A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future research

By Reza Khatami, Giorgos Mountrakis, Stephen V. Stehman. 2016

543 citations recorded by the source corpus.

Subfields: Ecology, Global and Planetary Change, Media Technology

View Paper: https://www.semanticscholar.org/paper/ccdf098fd7d0ae80895b700da36b112832015d8a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology

1. Joanne C. White
2. M. Wulder
3. Abdulhakim M. Abdi
4. David P. Roy
5. Charlotte Pelletier
6. Silvia Valero
7. Ieda Del'Arco Sanches
8. Jordi Inglada

In Global and Planetary Change

1. Abdulhakim M. Abdi
2. Joanne C. White
3. M. Wulder
4. Ieda Del'Arco Sanches
5. Geoffrey I. Webb
6. Txomin Hermosilla
7. Nicholas C. Coops
8. Martin Kappas

In Media Technology

1. Abdulhakim M. Abdi
2. M. Wulder
3. Ieda Del'Arco Sanches
4. Jordi Inglada
5. Geoffrey I. Webb
6. Txomin Hermosilla
7. Nicholas C. Coops
8. Martin Kappas

Citations in the corpus

• Assessing the robustness of Random Forests to map land cover with high resolution satellite image time series over large areas (2016, background): “, 1996; Friedl and Brodley, 1997) — outperform unsupervised methods (Szuster et al., 2011; Khatami et al., 2016).”
• Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery (2017, methodology): “Remote sensing satellite images are considered as one of the most important data sources for land use/cover mapping [10] due to their extensive geographical coverage at an efficient cost while providing irreplaceable information on the earth’s surface [11].”
• Land cover 2.0 (2018, background): “Additional ancillary data options are evident in the land cover literature (e.g., as reviewed in Franklin and Wulder 2002; Khatami, Mountrakis, and Stehman 2016) and are often straightforward to incorporate given the current processing and classi fi cation environment.”
• Recent Applications of Landsat 8/OLI and Sentinel-2/MSI for Land Use and Land Cover Mapping: A Systematic Review (2020, background): “[11] provided guidance on the performance of different supervised pixel-based image classification processes, involving algorithms and input data.”
• Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series (2018, methodology): “The state-of-the-art classification algorithms used to produce maps are currently Support Vector Machines (SVMs) and Random Forests (RFs) [9].”
• Algorithms for semantic segmentation of multispectral remote sensing imagery using deep learning (2017, background): “The authors in [22] determined that spatial (texture) information influenced classification performance the most, which is why current state-of-theart methods extract spatial-spectral features from the image data.”
• Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data (2019, methodology): “Indeed, Khatami, Mountrakis, and Stehman (2016) found that SVM was the most e ffi cient algorithm for most applications and outperformed several classi fi er families, including RF, neural networks, and decision trees in direct comparison.”