DETECTING THE NATURE OF CHANGE IN AN URBAN ENVIRONMENT: A COMPARISON OF MACHINE LEARNING ALGORITHMS

The performance of different machine learning algorithms for detecting the nature of change was compared. To choose the best algorithms for a specific task, users are advised to consider not only classification accuracy, but also comprehensibility, compactness, and robustness in training and classification. The purpose of this paper is to provide information to users so that they can choose the best algorithms for specific tasks. To alleviate the problem of obtaining enough training data, simulated training data were generated from single-date images. A one-pass classification with four machine learning algorithms, namely, Multi-Layer Perceptrons (MLP), Learning Vector Quantization (LVQ), Decision Tree Classifiers (DTC), and the Maximum-Likelihood Classifier (MLC), were tested. Recognition rates, ease of use, and degree of automation of the four algorithms were assessed. The results showed that the incorporation of cross-combined simulated training data enhanced the detection of nature of change. Compared to conventional post-classification comparison methods, LVQ and DTC did better in terms of overall accuracy. In terms of average accuracy of the change classes, LVQ was the best performer. DTC was the easiest to use and the most robust in training. MLP procedures were the most difficult to replicate.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00991112
• Corporate Authors:

  American Society for Photogrammetry and Remote Sensing

  5410 Grosvenor Lane, Suite 210
  Bethesda, MD  United States  20814-2160
• Authors:
  • Chan, JC-W
  • Chan, K-P
  • Yeh, AG-O
• Publication Date: 2001-2

Language

• English

Media Info

• Features: Figures; Photos; References; Tables;
• Pagination: p. 213-225
• Serial:
  • Photogrammetric Engineering and Remote Sensing
  • Volume: 67
  • Issue Number: 2
  • Publisher: American Society for Photogrammetry and Remote Sensing
  • ISSN: 0099-1112

Subject/Index Terms

• TRT Terms: Accuracy; Algorithms; Alternatives analysis; Decision trees; Detection and identification; Evaluation and assessment; Land use; Machine learning; Maximum likelihood method; Urban areas
• Uncontrolled Terms: Learning Vector Quantization; Maximum-Likelihood Classifier; Multi-Layer Perceptrons; Nature of change
• Subject Areas: Design; Highways; Planning and Forecasting; I21: Planning of Transport Infrastructure;

Filing Info

• Accession Number: 00807447
• Record Type: Publication
• Files: TRIS
• Created Date: Mar 6 2001 12:00AM