Original Articles Open access

A METHOD FOR SEGMENTING FOOD COLOUR IMAGES

Authors

Published
30 June 2008
Section
Original Articles
Pages
53-56

Abstract

In this note, a method for segmenting foods from their backgrounds in colour images is presented. The proposed method has three steps: i) determination of the optimal decision plane for the segmentation of an image; ii) coarse segmentation of the image; iii) morphological operations in order to correct possible errors in the segmented image. The method was implemented in MATLAB and tested on 40 colour images of foodstuff with very different morphological and chromatic characteristics, including meat, baked products, fruit and tubers. The experimental results are presented and the performance of method in the segmentation process is assessed. The method has shown to be both effective and efficient also for colour images with high spatial resolution.

Metrics

Citations and online attention from Dimensions, Altmetric and PlumX.