CATI: Application of image analysis systems for seed coat fragment detection and on other fibre characterization
Research conducted by Cirad in the 1990s focussed on reducing the number of seed coat fragments (SCF) in cotton lint. Technologies were therefore developed to characterize cotton samples and the TRASHCAM project was initiated to develop a sample preparation technique and the software for counting and sizing SCF in cotton webs. Various CATI software systems were produced to analyse images captured on a grey scale (256 grey levels). This technology was then showed to be effective for SCF in breeding programmes, but was dependent upon the preparation technique, which involved using a mini-card that is assumed to remove everything in the lint but SCF. Other preparation devices were tested, but impurities remained in the cotton web together with the SCF. To overcome this problem, software was developed that took into account the colour of the foreign matter in the lint. This software, called CATI, proved to be effective and thanks to its versatility was considered as potentially of interest for determining the number of dark or stained wool fibres in wool webs. Results obtained with this software in the measurement of dark and stained fibres in wool webs are presented in this paper.
Main Authors: | , , , , |
---|---|
Format: | conference_item biblioteca |
Language: | eng |
Published: |
NCCA
|
Subjects: | Q60 - Traitement des produits agricoles non alimentaires, coton, Gossypium, graine de coton, égrenage du coton, pureté, sélection, qualité technologique, méthode, analyse d'image, http://aims.fao.org/aos/agrovoc/c_1926, http://aims.fao.org/aos/agrovoc/c_3335, http://aims.fao.org/aos/agrovoc/c_25473, http://aims.fao.org/aos/agrovoc/c_34005, http://aims.fao.org/aos/agrovoc/c_26812, http://aims.fao.org/aos/agrovoc/c_6951, http://aims.fao.org/aos/agrovoc/c_29772, http://aims.fao.org/aos/agrovoc/c_4788, http://aims.fao.org/aos/agrovoc/c_36762, |
Online Access: | http://agritrop.cirad.fr/527190/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|