Advertisement

A Hybrid CNN + Random Forest Approach to Delineate Debris Covered Glaciers Using Deep Features

  • Rahul Nijhawan
  • Josodhir Das
  • Raman Balasubramanian
Research Article
  • 139 Downloads

Abstract

The main aim of this study is to propose a novel hybrid deep learning framework approach for accurate mapping of debris covered glaciers. The framework comprises of integration of several CNNs architecture, in which different combinations of Landsat 8 multispectral bands (including thermal band), topographic and texture parameters are passed as input for feature extraction. The output of an ensemble of these CNNs is hybrid with random forest model for classification. The major pillars of the framework include: (1) technique for implementing topographic and atmospheric corrections (preprocessing), (2) the proposed hybrid of ensemble of CNNs and random forest classifier, and (3) procedures to determine whether a pixel predicted as snow is a cloud edge/shadow (post-processing). The proposed approach was implemented on the multispectral Landsat 8 OLI (operational land imager)/TIRS (thermal infrared sensor) data and Shuttle Radar Topography Mission Digital Elevation Model for the part of the region situated in Alaknanda basin, Uttarakhand, Himalaya. The proposed framework was observed to outperform (accuracy 96.79%) the current state-of-art machine learning algorithms such as artificial neural network, support vector machine, and random forest. Accuracy assessment was performed by means of several statistics measures (precision, accuracy, recall, and specificity).

Keywords

Classification CNN Texture Random forest Debris Glaciers 

References

  1. Archer, K. J., & Kimes, R. V. (2008). Empirical characterization of random forest variable importance measures. Computational Statistics and Data Analysis, 52(4), 2249–2260.CrossRefGoogle Scholar
  2. Bishop, M. P., Bonk, R., Kamp, U., Jr., & Shroder, J. F., Jr. (2001). Terrain analysis and data modeling for alpine glacier mapping. Polar Geography, 25(3), 182–201.CrossRefGoogle Scholar
  3. Bishop, M. P., Shroder, J. F., Jr., & Hickman, B. L. (1999). SPOT panchromatic imagery and neural networks for information extraction in a complex mountain environment. Geocarto International, 14(2), 19–28.CrossRefGoogle Scholar
  4. Bolch, T., Buchroithner, M. F., Kunert, A., & Kamp, U. (2007). Automated delineation of debris-covered glaciers based on ASTER data. In Proceedings of the 27th EARSeL symposium geoinformation in Europe (pp. 4–6).Google Scholar
  5. Bolch, T., & Kamp, U. (2006). Glacier mapping in high mountains using DEMs, landsat and ASTER data. Grazer Schriften der Geographie und Raumforschung, 41, 37–48.Google Scholar
  6. Brahmbhatt, R. M., Bahuguna, I., Rathore, B. P., Kulkarni, A. V., Shah, R. D., & Nainwal, H. C. (2012). Variation of snowline and mass balance of glaciers of Warwan and Bhut basins of western Himalaya using remote sensing technique. Journal of the Indian Society of Remote Sensing, 40(4), 629–637.CrossRefGoogle Scholar
  7. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.CrossRefGoogle Scholar
  8. Buchroithner, M. F., & Bolch, T. (2007). An automated method to delineate the ice extension of the debris-covered glaciers at Mt. Everest based on ASTER imagery. In Proceedings of the 9th international symposium on high mountain remote sensing cartography, 2006 held in Graz, Austria.Google Scholar
  9. Castelluccio, M., Poggi, G., Sansone, C., & Verdoliva, L. (2015). Land use classification in remote sensing images by convolutional neural networks. http://arxiv.org/abs/1508.00092.
  10. Change, I. P. O. C. (2007). Climate change 2007: The physical science basis. Agenda, 6(07), 333.Google Scholar
  11. Chen, Y., Zhao, X., & Jia, X. (2015). Spectral–spatial classification of hyperspectral data based on deep belief network. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(6), 2381–2392.CrossRefGoogle Scholar
  12. Foody, G. M., Campbell, N. A., Trodd, N. M., & Wood, T. F. (1992). Derivation and applications of probabilistic measures of class membership from the maximum-likelihood classification. Photogrammetric Engineering and Remote Sensing, 58(9), 1335–1341.Google Scholar
  13. Foody, G. M., & Mathur, A. (2004). A relative evaluation of multiclass image classification by support vector machines. IEEE Transactions on Geoscience and Remote Sensing, 42(6), 1335–1343.CrossRefGoogle Scholar
  14. Gao, J. (2008). Digital analysis of remotely sensed imagery. New York: McGraw-Hill Professional.Google Scholar
  15. Ghosh, S., Pandey, A. C., Nathawat, M. S., & Bahuguna, I. M. (2014). Contrasting signals of glacier changes in Zanskar valley, Jammu & Kashmir, India using remote sensing and GIS. Journal of the Indian Society of Remote Sensing, 42(4), 817–827.CrossRefGoogle Scholar
  16. Haykin, S. (1999). Neural networks: A comprehensive foundation (2nd ed.). Upper Saddle River, NJ: Prentice Hall.Google Scholar
  17. Kargel, J. S., Abrams, M. J., Bishop, M. P., Bush, A., Hamilton, G., Jiskoot, H., Kääb, A., Kieffer, H. H., Lee, E. M., Paul, F., & Rau, F. (2005). Multispectral imaging contributions to global land ice measurements from space. Remote Sensing of Environment, 99(1), 187–219.CrossRefGoogle Scholar
  18. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097–1105).Google Scholar
  19. Kulkarni, A. V., & Alex, S. (2003). Estimation of recent glacial variations in Baspa Basin using remote sensing technique. Journal of the Indian Society of Remote Sensing, 31(2), 81–90.CrossRefGoogle Scholar
  20. Li, T., Zhang, J., & Zhang, Y. (2014). Classification of hyperspectral image based on deep belief networks. In 2014 IEEE international conference on image processing (ICIP) (pp. 5132–5136) IEEE.Google Scholar
  21. Midhun, M. E., Nair, S. R., Prabhakar, V. T., & Kumar, S. S. (2014). Deep model for classification of hyperspectral image using restricted boltzmann machine. In Proceedings of the 2014 international conference on interdisciplinary advances in applied computing (p. 35) ACM.Google Scholar
  22. Nemmour, H., & Chibani, Y. (2010). Support vector machines for automatic multi-class change detection in Algerian Capital using landsat TM imagery. Journal of the Indian Society of Remote Sensing, 38(4), 585–591.CrossRefGoogle Scholar
  23. Pal, M. (2005). Random forest classifier for remote sensing classification. International Journal of Remote Sensing, 26(1), 217–222.CrossRefGoogle Scholar
  24. Paul, F., Huggel, C., & Kääb, A. (2004). Combining satellite multispectral image data and a digital elevation model for mapping debris-covered glaciers. Remote Sensing of Environment, 89(4), 510–518.CrossRefGoogle Scholar
  25. Paul, F., Kääb, A., & Haeberli, W. (2007). Recent glacier changes in the Alps observed by satellite: Consequences for future monitoring strategies. Global and Planetary Change, 56(1), 111–122.CrossRefGoogle Scholar
  26. Platt, J. C., Cristianini, N., & Shawe-Taylor, J. (2000). Large margin DAGs for multiclass classification. In nips (Vol. 12, pp. 547–553).Google Scholar
  27. Racoviteanu, A., & Williams, M. W. (2012). Decision tree and texture analysis for mapping debris-covered glaciers in the Kangchenjunga area, Eastern Himalaya. Remote Sensing, 4(10), 3078–3109.CrossRefGoogle Scholar
  28. Ren, J., Shen, Y., Ma, S., & Guo, L. (2004). Applying multi-class SVMs into scene image classification. In International conference on industrial, engineering and other applications of applied intelligent systems (pp. 924–934). Berlin: Springer.Google Scholar
  29. Richter, R. (1998). Correction of satellite imagery over mountainous terrain. Applied Optics, 37(18), 4004–4015.CrossRefGoogle Scholar
  30. Shukla, A., Arora, M. K., & Gupta, R. P. (2010). Synergistic approach for mapping debris-covered glaciers using optical-thermal remote sensing data with inputs from geomorphometric parameters. Remote Sensing of Environment, 114(7), 1378–1387.CrossRefGoogle Scholar
  31. Simonyan, K., & Zisserman, A. (2014). Two-stream convolutional networks for action recognition in videos. In Advances in neural information processing systems (pp. 568–576).Google Scholar
  32. Simpson, J. J., Jin, Z., & Stitt, J. R. (2000). Cloud shadow detection under arbitrary viewing and illumination conditions. IEEE Transactions on Geoscience and Remote Sensing, 38(2), 972–976.CrossRefGoogle Scholar
  33. Simpson, J. J., & Stitt, J. R. (1998). A procedure for the detection and removal of cloud shadow from AVHRR data over land. IEEE Transactions on Geoscience and Remote Sensing, 36(3), 880–897.CrossRefGoogle Scholar
  34. Simpson, J. J., Stitt, J. R., & Sienko, M. (1998). Improved estimates of the areal extent of snow cover from AVHRR data. Journal of Hydrology, 204(1–4), 1–23.CrossRefGoogle Scholar
  35. Singh, K. K., Kulkarni, A. V., & Mishra, V. D. (2010). Estimation of glacier depth and moraine cover study using ground penetrating radar (GPR) in the Himalayan region. Journal of the Indian Society of Remote Sensing, 38(1), 1–9.CrossRefGoogle Scholar
  36. Stokes, C. R., Popovnin, V., Aleynikov, A., Gurney, S. D., & Shahgedanova, M. (2007). Recent glacier retreat in the Caucasus Mountains, Russia, and associated increase in supraglacial debris cover and supra-/proglacial lake development. Annals of Glaciology, 46(1), 195–203.CrossRefGoogle Scholar
  37. Thakur, P. K., Aggarwal, S. P., Arun, G., Sood, S., Kumar, A. S., Mani, S., & Dobhal, D. P. (2017). Estimation of snow cover area, snow physical properties and glacier classification in parts of Western Himalayas using C-band SAR data. Journal of the Indian Society of Remote Sensing, 45(3), 525–539.CrossRefGoogle Scholar
  38. Wang, L., Zhang, J., Liu, P., Choo, K. K. R., & Huang, F. (2017). Spectral–spatial multi-feature-based deep learning for hyperspectral remote sensing image classification. Soft Computing, 21(1), 213–221.CrossRefGoogle Scholar
  39. Whalley, W. B., Martin, H. E., & Gellatly, A. F. (1986). The problem of hidden ice in glacier mapping. Annals of Glaciology, 8, 181–183.CrossRefGoogle Scholar
  40. Wold, S., Esbensen, K., & Geladi, P. (1987). Principal component analysis. Chemometrics and Intelligent Laboratory Systems, 2(1–3), 37–52.CrossRefGoogle Scholar

Copyright information

© Indian Society of Remote Sensing 2018

Authors and Affiliations

  1. 1.Department of Earthquake EngineeringIndian Institute of Technology RoorkeeRoorkeeIndia
  2. 2.Department of Computer Science and EngineeringIndian Institute of Technology RoorkeeRoorkeeIndia

Personalised recommendations