Disaster Assessment with High Resolution Remote Sensing Images Based on Hierarchical Knowledge Transfer Method

High resolution remote sensing images are more detailed observation data of small targets on the earth's surface, and have become one of important data support for disaster monitoring and assessment. At present, some achievements have been got on the application of high resolution remote sensing images in the field of disasters. However, the existing methods needed professional data and professional staff, which limits the wide application of these methods in actual business. In order to improve the automation level of disaster assessment services and the accuracy of evaluation results, this paper proposed a hierarchical knowledge model, which named "feature - rule - knowledge" model on the basis of the full analysis of disaster assessment process and earthquake disaster assessment application. An implementation framework of dynamic knowledge acquisition and hierarchical knowledge transfer was designed, and some key technologies were described in detail, including hierarchical knowledge management model, sample matching, feature selection, spatial rules mining and historical knowledge application. Finally, taking Ludian, Yunnan Province as a study area, an automatic experiment of earthquake disaster assessment based on high resolution remote sensing images was carried out. The comparison between experimental results and field survey results showed that this method could basically automate the evaluation process and effectively enhance the capability of disaster emergency response. At the same time, the stability of evaluation results was improved by reducing manual participation in assessment process.

[1]  Charles K. Huyck,et al.  Towards Rapid Citywide Damage Mapping Using Neighborhood Edge Dissimilarities in Very High-Resolution Optical Satellite Imagery—Application to the 2003 Bam, Iran, Earthquake , 2005 .

[2]  Zhang Da-peng.  Review of Road Extraction from High Resolution Remote Sensing Images , 2010 .

[3]  Lorenzo Bruzzone,et al.  Definition of Effective Training Sets for Supervised Classification of Remote Sensing Images by a Novel Cost-Sensitive Active Learning Method , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[4]  André Stumpf,et al.  Active Learning in the Spatial Domain for Remote Sensing Image Classification , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Tan Qu-lin,et al.  Urban Building Extraction from VHR Multi-spectral Images Using Object-based Classification , 2010 .

[6]  Zhiwen Zhu,et al.  Building extraction from high resolution remote sensing imagery based on spatial-spectral method , 2012 .

[7]  Jawad Iounousse,et al.  Using an unsupervised approach of Probabilistic Neural Network (PNN) for land use classification from multitemporal satellite images , 2015, Appl. Soft Comput..

[8]  Wang Weixing,et al.  A Method of Road Extraction from High-resolution Remote Sensing Images Based on Shape Features , 2009 .

[9]  Germain Forestier,et al.  Ontology-Based Object Recognition for Remote Sensing Image Interpretation , 2007 .

[10]  Francesca Bovolo,et al.  Updating Land-Cover Maps by Classification of Image Time Series: A Novel Change-Detection-Driven Transfer Learning Approach , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Robin Spence,et al.  Validating Assessments of Seismic Damage Made from Remote Sensing , 2011 .

[12]  M. Berthod,et al.  Automatic classification of planktonic foraminifera by a knowledge-based system , 1994, Proceedings of the Tenth Conference on Artificial Intelligence for Applications.

[13]  Siamak Khorram,et al.  Development of a modified neural network-based land cover classification system using automated data selector and multiresolution remotely sensed data , 2011 .

[14]  Zhang Jing.  Change detection of remote sensing image for earthquake damaged buildings and its application in seismic disaster assessment , 2002 .