Agricultural Crop Classification Research Articles

Classification of Agricultural Crops with Random Forest and Support Vector Machine Algorithms Using Sentinel-2 and Landsat-8 Images

Monitoring crop development and mapping cultivated areas are important for reducing risks to food security due to climate change. Remote sensing techniques contribute significantly to the efficient and effective management of agricultural production. In this study, agricultural fields (sunflower, wheat, maize, oat, chickpea, sugar beet, alfalfa, onion, fallow) and other fields (non-agricultural, pasture, lake) were identified by using Random Forest (RF) and Support Vector Machines (SVM) machine learning algorithms with Sentinel-2 and Landsat-8 images in the area covering Polatlı, Haymana and Gölbaşı districts of Ankara province Multi-temporal images were used to distinguish winter and summer crops, taking into account crop development periods. As a result of classification; the overall accuracy of RF and SVM models with S2 images are 89.5% and 84.6% and kappa coefficients are 0.88 and 0.83, while the overall accuracy of RF and SVM models with L8 images are 79% and 78.1% and kappa coefficients are 0.76 and 0.75. RF model was found to have higher prediction accuracy than SVM. Sentinel-2 imagery has a higher accuracy in all classes compared to Landsat-8, indicating that Sentinel-2 imagery with its high temporal and spatial resolution is more suitable and has a great potential for agricultural crop pattern detection.

Enhancing food crop classification in agriculture through dipper throat optimization and deep learning with remote sensing

Remote sensing images (RSIs), a keystone of modern agricultural technology, refer to spectral or visual data captured from drones, satellites, or aircraft without direct physical contact with the Earth's surface. These images provide a wide-ranging view of agricultural landscapes, providing valuable insights into land use, crop health, and environmental conditions. Agricultural food crop classification, a vital application within precision agriculture, includes the detection and classification of different crops cultivated in a certain region. Traditionally reliant on manual techniques, the development of technologies, particularly the incorporation of RSIs, has revolutionized this process. Agricultural food crop classification has become more sophisticated and automated by harnessing the wealth of data received from RS, which facilitates precise management and monitoring of crops on a large scale. Deep learning (DL), a branch of artificial intelligence, plays a more effective role in these synergies. The incorporation of DL into the RSI analysis enables high-precision and efficient detection of various crop types, assisting more informed decision-making in agriculture. This study proposes a new Dipper Throat Optimization Algorithm with Deep Learning based Food Crop Classification (DTOADL-FCC) algorithm using Remote Sensing Imaging for Agricultural Resource Management. The DTOADL-FCC method aims to apply DL algorithms for the classification of different crop types. In the DTOADL-FCC method, fully convolutional network (FCN) based segmentation process is performed. Next, the DTOADL-FCC method exploits the SE-ResNet model for learning intrinsic and complex features. The DTOADL-FCC method makes use of DTOA for the hyperparameter tuning process. Lastly, the classification of crop types takes place using the extreme learning machine (ELM) model. The study utilizes mathematical formulations including activation functions, loss functions, fitness calculations, and iterative update processes. A brief set of simulations showcases that the DTOADL-FCC method achieves remarkable performance over other techniques with much improved results.

Machine Learning-Based Fine Classification of Agricultural Crops in the Cross-Border Basin of the Heilongjiang River between China and Russia

The transboundary region along the Heilongjiang River, encompassing the Russian Far East and Northeast China, possesses abundant agricultural natural resources crucial for global food security. In the face of the challenge of disruptions in the global food supply chain, the precise monitoring and exploitation of agricultural resources in the Heilongjiang Basin becomes imperative. This study employed deep learning to classify crop status in 2023 in the Heilongjiang Basin using Sentinel-2 satellite remote sensing images at a 10 m resolution. Various vegetation indices, including the Normalized Difference Vegetation Index (NDVI), the Normalized Difference Water Index (NDWI), the Enhanced Vegetation Index (EVI), the Modified Soil Adjusted Vegetation Index (MSAVI), and others, were computed and analyzed for different crops. The Google Earth Engine (GEE) platform was utilized for validation point sampling based on plot objects. The random forest (RF) classification method was successfully employed to classify and identify major crops in the study area (wheat, maize, rice, and soybean), as well as wetlands, tree cover, grassland, water, and constructed land, with an overall classification accuracy of 86%. Tree cover dominated the land cover, constituting 62%, while wheat, maize, rice, and soybeans accounted for 7% of the total area. Of these, soybeans occupied the largest area (57,646.60 hectares), followed by rice (53,209.53 hectares), maize (39,998.37 hectares), and wheat (8782.31 hectares). This study demonstrated that sample selection based on plot objects facilitates efficient sample labeling, providing insights into crop classification in other, potentially larger, areas. This method simultaneously distinguishes wetland, cultivated land, and forest features, supporting further integrated investigations for more natural resources.

Leveraging Hyperspectral Remote Sensing Imaging for Agricultural Crop Classification using Coot Bird Optimization with Entropy-based Feature Fusion Model

Leveraging Hyperspectral Remote Sensing Imaging for Agricultural Crop Classification using Coot Bird Optimization with Entropy-based Feature Fusion Model

Classification of crops by NDVI time series of reduced dimensionality

The paper considers the problem of classification of agricultural crops. As is known, to solve this problem, it is much more efficient to use not instantaneous remote sensing data or calculated vegetation indices, but their historical series. Time series formed by index values for a fixed spatial point at different dates are characterized by a high level of missing values, caused primarily by cloudiness on some dates. A study of known methods of time series approximation has been carried out. The question of whether reducing the dimensionality of the approximated time series can improve the quality of crops classification is also investigated. In the experimental part of the work, NDVI time series calculated from the Sentinel-2 multispectral satellite data were used. The classification of corn, sunflower, wheat and soybeans was studied. The paper shows that UMAP usage for dimensionality reduction leads to 1.5 times increase of classification quality in terms of average the F1-measure compared to using the original dimension data. A new crop classification method based on cubic spline approximation of NDVI time series, extraction of features of low dimension by the UMAP algorithm and their classification by the k nearest neighbors method is proposed.

Fusion of spectral and spatial information for agricultural crop classification in multi-temporal Sentinel images (Case Study: Qorveh County)

امروزه کاربردهای تصاویر ماهواره‌ای، در پایش و مدیریت زمین‌های کشاورزی، رو به گسترش است. با توجه به قدرت تفکیک مکانی، طیفی و زمانی بالای تصاویر سنتینل‌ـ 2، در این مطالعه، در کشاورزی دقیق در شهرستان قروه از این تصاویر استفاده شده است. ابتدا با توجه به تقویم زراعی محصولات متفاوت آن منطقه، تصاویر سری زمانی جمع‌آوری شد. در روش پیشنهادی، نخست، فضای ویژگی طیفی براساس بازتاب طیفی باندها و همچنین شاخص‌های گیاهی، ایجاد شد. ابعاد فضای ویژگی طیفی، با استفاده از روش آنالیز مؤلفه‌های اصلی، کاهش یافت. سپس چهار طبقه‌بندی‌کنندة قدرتمند ماشین‌های بردار پشتیبان، شبکة عصبی پرسپترون چندلایه، نزدیک‌ترین k همسایه و جنگل‌های تصادفی نقشة طبقه‌بندی از اطلاعات طیفی تولید کردند. در ادامه، مکانی با هدف تعیین مرز مزارع، اطلاعات استخراج شد. برای این منظور، از شناسایی لبه‌ها در سری زمانی تصاویر سنتینل‌ـ 2 استفاده شد. در نهایت، نقشة طبقه‌بندی نهایی، با تلفیق اطلاعات مکانی و ادغام نتایج طبقه‌بندی‌کننده‌ها ایجاد شد. نتایج به‌دست‌آمده نشان داد که دقت طبقه‌بندی‌کننده‌های نزدیک‌ترین k همسایه، ماشین‌های بردار پشتیبان، شبکة عصبی پرسپترون چندلایه و جنگل‌های تصادفی روی فضای ویژگی طیفی اولیه، به‌ترتیب 78/77%، 16/79%، 41/76% و 89/76% است. با استفاده از روش پیشنهادی، دقت طبقه‌بندی به 72/94% افزایش پیدا کرد که حاکی از توانایی آن در منطقة مورد مطالعه است.

Açık kaynak kodlu Eo-learn kütüphanesi ve çok zamanlı Sentinel-2 görüntüleri ile tarımsal ürün sınıflandırması

In this study, agricultural crop classification for 2020 was carried out in Çivril-Baklan Plain, which is located between the borders of Denizli Province, Baklan, Çal and Çivril districts. The open-source Eo-Learn library that uses machine learning and deep learning algorithms in remote sensing studies and multi-temporal Sentinel-2 images was utilized in the classification process. In this study, the parcels registered in the Farmer Registration System (FRS) were used as reference parcels and before using FRS data as ground truth data, pre-editing and rule-based deletion processes were performed. By using Light Gradient Boosting Machines (LightGBM) algorithm, agricultural product pattern classification was carried out including cereal, maize, sugar beet, sunflower, hash, vineyard, fruit tree and clover crops. Classification results were evaluated using k-fold cross-validation with an overall accuracy of %93.5. A second accuracy assessment was performed with Agricultural Insurance Parcels (TARSİM) that were not included in the classification process as training data, achieving an overall accuracy of %91.1 and Kappa coefficient of 0.89.

CCTNet: Coupled CNN and Transformer Network for Crop Segmentation of Remote Sensing Images

Semantic segmentation by using remote sensing images is an efficient method for agricultural crop classification. Recent solutions in crop segmentation are mainly deep-learning-based methods, including two mainstream architectures: Convolutional Neural Networks (CNNs) and Transformer. However, these two architectures are not sufficiently good for the crop segmentation task due to the following three reasons. First, the ultra-high-resolution images need to be cut into small patches before processing, which leads to the incomplete structure of different categories’ edges. Second, because of the deficiency of global information, categories inside the crop field may be wrongly classified. Third, to restore complete images, the patches need to be spliced together, causing the edge artifacts and small misclassified objects and holes. Therefore, we proposed a novel architecture named the Coupled CNN and Transformer Network (CCTNet), which combines the local details (e.g., edge and texture) by the CNN and global context by Transformer to cope with the aforementioned problems. In particular, two modules, namely the Light Adaptive Fusion Module (LAFM) and the Coupled Attention Fusion Module (CAFM), are also designed to efficiently fuse these advantages. Meanwhile, three effective methods named Overlapping Sliding Window (OSW), Testing Time Augmentation (TTA), and Post-Processing (PP) are proposed to remove small objects and holes embedded in the inference stage and restore complete images. The experimental results evaluated on the Barley Remote Sensing Dataset present that the CCTNet outperformed the single CNN or Transformer methods, achieving 72.97% mean Intersection over Union (mIoU) scores. As a consequence, it is believed that the proposed CCTNet can be a competitive method for crop segmentation by remote sensing images.

Application of machine learning methods for classification of agricultural crops

Application of machine learning methods for classification of agricultural crops

Influence of Data Representations and Deep Architectures in Image Time Series Classification

Image time series, such as Satellite Image Time Series (SITS) or MRI functional sequences in the medical domain, carry both spatial and temporal information. In many pattern recognition applications such as image classification, taking into account such rich information may be crucial and discriminative during the decision making stage. However, the extraction of spatio-temporal features from image time series is difficult to handle due to the complex representation of the data cube. In this paper, we present a strategy based on Random Walk to build a novel segment-based representation of the data, passing from a 2D[Formula: see text] dimension to a 2D one, more easily manipulable and without losing too much spatial information. Such new representation is then used to feed a classical Convolutional Neural Network (CNN) in order to learn spatio-temporal features with only 2D convolutions and to classify image time series data for a particular classification problem. The influence of the way the 2D[Formula: see text] data are represented, as well as the impact of the network architectures on the results, are carefully studied. The interest of this approach is highlighted on a remote sensing application for the classification of complex agricultural crops.

Deep-STaR: Classification of image time series based on spatio-temporal representations

Image time series (ITS) represent complex 3D (2D+t in practice) data that are now daily produced in various domains, from medical imaging to remote sensing. They contain rich spatio-temporal information allowing the observation of the evolution of a sensed scene over time. In this work, we focus on the classification task of ITS, as often available in remote sensing tasks. An underlying problem here is to consider jointly the spatial and the temporal dimensions of the data. We present Deep-STaR, a method to learn such features from ITS data to proceed to their classification. Instead of reasoning in the original 2D+t space, we investigate novel 2D planar data representations, containing both temporal and spatial information. Such representations are a novel way to structure the ITS, compatible with deep learning architectures. They are used to feed a convolutional neural network to learn spatio-temporal features with 2D convolutions, leading ultimately to classification decision. To enhance the explainability of the results, we also propose a post-hoc attention mechanism, enabled by this new approach, providing a semantic map giving some insights for the taken decision. Deep-STaR is evaluated on a remote sensing application, for the classification of agricultural crops from satellite ITS. The results highlight the benefice of this method, compared to the literature, and its interest to make easier the interpretation of ITS to understand spatio-temporal phenomena.

Crop Monitoring and Classification Using Polarimetric RADARSAT-2 Time-Series Data Across Growing Season: A Case Study in Southwestern Ontario, Canada

Multitemporal polarimetric synthetic aperture radar (PolSAR) has proven as a very effective technique in agricultural monitoring and crop classification. This study presents a comprehensive evaluation of crop monitoring and classification over an agricultural area in southwestern Ontario, Canada. The time-series RADARSAT-2 C-Band PolSAR images throughout the entire growing season were exploited. A set of 27 representative polarimetric observables categorized into ten groups was selected and analyzed in this research. First, responses and temporal evolutions of each of the polarimetric observables over different crop types were quantitatively analyzed. The results reveal that the backscattering coefficients in cross-pol and Pauli second channel, the backscattering ratio between HV and VV channels (HV/VV), the polarimetric decomposition outputs, the correlation coefficient between HH and VV channelρ ρHHVV, and the radar vegetation index (RVI) show the highest sensitivity to crop growth. Then, the capability of PolSAR time-series data of the same beam mode was also explored for crop classification using the Random Forest (RF) algorithm. The results using single groups of polarimetric observables show that polarimetric decompositions, backscattering coefficients in Pauli and linear polarimetric channels, and correlation coefficients produced the best classification accuracies, with overall accuracies (OAs) higher than 87%. A forward selection procedure to pursue optimal classification accuracy was expanded to different perspectives, enabling an optimal combination of polarimetric observables and/or multitemporal SAR images. The results of optimal classifications show that a few polarimetric observables or a few images on certain critical dates may produce better accuracies than the whole dataset. The best result was achieved using an optimal combination of eight groups of polarimetric observables and six SAR images, with an OA of 94.04%. This suggests that an optimal combination considering both perspectives may be valuable for crop classification, which could serve as a guideline and is transferable for future research.

Classification and status monitoring of agricultural crops in central Morocco: a synergistic combination of OBIA approach and fused Landsat-Sentinel-2 data

Crop type mapping provides essential information to control and make decisions related to agricultural practices and their regulations. To map crop types accurately, it is important to capture their phenological stages and fine spatial details, especially in a temporally and spatially heterogeneous landscape. The data availability of new generation multispectral sensors of Landsat-8 (L8) and Sentinel-2 (S2) satellites offers unprecedented options for such applications. Given this, our study aims to display how the synergistic use of these optical sensors can efficiently support crop type mapping research while integrating an object-based image analysis (OBIA). Through the applied methods, we used the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) and the flexible spatiotemporal data-fusion model (FSDAF) to blend L8 and S2 data and obtain reliable normalized difference vegetation index (NDVI) datasets with fine spatial and temporal resolution. Then the crop phenological information was extracted using a Savitzky–Golay filter and fused NDVI time series. Finally, a model combining phenological metrics and fused reconstructed NDVI as classification features was developed using a random forest (RF) classifier/OBIA approach. The results show that the FSDAF method creates more accurate fused NDVI and keeps more spatial details than ESTARFM. The FSDAF model was then used to create fused, high-resolution time-series products that were able to extract crop phenology in single-crop fields while providing a very detailed pattern relative to that from individual sensor time-series data. Moreover, combined L8 and S2 data by FSDAF produced highly significant overall classification accuracies (90.03% for pixel-based RF to 93.12% OBIA RF), outperforming individual sensor use (82.57% for L8-only; 88.45% for S2-only). Our proposed workflow highlights the advantage of spatiotemporal fusing and OBIA environment in spatiotemporally heterogeneous areas and fragmented landscapes, which represents a promising step toward generating fast, accurate, and ready-to-use agricultural data products.

Combining ASNARO-2 XSAR HH and Sentinel-1 C-SAR VH/VV Polarization Data for Improved Crop Mapping

The Advanced Satellite with New system ARchitecture for Observation-2 (ASNARO-2), which carries the X-band Synthetic Aperture Radar (XSAR), was launched on 17 January 2018 and is expected to be used to supplement data provided by larger satellites. Land cover classification is one of the most common applications of remote sensing, and the results provide a reliable resource for agricultural field management and estimating potential harvests. This paper describes the results of the first experiments in which ASNARO-2 XSAR data were applied for agricultural crop classification. In previous studies, Sentinel-1 C-SAR data have been widely utilized to identify crop types. Comparisons between ASNARO-2 XSAR and Sentinel-1 C-SAR using data obtained in June and August 2018 were conducted to identify five crop types (beans, beetroot, maize, potato, and winter wheat), and the combination of these data was also tested. To assess the potential for accurate crop classification, some radar vegetation indices were calculated from the backscattering coefficients for two dates. In addition, the potential of each type of SAR data was evaluated using four popular supervised learning models: Support vector machine (SVM), random forest (RF), multilayer feedforward neural network (FNN), and kernel-based extreme learning machine (KELM). The combination of ASNARO-2 XSAR and Sentinel-1 C-SAR data was effective, and overall classification accuracies of 85.4 ± 1.8% were achieved using SVM.

Classification of Agricultural Crops from Middle-Resolution Satellite Images Using Gaussian Processes Based Method

Classification of Agricultural Crops from Middle-Resolution Satellite Images Using Gaussian Processes Based Method

SAR image time-series analysis framework using morphological operators and global and local information-based linear discriminant analysis

Fusion of spectral, spatial, and temporal information is an effective method used in many satellite remote sensing applications. On the other hand, one drawback of this fusion is an increase in complexity. In this paper, we focus on developing a fast and well-performed classification method for agricultural crops using time-series SAR data. In order to achieve this, a novel two-stage approach is proposed. In the first stage, a high-dimensional feature space is obtained using time-series dual-pol SAR data and morphological operators. Spectral, spatial, and temporal information is combined into a single high-dimensional feature space. In the second stage, a dimension reduction technique is applied to the feature vector in order to decrease time complexity and increase classification accuracy by considering the global and local pattern information in the high-dimensional feature space. The contribution of the morphological profiles to the classification performance is significant; however the time complexity is increased drastically. The proposed method overcomes the time complexity stemming from high-dimensional feature space; it also improves the classification performance. The superiority of the proposed method to the comparative methods in agricultural crop classification is experimentally shown with the improvements in both classification and time performance using time-series TerraSAR-X images.

A Survey and Analysis of Various Agricultural Crops Classification Techniques

Mining is way of providing and extracting some meaning information from the data so that the data can be classified and grouped easily and quickly. These mining algorithms can be applied in various fields including classification of agricultural crops production. In the fields of Data Mining various efficient algorithms are implemented for the classification of agricultural crops production. Here in this paper a survey of all the existing techniques as well as their advantages and issues are discussed. Hence by analyzing their various advantages and issues a new and efficient technique for the classification of agricultural crops production is proposed in future such as classification using Fuzzy Conclusion Tree by the Optimizing the Feature Withdrawal using PSO-SVM (Particle Swarm Optimization with Support Vector Machine).

Identification of Agricultural Crop Types in Northern Israel using Multitemporal RapidEye Data

Summary: Accurate land use / land cover classifi-cation (LU/LC) of agricultural crops still repre-sents a major challenge for multispectral remotesensing. In order to obtain reliable classificationaccuracies on the basis of multispectral satellitedata,mergingcropclassesinratherbroadclassesisoftennecessary.Withregardtotherising availabil-ity and the improving spatial resolution of satellitedata, multitemporal analyses become increasinglyimportant for remote sensing investigations. Forthe separation of spectrally similar crops, multi-datesatelliteimagesinclude differentgrowthchar-acteristics duringthephenologicalperiod.Thepre-sent study aims at investigating a way to performhighlyaccurateclassificationswithnumerousagri-cultural classesusing multitemporalRapidEyedata. The Jeffries-Matusita separability (JM) wasused for applying a pre-procedure in order to findthe best multitemporal setting of all available im-ages withinone crop cycle, consisting of twoculti-vation periods P1 with 16 agricultural classes andP2 with 27 agricultural classes. Only one criticalclass pairing occurred for both P1 and P2 takinginto account the best multitemporal dataset. Themaximum likelihood (ML) classifier and the sup-port vector machine (SVM) were compared usingthemostsuitable multitemporalimages.Bothalgo-rithms achieved very high overall accuracies(OAA)of over 90%. SVM was slightly better witha classification accuracy of P1-OAA = 96.13% andP2-OAA=94.01%. MLprovidedaresult of OAA =94.83% correctly classified pixels for P1 and OAA= 93.28% for P2. Theprocessingtimeof ML,how-ever, was significantly shorter compared to SVM,infact by a factor of five.

Agricultural Crops Classification Models Based on PCA-GA Implementation in Data Mining

Extraction of knowledge in agricultural data is a challenging task, from discovering patterns and relationships and interpretation. In order to obtain potentially interesting patterns and relationships from this data, it is therefore essential that a methodology be developed and take advantage of the sets of existing methods and tools available for data mining and knowledge discovery in databases. Data mining is relatively a new approach in the field of agriculture. Accurate information in characterizing crops depends on climatic, geographical, biological and other factors. These are very important inputs to generate characterization and prediction models in data mining. In this study, an efficient data mining methodology based on PCA-GA is explored, presented and implemented to characterize agricultural crops. The method draws improvements to classification problems by using Principal Components Analysis (PCA) as a pre processing method and a modified Genetic Algorithm (GA) as the function optimizer. The fitness function in GA is modified accordingly using efficient distance measures. The approach is to asses, the PCA-GA hybrid data mining method, using various agricultural field data sets, generate data mining classification models and establish meaningful relationships. The experimental results show improved classification rates and generated characterization models for agricultural crops. The domain model outcome may have benefits, to agricultural researchers and farmers. These generated classification models can also be utilized and readily incorporated into a decision support system.

Discrimination between Babul plantations and Mustard crop using Polarimetric C- and L- band data

Radar sensors can be used for large-scale vegetation mapping and monitoring using backscattering coefficients in different polarizations and wavelength bands. C-band space borne SAR is widely used for the classification of agricultural crops, but can only perform a limited discrimination of various tree species. This paper presents the results of discrimination between mustard crop and babul plantation (Prosopis sp.) using quad polarisation Radarsat 2 and ALOS PALSAR data. Study area is comprised of dense babul plantation along the canal, mustard crop on one side of the canal and Fallow land near to Ramgarh village of Jaisalmer district. Three bands of Radarsat (HH, HV and VV) acquired during peak mustard crop growth stage were integrated with four polarizations (HH, HV, VH and VV) of ALOS PALSAR acquired when crop cover was absent. Using only Radarsat data Jefferies-Matusita (JM) separability between mustard crop and babul plantation was found to be poor (710). Where as in the seven band combination the separability was observed to be high (1374). Among the different polarizations three layer combination, highest separability was observed using cross polarizations (HV and VH) of L-band with any one of the Radarsat Polarisation (HH/HV/VV). This combination of C- and L-band resulted in easy separation of mustard and babul plantation which was otherwise difficult using only Radarsat data.