PROJECT TITLE :
Automatic Change Analysis in Satellite Images Using Binary Descriptors and Lloyd–Max Quantization
In this letter, we gift a novel technique for unsupervised change analysis that leads to a method of ranking the changes that occur between two satellite pictures acquired at different moments of time. The proposed change analysis relies on binary descriptors and uses the Hamming distance as a similarity metric. So as to render a fully unsupervised solution, the obtained distances are additional classified using vector quantization methods (i.e., Lloyd's algorithm for optimal quantization). The ultimate goal in the amendment analysis chain is to create change intensity maps that provide an outline of the severeness of changes in the world under analysis. In addition, the proposed analysis technique will be simply adapted for change detection by choosing only two levels for quantization. This discriminative technique (i.e., between modified/unchanged zones) is compared with different previously developed techniques that use principal component analysis or Bayes theory as starting points for his or her analysis. The experiments are carried on Landsat images at a thirty-m spatial resolution, covering an area of roughly $59 times 51 mboxkm^2$ over the surroundings of Bucharest, Romania, and containing multispectral information.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here