GPU Framework for Change Detection of Multitemporal Hyperspectral Images

Nowadays, it is increasingly common to detect land cover changes through remote sensing multidimensional images captured at different timeframes. A great part of the available change detection methods focus on pixel-based operations. The use of spectral-spatial techniques helps to improve the results but increases the processing times. In this work, a GPU framework to find object-based differences in multitemporal remote sensing hyperspectral data is presented. The method is based on Change Vector Analysis (CVA) and involves different segmentation algorithms to take into account the spatial information of the images. The Spectral Angle Mapper (SAM) distance and Otsu's thresholding are also used. The projection in GPU allows change detection applications to be efficiently computed in real-time. The method is applied to agricultural hyperspectral images obtaining a speedup of 16.1x against the OpenMP implementation.

For information see the readme in the files.

* Santa Barbara

* Bay Area

For information see the readme in the files.

* outputscva.zip

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.