Pixel-Based vs Patch-Based Classifiers for the LULC Classification
Pixel-Based vs Patch-Based Classifiers for the LULC Classification |
||
 |
 |
|
| © 2024 by IJETT Journal | ||
| Volume-72 Issue-6 |
||
| Year of Publication : 2024 | ||
| Author : Keerti Kulkarni, Priyadarshini K Desai |
||
| DOI : 10.14445/22315381/IJETT-V72I6P107 | ||
How to Cite?
Keerti Kulkarni, Priyadarshini K Desai, "Pixel-Based vs Patch-Based Classifiers for the LULC Classification," International Journal of Engineering Trends and Technology, vol. 72, no. 6, pp. 64-73, 2024. Crossref, https://doi.org/10.14445/22315381/IJETT-V72I6P107
Abstract
A vast majority of the policy decisions by the civic authorities depend on the geography of the given area. The land use and the land cover of the area also influence the infrastructure facilities availed by the civic bodies. The problem with LULC mapping of the urban areas is the class imbalance. Very few existing algorithms take into consideration this class imbalance. The novelty of this work is the handling of this class imbalance at two levels: the data level and the classifier level. At the data level, uniformly sampled training samples and area-proportional training samples are considered and compared. At the classifier level, pixel-based and patch-based classifiers are considered and compared. A pixel-based Parametric Classifier (Maximum Likelihood Classifier), trained on uniformly sampled training samples, gives an overall accuracy of 73.21% and an overall accuracy of 75.81% when trained on area-proportional training samples. Pixel-based Non-parametric Classifier (multiclass Support Vector Machines), trained on area-proportional training samples, gives an overall accuracy of 83%. The study area is the Bangalore Urban District, and the remotely sensed images are from LANDSAT-8. Patch-based convolutional neural networks give a superior accuracy of 91.2%. Hence, for an imbalanced class dataset, a classifier-level approach (convolutional neural networks) works better, as they look at patches of images rather than individual pixels.
Keywords
Imbalanced dataset, Convolutional Neural Networks, Land Use Land Cover (LULC), Maximum Likelihood Classifier (MLC), Support Vector Machines (SVM).
References
[1] Saeid Amini et al., “Urban Land Use and Land Cover Change Analysis Using Random Forest Classification of Landsat Time Series,” Remote Sensing, vol. 14, no. 11, pp. 1-23, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Akanksha Balha et al., “A Comparative Analysis of Different Pixel and Object-Based Classification Algorithms Using Multi-Source High Spatial Resolution Satellite Data for Lulc Mapping,” Earth Science Informatics, vol. 14, pp. 2231-2247, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Paula D. Blanco et al., “A Land Cover Map of Latin America and the Caribbean in the Framework of the Serena Project,” Remote Sensing of Environment, vol. 132, pp. 13-31, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Rok Blagus, and Lara Lusa, “Class Prediction for High-Dimensional Class-Imbalanced Data,” BMC Bioinformatics, vol. 11, pp. 1-17, 2010.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Christina Bogner et al., “Classification of Rare Land Cover Types: Distinguishing Annual and Perennial Crops In an Agricultural Catchment in South Korea,” PLoS One, vol. 13, no. 1, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[6] N.V. Chawla et al., “SMOTE: Synthetic Minority Over-sampling Technique,” Journal of Artificial Intelligence Research, vol. 16, pp. 321-357, 2002.
[CrossRef] [Google Scholar] [Publisher Link]
[7] René Roland Colditz, “An Evaluation Of Different Training Sample Allocation Schemes for Discrete and Continuous Land Cover Classification using Decision Tree-Based Algorithms,” Remote Sensing, vol. 7, no. 8, pp. 9655-9681, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Georgios Douzas, and Fernando Bacao, “Geometric SMOTE a Geometrically Enhanced Drop-in Replacement for SMOTE,” Information Sciences, vol. 501, pp. 118-135, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Martina Deur, Mateo Gašparović, and Ivan Balenović, “Tree Species Classification in Mixed Deciduous Forests Using Very High Spatial Resolution Satellite Imagery and Machine Learning Methods,” Remote Sensing, vol. l2, no. 23, pp. 1-18, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Wei Feng, Wenjiang Huang, and Wenxing Bao, “Imbalanced Hyperspectral Image Classification with an Adaptive Ensemble Method Based on SMOTE and Rotation Forest with Differentiated Sampling Rates,” IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 12, pp. 1879-1883, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Alberto Fernández et al., “Analysing the Classification of Imbalanced Data-Sets with Multiple Classes: Binarization Techniques and AdHoc Approaches,” Knowledge-Based Systems, vol. 42, pp. 97-110, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Joao Fonseca, Georgios Douzas, and Fernando Bacao, “Improving Imbalanced Land Cover Classification with K-Means SMOTE: Detecting and Oversampling Distinctive Minority Spectral Signatures,” Information, vol. 12, no. 7, pp. 1-20, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Kpade O.L. Hounkpatin et al., “Predicting Reference Soil Groups Using Legacy Data: A Data Pruning and Random Forest Approach for Tropical Environment (Dano Catchment, Burkina Faso), Scientific Reports, vol. 8, pp. 1-16, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Huiran Jin, Stephen V. Stehman, and Giorgos Mountrakis, “Assessing the Impact of Training Sample Extraction on Accuracy of an Urban Classification: A Case Study in Denver, Colorado,” International Journal of Remote Sensing, vol. 35, no. 6, pp. 2067-2081, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Taeho Jo, and Nathalie Japkowicz, “Class Imbalances versus Small Disjuncts,” ACM Sigkdd Explorations Newsletter, vol. 6, no. 1, pp. 40-49, 2004.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Shahab Eddin Jozdani, Brian Alan Johnson, and Dongmei Chen, “Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification,” Remote Sensing, vol. 11, no. 14, pp. 1-24, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Harsurinder Kaur, Husanbir Singh Pannu, and Avleen Kaur Malhi, “A Systematic Review on Imbalanced Data Challenges in Machine Learning: Applications and Solutions,” ACM Computer Surveys, vol. 52, no. 4, pp. 1-36, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Land Use / Land Cover Analysis-Third Cycle Technical, Natural Resource Census, pp. 1-12, 2019. [Online]. Available: https://bhuvanapp1.nrsc.gov.in/2dresources/thematic/LULC503/lulc.pdf
[19] Taehyung Lee, Ki Bum Lee, and Chang Ouk Kim, “Performance of Machine Learning Algorithms for Class-Imbalanced Process Fault Detection Problems,” IEEE Transactions on Semiconductor Manufacturing, vol. 29, no. 4, pp. 436-445, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Congcong Li et al., “Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery,” Remote Sensing, vol. 6, no. 2, pp. 964-983, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Julián Luengo et al., Imbalanced Data Preprocessing for Big Data, Big Data Preprocessing, Springer, Cham, pp. 147-160, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Aaron E. Maxwell, Timothy A. Warner, and Fang Fang, “Implementation of Machine-Learning Classification in Remote Sensing: An Applied Review,” International Journal of Remote Sensing, vol. 39, no. 9, pp. 2784-2817, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Varun Narayan Mishra et al., “Performance Evaluation of Textural Features in Improving Land Use/Land Cover Classification Accuracy of Heterogeneous Landscape Using Multi-Sensor Remote Sensing Data,” Earth Science Informatics, vol. 12, pp. 71-86, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Ranjit Panigrahi et al., “A Consolidated Decision Tree-Based Intrusion Detection System for Binary and Multiclass Imbalanced Datasets,” Mathematics, vol. 9, no. 7, pp. 1-35, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Fernando Pech-May et al., “Mapping of Land Cover with Optical Images, Supervised Algorithms, and Google Earth Engine,” Sensors, vol. 22, no. 13, pp. 1-19, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Kwanele Phinzi et al., “Understanding the Role of Training Sample Size in the Uncertainty of High-Resolution Lulc Mapping Using Random Forest,” Earth Science Informatics, vol. 16, pp. 3667-3677, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Biswajeet Pradhan, “A Comparative Study On The Predictive Ability of the Decision Tree, Support Vector Machine and Neuro-Fuzzy Models in Landslide Susceptibility Mapping Using GIS,” Computers & Geosciences, vol. 51, pp. 350-365, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Julien Radoux et al., “Automated Training Sample Extraction for Global Land Cover Mapping,” Remote Sensing, vol. 6, no. 5, pp. 39653987, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Amin Sharififar, Fereydoon Sarmadian, and Budiman Minasny, “Mapping Imbalanced Soil Classes Using Markov Chain Random Fields Models Treated With Data Resampling Technique,” Computers and Electronics in Agriculture, vol. 159, pp. 110-118, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Yuan-Hai Shao et al., “An Efficient Weighted Lagrangian Twin Support Vector Machine for Imbalanced Data Classification,” Pattern Recognition, vol. 47, no. 9, pp. 3158-3167, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Asamaporn Sitthi et al., “Exploring Land Use and Land Cover of Geotagged Social-Sensing Images Using Naive Bayes Classifier,” Sustainability, vol. 8, no. 9, pp. 1-22, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[32] M. Zhu et al., “Land Cover Classification Using High Resolution Satellite Image Based on Deep Learning,” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 42, pp. 685-690, 2020.
[CrossRef] [Google Scholar] [Publisher Link]