Open AccessIdentification of fat, protein matrix, and water/starch on microscopy images of sausages by a principal component analysis‐based segmentation scheme
Author(s) -
Kohler Achim,
Høst Vibeke,
Enersen Grethe,
Ofstad Ragni
Publication year - 2006
Publication title -
scanning
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.359
H-Index - 47
eISSN - 1932-8745
pISSN - 0161-0457
DOI - 10.1002/sca.4950250302
Subject(s) - segmentation , principal component analysis , pixel , artificial intelligence , starch , computer vision , matrix (chemical analysis) , microscopy , computer science , image segmentation , pattern recognition (psychology) , biological system , materials science , chemistry , food science , biology , composite material , optics , physics
A color‐based segmentation scheme applied to microscopy images of cryosectioned sausages is proposed. The segmentation scheme is capable of segmenting three different levels on the microscopy images: the fat particles, the protein matrix, and water/starch. The method is based on principal component analysis. A user‐friendly program was developed for the manual segmentation of a selection of image pixels by microscopists. Principal component models based on the manually classified pixels are then used to segment fat, protein matrix, and starch/water on microscopy images. The program can also be used as a training tool for microscopists.