Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery

By Congcong Li, Jie Wang, Lei Wang, Luanyun Hu, Peng Gong. Remote Sensing, 2014

389 citations recorded by the source corpus.

Although a large number of new image classification algorithms have been developed, they are rarely tested with the same classification task. In this research, with the same Landsat Thematic Mapper (TM) data set and the same classification scheme over Guangzhou City, China, we tested two unsupervised and 13 supervised classification algorithms, including a number of machine learning algorithms that became popular in remote sensing during the past 20 years. Our analysis focused primarily on the spectral information provided by the TM data. We assessed all algorithms in a per-pixel classification decision experiment and all supervised algorithms in a segment-based experiment. We found that when sufficiently representative training samples were used, most algorithms performed reasonably well. Lack of training samples led to greater classification accuracy discrepancies than classification algorithms themselves. Some algorithms were more tolerable to insufficient (less representative) training samples than others. Many algorithms improved the overall accuracy marginally with per-segment decision making.

Subfields: Atmospheric Science, Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/c6e6463fed4cf0f1b436755a2207249a54b67988

How researchers see this paper

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

In Atmospheric Science

In Ecology

In Media Technology

Citations in the corpus

Sentinel-2 Data for Land Cover/Use Mapping: A Review (2020, methodology): “The comparison of accuracies for different classifiers needs to be interpreted with caution because the performance of these classifies depends on several factors such as number of training samples [113,114], number of land cover classes [112], the type of terrain [115] and pre-processing techniques”
Implementation of machine-learning classification in remote sensing: an applied review (2018, background): “However, one broad conclusion can be reached: large and accurate training data sets are generally preferable, no matter the algorithm used (Lu and Weng 2007; Li et al. 2014).”
Developments in Landsat Land Cover Classification Methods: A Review (2017, methodology): “Although OBIA has been commonly applied on fine spatial resolution images, most studies have indicated its superior performance on Landsat images because it combines different types of information in the classification procedure [89,128].”
Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery (2017, methodology): “Following the study of Li et al. [28] and pretested to our dataset, in this study, to ﬁnd the optimal parameters for SVM, ten values of C (2 − 2 , 2 − 1 , 2 0 , 2 1 , 2 2 , 2 3 , 2 4 , 2 5 , 2 6 , 2 7 ), and ten values of γ 3 , 2 4 ) were tested.”
Stable classification with limited sample: transferring a 30-m resolution sample set collected in 2015 to mapping 10-m resolution global land cover in 2017. (2019, background): “can be found in urban classification [8], crop field classification [9], and land cover mapping at the continental scale [10] but have never been tested at the global scale.”