Ranga Raju Vatsavai
ABSTRACT
In this paper, we present a computationally efficient algorithm based on multiple instance learning for mapping informal settlements (slums) using very high-resolution remote sensing imagery. From remote sensing perspective, informal settlements share unique spatial characteristics that distinguish them from other urban structures like industrial, commercial, and formal residential settlements. However, regular pattern recognition and machine learning methods, which are predominantly single-instance or per-pixel classifiers, often fail to accurately map the informal settlements as they do not capture the complex spatial patterns. To overcome these limitations we employed a multiple instance based machine learning approach, where groups of contiguous pixels (image patches) are modeled as generated by a Gaussian distribution. We have conducted several experiments on very high-resolution satellite imagery, representing four unique geographic regions across the world. Our method showed consistent improvement in accurately identifying informal settlements.
- Quantum GIS: open source geographic information system (gis). http://www.qgis.org/.Google Scholar
- A. Benediktsson, P. Swain, and O. Ersoy. Neural network approaches versus statistical methods in classificaion of multisource remote sensing data. IEEE Transactions on Geoscience and Remote Sensing, 28(4):550, 1990.Google ScholarCross Ref
- T. Blaschke, S. Lang, and G. Hay. Object-Based Image Analysis. Springer, 2008. Google ScholarDigital Library
- J. Bolton and P. Gader. Application of multiple-instance learning for hyperspectral image analysis. Geoscience and Remote Sensing Letters, IEEE, 8(5):889--893, sept. 2011.Google ScholarCross Ref
- L. Bruzzone, C. Consese, F. Masellit, and F. Roli. Multisource classification of complex rural areas by statistical and neural-network approaches. Photogrammetric Engineering & Remote Sensing, 63(5):523--533, May 1997.Google Scholar
- V. Chandola and R. R. Vatsavai. Scalable time series change detection for biomass monitoring using gaussian process. In CIDU, pages 69--82, 2010.Google Scholar
- V. Chandola and R. R. Vatsavai. A scalable gaussian process analysis algorithm for biomass monitoring. Statistical Analysis and Data Mining, 4(4):430--445, 2011. Google ScholarDigital Library
- T. G. Dietterich, R. H. Lathrop, T. Lozano-Perez, and A. Pharmaceutical. Solving the multiple-instance problem with axis-parallel rectangles. Artificial Intelligence, 89:31--71, 1997. Google ScholarDigital Library
- S. Dreiseitl and L. Ohno-Machado. Logistic regression and artificial neural network classification models: a methodology review. Journal of Biomedical Informatics, 35(5--6):352 -- 359, 2002. Google ScholarDigital Library
- J. Graesser, A. Cheriyadat, R. Vatsavai, V. Chandola, J. Long, and E. Bright. Image based characterization of formal and informal neighborhoods in an urban landscape. Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of, 5(4):1164 --1176, aug. 2012.Google Scholar
- J. Ham, Y. Chen, M. M. Crawford, and J. Ghosh. Investigation of the random forest framework for classification of hyperspectral data. Geoscience and Remote Sensing, IEEE Transactions on, 43(3):492 -- 501, Mar. 2005.Google Scholar
- I. Kanellopoulos and G. G. Wilkinson. Strategies and best practice for neural network image classification. International Journal of Remote Sensing, 18(4):711--725, 1997.Google ScholarCross Ref
- B. Krishnapuram, L. Carin, M. A. Figueiredo, and A. J. Hartemink. Sparse multinomial logistic regression: fast algorithms and generalization bounds. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 27(6):957 --968, June 2005. Google ScholarDigital Library
- O. Maron and T. Lozano-Pérez. A framework for multiple-instance learning. In ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, pages 570--576. MIT Press, 1998. Google ScholarDigital Library
- S. Nussbaum and G. Menz. Object-Based Image Analysis and Treaty Verification. Springer, 2008.Google Scholar
- J. Paola and R. Schowengerdt. A detailed comparison of backpropagation neural network and maximum likelihood classifiers for urban land use classification. IEEE Transactions on Geoscience and Remote Sensing, 33(4):981--996, 1995.Google ScholarCross Ref
- J. Paola and R. Schowengerdt. The effect of neural-network structure on a multispectral land-use/land-cover classification. Photogrammetric Engineering & Remote Sensing, 63(5):535--544, 1997.Google Scholar
- S. Shekhar, P. Schrater, R. Vatsavai, W. Wu, and S. Chawla. Spatial contextual classification and prediction models for mining geospatial data. IEEE Transaction on Multimedia, 4(2):174--188, 2002. Google ScholarDigital Library
- A. Skidmore, B. Turner, W. Brinkhof, and E. Knowles. Performance of a neural network: Mapping forest using gis and remorely sensed data. Photogrammetric Engineering & Remote Sensing, 63(5):501--514, May 1997.Google Scholar
- P. Torrione, C. Ratto, and L. Collins. Multiple instance and context dependent learning in hyperspectral data. In Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, 2009. WHISPERS '09. First Workshop on, pages 1 --4, aug. 2009.Google ScholarCross Ref
- R. R. Vatsavai. High-resolution urban image classification using extended features. In ICDM Workshops, pages 869--876, 2011. Google ScholarDigital Library
- R. R. Vatsavai. A data mining framework for monitoring nuclear facilities. In ICDM Workshops (Industry/Government Track), page 917, 2012. Google ScholarDigital Library
- R. R. Vatsavai, B. Bhaduri, and J. Graesser. Complex settlement pattern extraction with multi-instance learning. In Joint Urban Remote Sensing Event, pages 1 --4, apr. 2013.Google ScholarCross Ref
- R. R. Vatsavai, B. L. Bhaduri, A. Cheriyadat, L. F. Arrowood, E. A. Bright, S. S. Gleason, C. Diegert, A. K. Katsaggelos, T. Pappas, R. Porter, J. Bollinger, B. Chen, and R. Hohimer. Geospatial image mining for nuclear proliferation detection: Challenges and new opportunities. In IGARSS, pages 48--51, 2010.Google ScholarCross Ref
- R. R. Vatsavai, E. A. Bright, V. Chandola, B. L. Bhaduri, A. Cheriyadat, and J. Graesser. Machine learning approaches for high-resolution urban land cover classification: a comparative study. In COM.Geo, page 11, 2011. Google ScholarDigital Library
- J. Wang. Solving the multiple-instance problem: A lazy learning approach. In In Proc. 17th International Conf. on Machine Learning, pages 1119--1125. Morgan Kaufmann, 2000. Google ScholarDigital Library
Index Terms
- Gaussian multiple instance learning approach for mapping the slums of the world using very high resolution imagery
Recommendations
Digital mapping of coffee ripeness using UAV-based multispectral imagery
Highlights- UAV imagery provides a feasible method for monitoring the coffee fruit ripeness.
AbstractTimely and accurate monitoring of coffee ripeness is essential for harvest planning, especially in mountainous areas where the harvest is performed manually due to the limited use of agricultural mechanization. The increasing temporal ...
Quantitative mapping of pasture biomass using satellite imagery
A knowledge of the amount of pasture biomass available in farm paddocks is crucial for improving utilization and productivity in the Australian grazing industry. A method to quantitatively map the biomass of annual pastures under grazing has been ...
Using QuickBird imagery and a production efficiency model to improve crop yield estimation in the semi-arid hilly Loess Plateau, China
Crop yield is a key element in rural development and an indicator of national food security. A method that could estimate crop yield over large hilly areas would be highly desirable. Methods including high spatial resolution satellite imagery have the ...
Comments