TY - GEN
T1 - Multispectral imaging based on natural neighbor interpolation with rendering under various illuminations
AU - Chou, Tzren Ru
AU - Hsieh, Chi Heng
N1 - Publisher Copyright:
© 2017 Association for Computing Machinery.
PY - 2017/8/25
Y1 - 2017/8/25
N2 - In this study, we constructed the multispectral imaging by the reconstruction of spectral reflectance. This reconstruction method was based on natural neighbor interpolation, abbreviated as NNI. The experiments focused on the feasibility of the reconstruction of reflectance spectra with multispectral images, and the experimental targets were divided into two types; paper packages, and portraits. The performance was evaluated in terms of root mean square error (RMSE), color difference (?E), and peak signal-tonoise ratio (PSNR). The experimental results showed that the proposed NNI reconstruction method was able to directly generate target spectra from sRGB channel values accurately. Furthermore, even after rendering under various illuminations, the average of color difference was still within an acceptable range.
AB - In this study, we constructed the multispectral imaging by the reconstruction of spectral reflectance. This reconstruction method was based on natural neighbor interpolation, abbreviated as NNI. The experiments focused on the feasibility of the reconstruction of reflectance spectra with multispectral images, and the experimental targets were divided into two types; paper packages, and portraits. The performance was evaluated in terms of root mean square error (RMSE), color difference (?E), and peak signal-tonoise ratio (PSNR). The experimental results showed that the proposed NNI reconstruction method was able to directly generate target spectra from sRGB channel values accurately. Furthermore, even after rendering under various illuminations, the average of color difference was still within an acceptable range.
KW - Natural neighbor interpolation
KW - Reconstruction of spectral reflectance
KW - Spectral reflectance
UR - http://www.scopus.com/inward/record.url?scp=85033570212&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85033570212&partnerID=8YFLogxK
U2 - 10.1145/3133264.3133269
DO - 10.1145/3133264.3133269
M3 - Conference contribution
AN - SCOPUS:85033570212
T3 - ACM International Conference Proceeding Series
SP - 45
EP - 49
BT - Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017
PB - Association for Computing Machinery
T2 - 2017 International Conference on Advances in Image Processing, ICAIP 2017
Y2 - 25 August 2017 through 27 August 2017
ER -