Towards improved land use mapping of irrigated croplands: performance assessment of different image classification algorithms and approaches

Basukala, Amit Kumar; Oldenburg, Carsten; Schellberg, J.; Sultanov, Murodjon; Dubovyk, Olena

doi:10.1080/22797254.2017.1308235

Cited by 45 publications

(26 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It should be mentioned that in Figure 1 , the high number of papers with the keyword MLC does not mean that there is much research using MLC for classification. In fact, most studies from our searched list used the MLC method as one of the criteria to compare to other machine learning algorithms [ 11 , 29 , 30 ].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

Phan

Kappas

2017

Sensors

831

303

View full text Add to dashboard Cite

In previous classification studies, three non-parametric classifiers, Random Forest (RF), k-Nearest Neighbor (kNN), and Support Vector Machine (SVM), were reported as the foremost classifiers at producing high accuracies. However, only a few studies have compared the performances of these classifiers with different training sample sizes for the same remote sensing images, particularly the Sentinel-2 Multispectral Imager (MSI). In this study, we examined and compared the performances of the RF, kNN, and SVM classifiers for land use/cover classification using Sentinel-2 image data. An area of 30 × 30 km2 within the Red River Delta of Vietnam with six land use/cover types was classified using 14 different training sample sizes, including balanced and imbalanced, from 50 to over 1250 pixels/class. All classification results showed a high overall accuracy (OA) ranging from 90% to 95%. Among the three classifiers and 14 sub-datasets, SVM produced the highest OA with the least sensitivity to the training sample sizes, followed consecutively by RF and kNN. In relation to the sample size, all three classifiers showed a similar and high OA (over 93.85%) when the training sample size was large enough, i.e., greater than 750 pixels/class or representing an area of approximately 0.25% of the total study area. The high accuracy was achieved with both imbalanced and balanced datasets.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

Phan

Kappas

2017

Sensors

831

303

View full text Add to dashboard Cite

show abstract

“…However, classification does not seem to be as reliably specific for potatoes, as may be seen by frequent contrasts with differences in the fields of training and validation. (Basukala et al 2017) Based on random forest (RF), multiple Land sat 8 images, support vector machine (SVM), robust nonparametric machine learning algorithms, and a common maximum likelihood parametric algorithm (MLC).…”

Section: Accuracy In % = 75mentioning

confidence: 99%

Hyper Spectral Remote Sensing for Damage Detection and Classification Models in Agriculture - A Review

Dhande¹,

Malik²

2021

ITII

View full text Add to dashboard Cite

Agriculture is the pillar of the financial system of any country and it's far enormously crucial for financial improvement and social stability. Remote sensing has numerous benefits within the discipline of agronomical studies purpose. Remote sensing has the capacity to face this big obstacle in the evolutionary changes in agricultural practices, by delivering recurrent information on crop condition at multiple scales and for different actors during the season. Because of the rapid changes in farming started from mid to past 1980, remote sensing has become an indispensable a part of precision agriculture. Agricultural systems need to be monitored in regular interval. Remote sensing act as vital method in well timed tracking and provides a correct snapshot of the agricultural with good accuracy. The tracking of agricultural strongly follows seasonal patterns on the subject of the organic life of crops. Because of this background we studied various remote sensing methods like crop classification, crop tracking and yield assessment. We discovered that almost all of the remote sensing methods have been exploratory investigations, examined on a nearby scale with direct dependence on ground data. In tandem with other specialized methods such as geographical information systems and global positioning systems, remote sensing plays an important role in the measurement and monitoring of agricultural activities. We begin this review by presenting an outline of the existing methods of remote sensing relevant to the agricultural sector. This paper analyses remote sensing methods progress, problems and future in agricultural context. This article will offer an outline of commercially remote sensors having high resolution which have been used or have ability for accuracy in collecting agriculture information. Further, overview illustrates current research trends in remote sensing methods that allow reinforcing applicative skills with precise necessities for specific forms of stakeholders. We offer a synthesis of the rising possibilities in remote sensing techniques that have to reinforce the role in imparting long time and efficient services for agricultural purpose. SAR remote sensing techniques have extremely good capability and could play an important role in diverse fields of agricultural remote sensing. Finally, the paper outlines challenges, limitation and future trends of remote sensing methods in crop evaluation. This paper concludes that current utilization of remote sensing methods in crop evaluation still remains in an experimental stage.

show abstract

“…Although support vector machine (SVM) has been used successfully for LULC classification in various regions worldwide (e.g., [18][19][20]), it has been applied in study regions in Burkina Faso for mapping soil properties [21] or urban development patterns [22], and has yet to be investigated for monitoring irrigated areas. The aims of this study were to (1) assess the changes in irrigated agricultural areas around the Mogtedo water reservoir in Burkina Faso between 1987 and 2015 using Landsat imagery and SVM classification, and (2) determine whether the irrigable potential of this reservoir could sustainably meet the agricultural water needs under a more variable and changing climate.…”

Section: Introductionmentioning

confidence: 99%

Using Multi-Temporal Landsat Images and Support Vector Machine to Assess the Changes in Agricultural Irrigated Areas in the Mogtedo Region, Burkina Faso

et al. 2019

View full text Add to dashboard Cite

Over the last few decades, small-scale irrigation has been implemented in Burkina Faso as a strategy to mitigate the impacts of adverse climate conditions. However, the development of irrigated perimeters around small and medium water reservoirs has put the water resources under significant pressure, given the uncontrolled exploitation and lack of efficacious management plan. Insights into changes in irrigated areas around these reservoirs are therefore crucial for their sustainable management while meeting the different agricultural water needs. They will help to center policy priorities in terms of major impacts on the reservoirs; and thereby elaborate relevant mitigation and/or adaptation strategies. The main objectives of this study were to (1) quantify the changes in irrigated land areas surrounding the Mogtedo water reservoir between 1987 and 2015; and (2) determine whether the irrigable potential of this reservoir could sustainably meet the agricultural water needs under a more variable and changing climate. A low-cost remote sensing method based on Landsat imagery (Thematic Mapper, Enhanced Thematic Mapper Plus, and Operational Land Imager) and using Support Vector Machine (SVM) classification was developed to detect the changes in proportion of land use/land cover (LULC) in the Mogtedo region. A forward and backward change detection analysis requiring agronomic expertise was also applied to correct the pixels temporal trajectories. In addition, an intensity analysis was performed to assess land changes at time intervals, category, and transition levels. Five main LULC classes were identified: bare and hydromorphic soils, irrigated and rainfed agricultural areas, and water bodies. Overall, the classification of LULC was satisfactory with the overall accuracy and kappa coefficients ranging from 94.22 to 95.60% and 0.92 to 0.94, respectively. Results showed that LULC transformations were faster between 2000 and 2015, compared to the 1987–2000 period. The majority of categories (LULC classes) were active in terms of intensity of change (gain or loss) during the 1987–2000 and 2000–2015 periods, except hydromorphic soils. During these periods, the transition from rainfed agricultural areas to irrigated agricultural areas were targeted and stationary. Our findings revealed a 54% increase in irrigated areas between 1987 and 2015. The reservoir water volume decreased markedly from 9,077,000 m3 to 7,100,000 m3 during the same period. Such a decrease threatens the satisfaction of agricultural water requirements, since the reservoir is the unique source of irrigation water in the region. It could potentially lead to conflicts between users if adequate strategies for the sustainable management of the Mogtedo reservoir are not implemented. The methodology used in this study also addressed the challenge of building up historical spatial information database in data-scarce environments, and could be replicated readily in regions or countries like Burkina Faso.

show abstract

Towards improved land use mapping of irrigated croplands: performance assessment of different image classification algorithms and approaches

Abstract: View related articlesView Crossmark data Citing articles: 1 View citing articles Towards improved land use mapping of irrigated croplands: performance assessment of different image classification algorithms and approaches

Cited by 45 publications

References 35 publications

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

Hyper Spectral Remote Sensing for Damage Detection and Classification Models in Agriculture - A Review

Using Multi-Temporal Landsat Images and Support Vector Machine to Assess the Changes in Agricultural Irrigated Areas in the Mogtedo Region, Burkina Faso

Contact Info

Product

Resources

About