Hyperspectral near infrared image calibration and regression

Mikko Mäkelä; Paul Geladi; Marja Rissanen; Lauri Rautkari; Olli Dahl

doi:10.1016/j.aca.2020.01.019

Hyperspectral near infrared image calibration and regression

Mikko Mäkelä^*
, Paul Geladi
, Marja Rissanen
, Lauri Rautkari
, Olli Dahl

^*Corresponding author for this work

Not published at VTT

Aalto University
Swedish University of Agricultural Sciences

Research output: Contribution to journal › Article › Scientific › peer-review

17 Citations (Scopus)

Abstract

Reference materials are used in diffuse reflectance imaging for transforming the digitized camera signal into reflectance and absorbance units for subsequent interpretation. Traditional white and dark reference signals are generally used for calculating reflectance or absorbance, but these can be supplemented with additional reflectance targets to improve the accuracy of reflectance transformations. In this work we provide an overview of hyperspectral image regression and assess the effects of reflectance calibration on image interpretation using partial least squares regression. Linear and quadratic reflectance transformations based on additional reflectance targets decrease average measurement errors and make it easier to estimate model pseudorank during image regression. The lowest measurement and prediction errors were obtained with the column and wavelength specific quadratic transformations which retained the spatial information provided by the line-scanning instrument and reduced errors in the predicted concentration maps.

Original language	English
Pages (from-to)	56-63
Journal	Analytica Chimica Acta
Volume	1105
DOIs	https://doi.org/10.1016/j.aca.2020.01.019
Publication status	Published - 8 Apr 2020
MoE publication type	A1 Journal article-refereed

Funding

We acknowledge the contribution of Håkan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project. We acknowledge the contribution of Håkan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project.

Keywords

Hyperspectral imaging
Partial least squares
Prediction
Pseudorank
Reflectance calibration
Textile analysis

Access to Document

10.1016/j.aca.2020.01.019

Cite this

@article{4d865fe132c74da493be849ca6eaf725,

title = "Hyperspectral near infrared image calibration and regression",

abstract = "Reference materials are used in diffuse reflectance imaging for transforming the digitized camera signal into reflectance and absorbance units for subsequent interpretation. Traditional white and dark reference signals are generally used for calculating reflectance or absorbance, but these can be supplemented with additional reflectance targets to improve the accuracy of reflectance transformations. In this work we provide an overview of hyperspectral image regression and assess the effects of reflectance calibration on image interpretation using partial least squares regression. Linear and quadratic reflectance transformations based on additional reflectance targets decrease average measurement errors and make it easier to estimate model pseudorank during image regression. The lowest measurement and prediction errors were obtained with the column and wavelength specific quadratic transformations which retained the spatial information provided by the line-scanning instrument and reduced errors in the predicted concentration maps.",

keywords = "Hyperspectral imaging, Partial least squares, Prediction, Pseudorank, Reflectance calibration, Textile analysis",

author = "Mikko M{\"a}kel{\"a} and Paul Geladi and Marja Rissanen and Lauri Rautkari and Olli Dahl",

note = "Funding Information: We acknowledge the contribution of H{\aa}kan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project. Funding Information: We acknowledge the contribution of H{\aa}kan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project. Publisher Copyright: {\textcopyright} 2020 Elsevier B.V.",

year = "2020",

month = apr,

day = "8",

doi = "10.1016/j.aca.2020.01.019",

language = "English",

volume = "1105",

pages = "56--63",

journal = "Analytica Chimica Acta",

issn = "0003-2670",

publisher = "Elsevier",

}

TY - JOUR

T1 - Hyperspectral near infrared image calibration and regression

AU - Mäkelä, Mikko

AU - Geladi, Paul

AU - Rissanen, Marja

AU - Rautkari, Lauri

AU - Dahl, Olli

N1 - Funding Information: We acknowledge the contribution of Håkan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project. Funding Information: We acknowledge the contribution of Håkan Nilsson from the Swedish University of Agricultural Sciences in the practical arrangements for the imaging procedure and Anna Leinonen from Aalto University in the preparation of the textile samples. This work was financially supported by the Academy of Finland under grant agreement no. 309881 and the Strategic Research Council of the Academy of Finland under grant agreement no. 327296 - the FINIX project. Publisher Copyright: © 2020 Elsevier B.V.

PY - 2020/4/8

Y1 - 2020/4/8

N2 - Reference materials are used in diffuse reflectance imaging for transforming the digitized camera signal into reflectance and absorbance units for subsequent interpretation. Traditional white and dark reference signals are generally used for calculating reflectance or absorbance, but these can be supplemented with additional reflectance targets to improve the accuracy of reflectance transformations. In this work we provide an overview of hyperspectral image regression and assess the effects of reflectance calibration on image interpretation using partial least squares regression. Linear and quadratic reflectance transformations based on additional reflectance targets decrease average measurement errors and make it easier to estimate model pseudorank during image regression. The lowest measurement and prediction errors were obtained with the column and wavelength specific quadratic transformations which retained the spatial information provided by the line-scanning instrument and reduced errors in the predicted concentration maps.

AB - Reference materials are used in diffuse reflectance imaging for transforming the digitized camera signal into reflectance and absorbance units for subsequent interpretation. Traditional white and dark reference signals are generally used for calculating reflectance or absorbance, but these can be supplemented with additional reflectance targets to improve the accuracy of reflectance transformations. In this work we provide an overview of hyperspectral image regression and assess the effects of reflectance calibration on image interpretation using partial least squares regression. Linear and quadratic reflectance transformations based on additional reflectance targets decrease average measurement errors and make it easier to estimate model pseudorank during image regression. The lowest measurement and prediction errors were obtained with the column and wavelength specific quadratic transformations which retained the spatial information provided by the line-scanning instrument and reduced errors in the predicted concentration maps.

KW - Hyperspectral imaging

KW - Partial least squares

KW - Prediction

KW - Pseudorank

KW - Reflectance calibration

KW - Textile analysis

UR - https://www.scopus.com/pages/publications/85077914958

U2 - 10.1016/j.aca.2020.01.019

DO - 10.1016/j.aca.2020.01.019

M3 - Article

C2 - 32138926

AN - SCOPUS:85077914958

SN - 0003-2670

VL - 1105

SP - 56

EP - 63

JO - Analytica Chimica Acta

JF - Analytica Chimica Acta

ER -

Hyperspectral near infrared image calibration and regression

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this