Game Theory Based Pixel Approximation for Remote Sensing Imagery

Abstract Classification of remote sensing images faces several challenges due to mixed pixels. Such pixels that are wrongly classified are called mixed pixels. There is uncertainty about the class label of mixed pixels as they represent the average energy emitted from different objects present within their spatial extent. Rough set theory can address the vagueness in data. Therefore, in this work rough set concept is used to identify the mixed pixels. In a multi-player environment, the Game theory is a science of making rational decisions. We propose a game theory-based approach to approximate the mixed pixels to lower approximations of a class. For pixel approximation, we have applied the spatial information of neighbouring pixels. Experiment for the implementation of the proposed approach is carried on six Landsat 5 Thematic Mapper images with a different region of interest. These datasets vary in size and have different dominant classes. The impact of the proposed method on the quality of clusters is studied with the help of cluster quality parameters. The results of the percentage of approximation demonstrate the significant number of mixed pixels approximated to one of the class.

[1]  Shashi Shekhar,et al.  Identifying patterns in spatial information: A survey of methods , 2011, WIREs Data Mining Knowl. Discov..

[2]  Pawan Lingras,et al.  Applying Rough Set Concepts to Clustering , 2012 .

[3]  Valerio Cozzani,et al.  Applying game theory for securing oil and gas pipelines against terrorism , 2019, Reliab. Eng. Syst. Saf..

[4]  S. Swenson,et al.  Remote sensing of groundwater storage changes in Illinois using the Gravity Recovery and Climate Experiment (GRACE) , 2006 .

[5]  Fangju Wang,et al.  Fuzzy supervised classification of remote sensing images , 1990 .

[6]  Pramod K. Varshney,et al.  Decision tree regression for soft classification of remote sensing data , 2005 .

[7]  Andrzej Skowron,et al.  Data science, big data and granular mining , 2015, Pattern Recognit. Lett..

[9]  T. Caliński,et al.  A dendrite method for cluster analysis , 1974 .

[10]  L. P. C. Verbeke,et al.  Using genetic algorithms in sub-pixel mapping , 2003 .

[11]  Liang-pei Zhang,et al.  Spectral-spatial classification of hyperspectral imagery with cooperative game , 2018 .

[12]  Qinghua Ye,et al.  Handling uncertainties in image mining for remote sensing studies , 2009 .

[13]  Roger B. Myerson,et al.  Game theory - Analysis of Conflict , 1991 .

[14]  Lorenzo Bruzzone,et al.  Hyperspectral Band Selection Based on Rough Set , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Lori M. Bruce.  Game theory applied to big data analytics in geosciences and remote sensing , 2013, 2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS.

[16]  Pi-Fuei Hsieh,et al.  Effect of spatial resolution on classification errors of pure and mixed pixels in remote sensing , 2001, IEEE Trans. Geosci. Remote. Sens..

[17]  P. Rousseeuw.  Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[18]  João Paulo Ramos Teixeira,et al.  Remote sensing of drought: Progress, challenges and opportunities , 2015 .

[19]  Sonajharia Minz,et al.  Rough-Set Based Hotspot Detection in Spatial Data , 2019, ICACDS.

[20]  Gerard B. M. Heuvelink,et al.  Hybrid Constraints of Pure and Mixed Pixels for Soft-Then-Hard Super-Resolution Mapping With Multiple Shifted Images , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[21]  Andrzej Skowron,et al.  Rough sets: past, present, and future , 2018, Natural Computing.

[22]  Sankar K. Pal,et al.  Multispectral image segmentation using the rough-set-initialized EM algorithm , 2002, IEEE Trans. Geosci. Remote. Sens..

[23]  G. Leibniz.  Discourse on Metaphysics , 1902 .

[24]  Heesung Kwon,et al.  Coalition Game Theory-Based Feature Subspace Selection for Hyperspectral Classification , 2016, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[25]  M. Mohan,et al.  Impact of urbanization and land-use/land-cover change on diurnal temperature range: a case study of tropical urban airshed of India using remote sensing data. , 2015, The Science of the total environment.

[26]  Donald W. Bouldin,et al.  A Cluster Separation Measure , 1979, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[27]  Giles M. Foody,et al.  Incorporating mixed pixels in the training, allocation and testing stages of supervised classifications , 1996, Pattern Recognit. Lett..

[28]  Dipti Patra,et al.  An ensemble approach for the detection and classification of mixed pixels of remotely sensed images , 2018, 2018 4th International Conference on Recent Advances in Information Technology (RAIT).