An Assessment of Artificial Neural Networks, Support Vector Machines and Decision Trees for Land Cover Classification Using Sentinel-2A Data

Hamad, Rahel

doi:10.12691/aees-8-6-18

Cited by 8 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study covered the literature review about ANN classifier from previous studies as follow: (Mishra et al, 2017;Kadavi and Lee, 2018;Dibs et al, 2020;Dixit and Agarwal, 2020;Ekumah et al, 2020;Hamad, 2020;Kaya and Görgün, 2020;MohanRajan et al, 2020;Navin and Agilandeeswari, 2020;Rojas et al, 2020;Saddique et al, 2020;Xu et al, 2020;Angessa et al, 2021;Bhattacharya et al, 2021;Dede et al, 2021;Ghayour et al, 2021;Sang et al, 2021;Xie et al, 2021;Yusof et al, 2021;Ambinakudige and Intsiful, 2022;Fantinel et al, 2022;Gogumalla et al, 2022;Rizvon and Jayakumar, 2022;Theres and Selvakumar, 2022).…”

Section: Ann Classifiermentioning

confidence: 99%

Classification of land use/land cover using artificial intelligence (ANN-RF)

Alshari

Abdulkareem

Gawali

2023

Front. Artif. Intell.

View full text Add to dashboard Cite

Because deep learning has various downsides, such as complexity, expense, and the need to wait longer for results, this creates a significant incentive and impetus to invent and adopt the notion of developing machine learning because it is simple. This study intended to increase the accuracy of machine-learning approaches for land use/land cover classification using Sentinel-2A, and Landsat-8 satellites. This study aimed to implement a proposed method, neural-based with object-based, to produce a model addressed by artificial neural networks (limited parameters) with random forest (hyperparameter) called ANN_RF. This study used multispectral satellite images (Sentinel-2A and Landsat-8) and a normalized digital elevation model as input datasets for the Sana'a city map of 2016. The results showed that the accuracy of the proposed model (ANN_RF) is better than the ANN classifier with the Sentinel-2A and Landsat-8 satellites individually, which may contribute to the development of machine learning through newer researchers and specialists; it also conventionally developed traditional artificial neural networks with seven to ten layers but with access to 1,000's and millions of simulated neurons without resorting to deep learning techniques (ANN_RF).

show abstract

Section: Ann Classifiermentioning

confidence: 99%

Classification of land use/land cover using artificial intelligence (ANN-RF)

Alshari

Abdulkareem

Gawali

2023

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Compared to SVM and Decision Trees, Artificial neural networks have given better results in many specific applications of LULC classification [99]. manual effort required in annotation and object extraction tasks.…”

Section: Image Classification Techniquesmentioning

confidence: 99%

A Systematic Review of Machine Learning Applications in Land Use Land Cover Change Detection using Remote Sensing

Sumangala

Kini²

2022

IJAEML

View full text Add to dashboard Cite

Background/Purpose: The objective of this literature review is to explore different land use and land cover methods using machine learning techniques and also their applications in change detection. Reviewing various methods adopted in this domain opens up a new path for taking up further research by extending the current approaches. Design/Methodology/Approach: The research findings presented in various scholarly articles are collected from secondary resources including scholarly journal publications. These articles are analyzed, and the interpretations are highlighted in this review paper. Findings/Result: This research provides insight into various techniques used to classify remote sensing imagery. The gaps identified during the analysis with different approaches have helped to get a clear picture when formulating research questions in the remote sensing geographic information systems domain. Research limitations/implications: This study has surveyed various applications of remote sensing in GIS. This study is limited to a review of the various machine-learning approaches used for implementing change detection. The various deep learning architectures for image classification could be further explored. Originality/Value: The articles selected for review in this study are from scholarly research journals and are cited by other authors in their publications. The papers selected for review are relevant to the research work and research proposal presented in this paper. Paper Type: Literature review paper.

show abstract

“…Shetty S (2019) [17] while analysing efficiency of classifiers for LULC using Random Forest, Support Vector Machine and Maximum likelihood classier applied on Google Earth Engine image found that random forest (RF) classifier to be the best with highest accuracy. R. Hamad (2020) [18] in his research paper, An assessment of Artificial Neural Network (ANN), Support Vector Machine (SVM) and Decision Tree (DT) for land cover classification using sentinel-2A data found that ANN is the best classifier with 90% accuracy followed by SVM with 65% and DT with 60%. Mahendra et al (2019) [19] used Landsat-3 image to find the efficient classifiers among Mahalanobis distance, Minimum distance, Maximum likelihood and SVM and fond that SVM as the best classifier with 95.35% accuracy with Kappa of 0.94.…”

Section: Related Workmentioning

confidence: 99%

Change Detection of LULC using Machine Learning

Geetha¹,

Gowda²,

Nandeesha³

et al. 2021

Atlantis Highlights in Computer Sciences

View full text Add to dashboard Cite

This paper discusses detection of change in land usage in Davangere (Karnataka State, India) between the years 2016 and 2021. After the place has been declared as one of the smart cities identified by the Govt. of India in 2014 and subsequent to the international price crash for sugar, there were noticeable changes in land utilization in terms of urbanization and shift in traditional cropping pattern. The objective of this research work is to capture this change using remote sensing, the images from MSI Sentinel-2 were collected at two points of time and processed for LULC with the help of supervised machine learning classifiers such as Minimum Distance, Mahalanobis Distance and Maximum Likelihood to ascertain the accurate one. It was found that Maximum Likelihood classifier ensures highest accuracy of 95.2%. It was also found that during the study period, there was a significant change in the land use with respect to Built-up area and Area under cultivation of Paddy.

show abstract

An Assessment of Artificial Neural Networks, Support Vector Machines and Decision Trees for Land Cover Classification Using Sentinel-2A Data

Cited by 8 publications

References 32 publications

Classification of land use/land cover using artificial intelligence (ANN-RF)

Classification of land use/land cover using artificial intelligence (ANN-RF)

A Systematic Review of Machine Learning Applications in Land Use Land Cover Change Detection using Remote Sensing

Change Detection of LULC using Machine Learning

Contact Info

Product

Resources

About