We present an introduction to Multivariate Image Regression (MIR) with a se
lection of illustrative application studies. Generalisation from two-way mu
ltivariate calibration to the three-way regimen leads to-at least-three alt
ernative image regression cases depending on the nature of the available Y-
data: IPLS-Y-discrim; IPLS-Y-grid; IPLS-Y-total. A systematic image regress
ion typology is briefly introduced.
We here present the core of the principles of applied MIR. Two major MIR ap
plication studies are worked through, a food mass product industrial inspec
tion study (IPLS-Y-discrim) and a food product (fruit) storage stability im
age analytical monitoring (IPLS-Y-grid). These exemplifications are present
ed as archetypes, representing a much wider range of potential industrial/t
echnological application areas. Based on simple three-channel imagery (in o
rder to simulate many industrial systems), they nevertheless represent all
higher-dimensional multivariate image cases as well, since the pertinent MI
R principles and software are invariant w.r.t. any number of channels/varia
bles employed.
The present paper represents one major element of our work towards establis
hing a complete, stand-alone facility for Multivariate Image Regression (MI
R); the second paper in this series deals with the development, implementat
ion and extensive exemplifications of a complementary cross-validation faci
lity. (C) 2001 Elsevier Science B.V. All rights reserved.