattered settlements. Using only the PL dataset with a typical RF PB approach produced the worse overall accuracy (OA) results (67%), whereas adopting an OB approach for the same dataset yielded very good OA (82%). The integration of PL data with the S2 and S1 datasets improved both PB and OB overall accuracy outputs (82 vs. 67% and 91 vs. 82%, respectively). Moreover, this research demonstrated the OB approaches’ applicability in GEE, even in vast study areas and using high-resolution imagery. Although additional applications are necessary, the proposed methodology appears to be very promising for properly exploiting the potential of PL data in GEE.

Identifying Grassland Distribution in a Mountainous Region in Southwest China Using Multi-Source Remote Sensing Images

Southwest China has abundant grassland resources, but they are mainly scattered across fragmented mountainous terrain with frequently cloudy and rainy weather, making their accurate identification by remote sensing challenging. Therefore, the goal of this study was to generate prefecture-level city-scale mountainous grassland distribution data to support the development of sustainable grassland husbandry. Here, we proposed a sample selection method and comprehensively utilized multi-source data to obtain the quasi-10 m southwest grassland distribution data. The sample selection method was to first determine the sample selection range based on multi-source land use/cover database, and then to randomly select the samples under the constraint of secondary land use types, multiple factors of terrain and pure pixels. This method can deal with the difficulty in identifying the fragmented grassland distribution caused by steep mountains and hills. In addition, a multispectral time series dataset was constructed based on the fusion of Landsat 8 OLI and Sentinel-2A/B data due to cloudy and rainy weather and was used as one of the input features along with synthetic aperture radar Sentinel-1 time series data and the terrain multi-factor data. Finally, a remote sensing method to accurately identify grassland distribution in southwest China was constructed based on the Google Earth Engine (GEE) platform. Taking Zhaotong City, a prefecture-level city in Yunnan Province, as an example, a thematic map of grassland distribution with an overall accuracy of 88.21% was obtained using the above method. This map has been used by the local government of Zhaotong City in their planning of the development of sustainable grassland husbandry.

Identification of Crop Type Based on C-AENN Using Time Series Sentinel-1A SAR Data

Crop type identification is the initial stage and an important part of the agricultural monitoring system. It is well known that synthetic aperture radar (SAR) Sentinel-1A imagery provides a reliable data source for crop type identification. However, a single-temporal SAR image does not contain enough features, and the unique physical characteristics of radar images are relatively lacking, which limits its potential in crop mapping. In addition, current methods may not be applicable for time -series SAR data. To address the above issues, a new crop type identification method was proposed. Specifically, a farmland mask was firstly generated by the object Markov random field (OMRF) model to remove the interference of non-farmland factors. Then, the features of the standard backscatter coefficient, Sigma-naught (σ0), and the normalized backscatter coefficient by the incident angle, Gamma-naught (γ0), were extracted for each type of crop, and the optimal feature combination was found from time -series SAR images by means of Jeffries-Matusita (J-M) distance analysis. Finally, to make efficient utilization of optimal multi-temporal feature combination, a new network, the convolutional-autoencoder neural network (C-AENN), was developed for the crop type identification task. In order to prove the effectiveness of the method, several classical machine learning methods such as support vector machine (SVM), random forest (RF), etc., and deep learning methods such as one dimensional convolutional neural network (1D-CNN) and stacked auto-encoder (SAE), etc., were used for comparison. In terms of quantitative assessment, the proposed method achieved the highest accuracy, with a macro-F1 score of 0.9825, an overall accuracy (OA) score of 0.9794, and a Kappa coefficient (Kappa) score of 0.9705. In terms of qualitative assessment, four typical regions were chosen for intuitive comparison with the sample maps, and the identification result covering the study area was compared with a contemporaneous optical image, which indicated the high accuracy of the proposed method. In short, this study enables the effective identification of crop types, which demonstrates the importance of multi-temporal radar images in feature combination and the necessity of deep learning networks to extract complex features.

Research on cropping intensity mapping of the Huai River Basin (China) based on multi-source remote sensing data fusion

As a key input variable to many global climates, land surfaces and crop models, cropping intensity (CI) accurately assesses and predicts crops' output, in view of the global decline in food production in recent years due to declining natural resources, urban expansion and declining quality of arable land. Hence, research on CI mapping can have a contribution to solve this problem. Unfortunately, existing remote sensing data for CI mapping research, including Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat images, are not adequate for obtaining CI information at higher spatial and temporal resolution. In this regard, we develop an algorithm to extract CI based on per-pixel physiognomy. To be specific, the algorithm is based on the Google Earth Engine (GEE) platform and constructs a high temporal (10 days) spatial (30 m) resolution dataset with the fusion of Landsat 7/8 and Sentinel-2 A/B image data and extracts CI information using a time series of peak discovery method, threshold method and phenological period feature extraction to obtain the 2018 Chinese Huai River Basin (HRB) CI map. Our results suggest that the overall accuracy (OA) of CI extraction in the HRB is 92.72%, with a kappa coefficient of 0.864. The single-season crop, double-season crop and three-season crop account for 41.6%, 57.7% and 0.7% of the total farmland area, respectively. Compared to existing CI identification and extraction methods, this approach achieves higher accuracy in the identification and extraction of CI information over a larger area.

Crop Mapping in the Sanjiang Plain Using an Improved Object-Oriented Method Based on Google Earth Engine and Combined Growth Period Attributes

In recent years, the scale of rural land transfer has gradually expanded, and the phenomenon of non-grain-oriented cultivated land has emerged. Obtaining crop planting information is of the utmost importance to guaranteeing national food security; however, the acquisition of the spatial distribution of crops in large-scale areas often has the disadvantages of excessive calculation and low accuracy. Therefore, the IO-Growth method, which takes the growth stage every 10 days as the index and combines the spectral features of crops to refine the effective interval of conventional wavebands for object-oriented classification, was proposed. The results were as follows: (1) the IO-Growth method obtained classification results with an overall accuracy and F1 score of 0.92, and both values increased by 6.98% compared to the method applied without growth stages; (2) the IO-Growth method reduced 288 features to only 5 features, namely Sentinel-2: Red Edge1, normalized difference vegetation index, Red, short-wave infrared2, and Aerosols, on the 261st to 270th days, which greatly improved the utilization rate of the wavebands; (3) the rise of geographic data processing platforms makes it simple to complete computations with massive data in a short time. The results showed that the IO-Growth method is suitable for large-scale vegetation mapping.