Evaluation of Retrieved Categories from a Terrasar-X Benchmarking Data Set

Advanced interpretation of satellite images calls for automated content analysis as well as interactive content search. A typical example of such systems is EOLib, an ESA funded project that already demonstrated the application potential of TerraSAR-X data within a satellite payload ground segment. In this paper, we analyze the validation results of image content classification using a large set of selected TerraSAR-X images. The classification was done with a cascaded learning method. The main advantage of this method is a coarse-to-fine approach for semantically annotating pixel patches with decreasing size. Once a reliable label is found for a given pixel patch, no further subdivision into still smaller patch sizes is necessary. This leads to a considerable reduction of the computational effort during classification of large-size satellite images.

[1]  Shiyong Cui,et al.  Improved image classification by proper patch size selection: TerraSAR-X vs. Sentinel-1A , 2016, 2016 International Conference on Systems, Signals and Image Processing (IWSSIP).

[2]  Shiyong Cui,et al.  Data Mining and Knowledge Discovery for the TerraSAR-X Payload Ground Segment , 2015 .

[3]  Shiyong Cui,et al.  Pattern Retrieval in Large Image Databases Using Multiscale Coarse-to-Fine Cascaded Active Learning , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[4]  Shiyong Cui,et al.  Cascade Active Learning for Evolution Pattern Extraction from SAR Image Time Series , 2013 .

[5]  Geshi Tang,et al.  Matching suitable feature construction for SAR images based on evolutionary synthesis strategy , 2013 .

[6]  Carlos López-Martínez,et al.  A Public Database of Simulated Multidimensional SAR Data for Techniques Validation , 2008, IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium.

[7]  Sungho Kim,et al.  Synthetic SAR/IR database generation for sensor fusion-based A.T.R. , 2015, 2015 12th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI).

[8]  Lanqing Huang,et al.  OpenSARShip: A Dataset Dedicated to Sentinel-1 Ship Interpretation , 2018, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[9]  Shiyong Cui,et al.  The Earth Observation Image Librarian (EOLIB): The Data Mining Component of the TerraSAR-X Payload Ground Segment , 2016 .

[10]  Shiyong Cui,et al.  Information Content of Very-High-Resolution SAR Images: Semantics, Geospatial Context, and Ontologies , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.