File(s) under permanent embargo
Context free band reduction using a convolutional neural network
conference contribution
posted on 2018-01-01, 00:00 authored by Ran Wei, Antonio Robles-KellyAntonio Robles-Kelly, Jose M AlvarezIn this paper, we present a method for content-free band selection and reduction for hyperspectral imaging. Here, we reconstruct the spectral image irradiance in the wild making use of a reduced set of wavelength-indexed bands at input. To this end, we use of a deep neural net which employs a learnt sparse input connection map to select relevant bands at input. Thus, the network can be viewed as learning a non-linear, locally supported generic transformation between a subset of input bands at a pixel neighbourhood and the scene irradiance of the central pixel at output. To obtain the sparse connection map we employ a variant of the Levenberg-Marquardt algorithm (LMA) on manifolds which is devoid of the damping factor often used in LMA approaches. We show results on band selection and illustrate the utility of the connection map recovered by our approach for spectral reconstruction using a number of alternatives on widely available datasets.
History
Event
International Association of Pattern Recognition. Workshops (2018 : Beijing, China)Volume
11004Series
International Association of Pattern Recognition WorkshopsPagination
86 - 96Publisher
SpringerLocation
Beijing, ChinaPlace of publication
Cham, SwitzerlandPublisher DOI
Start date
2018-08-17End date
2018-08-19ISBN-13
9783319977843Language
engPublication classification
E Conference publication; E1 Full written paper - refereedCopyright notice
2018, Springer Nature Switzerland AGEditor/Contributor(s)
Xiao Bai, Edwin Hancock, Tin Ho, Richard Wilson, Battista Biggio, Antonio Robles-KellyTitle of proceedings
S+SSPR 2018 : Structural, syntactic, and statistical pattern recognition : Proceedings of joint IAPR international workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR)Usage metrics
Categories
Keywords
Convolutional Neural NetworkContext Free Band ReductionHyperspectral imagingSpectral image irradianceScience & TechnologyTechnologyPhysical SciencesComputer Science, Artificial IntelligenceComputer Science, Interdisciplinary ApplicationsComputer Science, Software EngineeringComputer Science, Theory & MethodsMathematics, AppliedStatistics & ProbabilityComputer ScienceMathematicsSELECTIONInformation SystemsArtificial Intelligence and Image ProcessingDistributed Computing
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC