Spectral-spatial classification from multi-sensor compressive measurements using superpixels.

Abstract

Compressive spectral imaging (CSI) acquires coded projections of a spectral image by performing a modulation of the data cube followed by a spectral-wise integration. To avoid the spectral image reconstruction procedure, this paper proposes a classification approach that extracts features directly from multi-sensor CSI measurements. Particularly, the proposed method obtains the features by considering the spectral information extracted from Hyperspectral CSI measurements, and the local spatial information extracted by clustering the Multispectral CSI measurements using a superpixel algorithm. This approach is evaluated on Pavia University and Salinas Valley datasets. Extensive simulations show that considering the local spatial information boosts the overall accuracy up to 3% in comparison with traditional approaches that only uses the spectral information. Furthermore, the computation time of the approach that reconstructs, fuses and classifies takes approximately 87.43 [s], while classifying directly from multi-sensor compressive measurements takes only 0.74 [s], achieving similar classification results.

Publication
In 2019 IEEE International Conference on Image Processing (ICIP)
Spectral Image Classification Superpixels Algorithms Compressive Spectral Imaging Multi-sensor Measurements
Avatar
Carlos Hinojosa
Ph.D Student In Computer Science

I’m Carlos Hinojosa. Computer scientist and engineer with over six years of experience in scientific research and software development. My research interests are in computer vision, machine learning, computational imaging, sparse representation, and signal processing.