Remote sensing segmentation benchmark

In this work we present the enrichment of the Prague texture segmentation data-generator and benchmark (PTSDB) also for the assessment of the remote sensing image segmenters. The PTSDB tool is a web based (http://mosaic.utia.cas.cz) service designed for real-time performance evaluation, mutual comparison, and ranking of various supervised or unsupervised static or dynamic image segmenters. PTSDB supports rapid verification and development of new segmentation approaches. The remote sensing datasets contain tenspectral ALI satellite images and their RGB subsets, with optional additive noise resistance checking. Alternative setting options allow to test also scale, rotation or illumination invariance. The benchmark functionality is demonstrated by testing and comparing six remote sensing segmentation algorithms.

[1]  Charless C. Fowlkes,et al.  Contour Detection and Hierarchical Image Segmentation , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Giuseppe Scarpa,et al.  Recursive Texture Fragmentation and Reconstruction segmentation algorithm applied to VHR images , 2009, 2009 IEEE International Geoscience and Remote Sensing Symposium.

[3]  Michal Haindl,et al.  Illumination invariant unsupervised segmenter , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[4]  Michal Haindl,et al.  Texture segmentation benchmark , 2008, 2008 19th International Conference on Pattern Recognition.

[5]  Pierre Soille,et al.  Morphological Texture Features for Unsupervised and Supervised Segmentations of Natural Landscapes , 2007, IEEE Transactions on Geoscience and Remote Sensing.