Mapping cotton cultivated area combining remote sensing with a fused representation-based classification algorithm

Xun, Lan; Zhang, Jiahua; Cao, Dan; Wang, Jingwen; Zhang, Sha; Yao, Fengmei

doi:10.1016/j.compag.2020.105940

Cited by 30 publications

(16 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Huang et al [22] developed a crop model and data assimilation framework that assimilated leaf area index (LAI) derived from time-series Landsat TM and MODIS data to estimate regional wheat yield predictions. Xun et al [23] explored the feasibility of combining time-series enhanced vegetation index (EVI) computed from MODIS satellite data with a fused representation-based classification (FRC) algorithm to identify cotton pixels and map cotton acreage. However, due to the mixed pixels of low-and medium-resolution (hundreds of meters) remote sensing images, it is difficult to obtain accurate classification results.…”

Section: Introductionmentioning

confidence: 99%

Cotton Classification Method at the County Scale Based on Multi-Features and Random Forest Feature Selection Algorithm and Classifier

et al. 2022

View full text Add to dashboard Cite

Accurate cotton maps are crucial for monitoring cotton growth and precision management. The paper proposed a county-scale cotton mapping method by using random forest (RF) feature selection algorithm and classifier based on selecting multi-features, including spectral, vegetation indices, and texture features. The contribution of texture features to cotton classification accuracy was also explored in addition to spectral features and vegetation index. In addition, the optimal classification time, feature importance, and the best classifier on the cotton extraction accuracy were evaluated. The results showed that the texture feature named the gray level co-occurrence matrix (GLCM) is effective for improving classification accuracy, ranking second in contribution among all studied spectral, VI, and texture features. Among the three classifiers, the RF showed higher accuracy and better stability than support vector machines (SVM) and artificial neural networks (ANN). The average overall accuracy (OA) of the classification combining multiple features was 93.36%, 7.33% higher than the average OA of the single-time spectrum, and 2.05% higher than the average OA of the multi-time spectrum. The classification accuracy after feature selection by RF can still reach 92.12%, showing high accuracy and efficiency. Combining multiple features and random forest methods may be a promising county-scale cotton classification method.

show abstract

Section: Introductionmentioning

confidence: 99%

Cotton Classification Method at the County Scale Based on Multi-Features and Random Forest Feature Selection Algorithm and Classifier

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In most traditional threshold methods, in order to obtain a group of thresholds with high classification accuracy, the threshold parameters of the key phenological periods need to be manually adjusted many times [60]. In recent years, a series of studies on reasonable threshold setting have been carried out [61,62], but the automation of threshold setting still needs improvement. In this study, the SCE-UA global optimization algorithm was used to realize the automatic optimization of the threshold in the winter wheat extraction model, which not only avoided the time-consuming manual adjustment of parameters, but also improved the accuracy of crop distribution extraction and mapping.…”

Section: Threshold Optimization Of the Crop Extraction Modelmentioning

confidence: 99%

Comparison of Regional Winter Wheat Mapping Results from Different Similarity Measurement Indicators of NDVI Time Series and Their Optimized Thresholds

Ren

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Generally, there is an inconsistency between the total regional crop area that was obtained from remote sensing technology and the official statistical data on crop areas. When performing scale conversion and data aggregation of remote sensing-based crop mapping results from different administrative scales, it is difficult to obtain accurate crop planting area that match crop area statistics well at the corresponding administrative level. This problem affects the application of remote sensing-based crop mapping results. In order to solve the above problem, taking Fucheng County of Hebei Province in the Huanghuaihai Plain of China as the study area, based on the Sentinel-2 normalized difference vegetation index (NDVI) time series data covering the whole winter wheat growth period, the statistical data of the regional winter wheat planting area were regarded as reference for the winter wheat planting area extracted by remote sensing, and a new method for winter wheat mapping that is based on similarity measurement indicators and their threshold optimizations (WWM-SMITO) was proposed with the support of the shuffled complex evolution-University of Arizona (SCE-UA) global optimization algorithm. The accuracy of the regional winter wheat mapping results was verified, and accuracy comparisons with different similarity indicators were carried out. The results showed that the total area accuracy of the winter wheat area extraction by the proposed method reached over 99.99%, which achieved a consistency that was between the regional remote sensing-based winter wheat planting area and the statistical data on the winter wheat planting area. The crop recognition accuracy also reached a high level, which showed that the proposed method was effective and feasible. Moreover, in the accuracy comparison of crop mapping results based on six different similarity indicators, the winter wheat distribution that was extracted by root mean square error (RMSE) had the best recognition accuracy, and the overall accuracy and kappa coefficient were 94.5% and 0.8894, respectively. The overall accuracies of winter wheat that were extracted by similarity indicators, such as Euclidean distance (ED), Manhattan distance (MD), spectral angle mapping (SAM), and spectral correlation coefficient (SCC) were 94.1%, 93.9%, 93.3%, and 92.8%, respectively, and the kappa coefficients were 0.8815, 0.8776, 0.8657, and 0.8558, respectively. The accuracy of the winter wheat results extracted by the similarity indicator of dynamic time warping (DTW) was relatively low. The results of this paper could provide guidance and serve as a reference for the selection of similarity indicators in crop distribution extraction and for obtaining large-scale, long-term, and high-precision remote sensing-based information on a regional crop spatial distribution that is highly consistent with statistical crop area data.

show abstract

“…The above classification models are all based on machine learning algorithms. Currently, deep learning algorithms (DL), represented by convolutional neural networks (CNN), Sparse Coding, and Deep Belief Network (DBN), are gradually being used in RS vegetation classification owing to their ability of deep feature mining [83]. DL with deep network structures allows end-to-end learning thus it can extract deep characteristics of vegetation from RS images without human intervention [84].…”

Section: The Effect Of Classification Model On Classification Accuracymentioning

confidence: 99%

Superpixel-Based Regional-Scale Grassland Community Classification Using Genetic Programming with Sentinel-1 SAR and Sentinel-2 Multispectral Images

Zhang

Fan

et al. 2021

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Grasslands are one of the most important terrestrial ecosystems on the planet and have significant economic and ecological value. Accurate and rapid discrimination of grassland communities is critical to the conservation and utilization of grassland resources. Previous studies that explored grassland communities were mainly based on field surveys or airborne hyperspectral and high-resolution imagery. Limited by workload and cost, these methods are typically suitable for small areas. Spaceborne mid-resolution RS images (e.g., Sentinel, Landsat) have been widely used for large-scale vegetation observations owing to their large swath width. However, there still keep challenges in accurately distinguishing between different grassland communities using these images because of the strong spectral similarity of different communities and the suboptimal performance of models used for classification. To address this issue, this paper proposed a superpixel-based grassland community classification method using Genetic Programming (GP)-optimized classification model with Sentinel-2 multispectral bands, their derived vegetation indices (VIs) and textural features, and Sentinel-1 Synthetic Aperture Radar (SAR) bands and the derived textural features. The proposed method was evaluated in the Siziwang grassland of China. Our results showed that the addition of VIs and textures, as well as the use of GP-optimized classification models, can significantly contribute to distinguishing grassland communities, and the proposed approach classified the seven communities in Siziwang grassland with an overall accuracy of 84.21% and a kappa coefficient of 0.81. We concluded that the classification method proposed in this paper is capable of distinguishing grassland communities with high accuracy at a regional scale.

show abstract

Mapping cotton cultivated area combining remote sensing with a fused representation-based classification algorithm

Cited by 30 publications

References 50 publications

Cotton Classification Method at the County Scale Based on Multi-Features and Random Forest Feature Selection Algorithm and Classifier

Cotton Classification Method at the County Scale Based on Multi-Features and Random Forest Feature Selection Algorithm and Classifier

Comparison of Regional Winter Wheat Mapping Results from Different Similarity Measurement Indicators of NDVI Time Series and Their Optimized Thresholds

Superpixel-Based Regional-Scale Grassland Community Classification Using Genetic Programming with Sentinel-1 SAR and Sentinel-2 Multispectral Images

Contact Info

Product

Resources

About