Paper published in a book (Scientific congresses and symposiums)
Caracterización y cuantificación autmatizadas de menas metálicas mediante visión artificial : proyecto cameva
Castroviejo, Ricardo; Catalina, Juan-Carlos; Bernhardt, Heinz-Juergen et al.
2008In Actos del XIII Congreso Latino Americano de Geologia
 

Files


Full Text
PUB_2008_06_RC Cuantificacion menas.pdf
Author postprint (602.02 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Abstract :
[en] The CAMEVA project (Automated system for the identification and quantitative measurement of ore minerals) aims to develop an automated system able to perform the identification and quantification of the different minerals present in a sample of metallic ores, in order to facilitate their industrial use. The system comprises a motorised reflected light microscope, a filter wheel (monochromator), a research BW camera, and a computer running a digital image analysis software associated to an expert system specially developed for this application. Multispectral reflectance values are measured for visible light and near IR and UV (350-1000 nm), and no polarisation is usually employed in order to reduce measurement variability due to mineral anisotropy. Mineral identification is carried out by comparison of measured reflectances with spectral reflectance databases (e.g. IMA-COM). Previous experiences showed that the most common ores can be differentiated through the analysis of a microscopic RGB colour image, by constraining the search with additional information such as the type of deposit and the corresponding mineral association. The project goes further, and intends to identify most of the ores from multi-spectral images, including non-visible bands, resorting to an expert system specifically built to deal with the different types of information available (spectral reflectance, sample origin, mineral associations, morphology, etc). 2 Validation of the method will be carried out by comparative análisis of standards and samples exchanged between the laboratories. Development pays special attention to the total automation of the process (including calibration) and to the self-learning capabilities of the system. Contacts with research institutions and with the industry will allow to apply the system to real problems. The system can be used to control the processing of ores as well as to prevent pollution for the protection of the environment (e.g. sulphides in coal or in industrial tailings). Further applications can be envisaged in the future: e.g. other minerals, coal or materials, particles, etc.
Disciplines :
Geological, petroleum & mining engineering
Author, co-author :
Castroviejo, Ricardo
Catalina, Juan-Carlos
Bernhardt, Heinz-Juergen
Espí, Jose Antonio
Pirard, Eric  ;  Université de Liège - ULiège > Département Argenco : Secteur GeMMe > Géoressources minérales & Imagerie géologique
Samper, Josefina
Brea, Carolina
Segundo, Fernando
Locutura, Juan
Perez-Barnuevo, Laura
Sanchez, Lazaro
Fidalgo, Angel
Language :
English
Title :
Caracterización y cuantificación autmatizadas de menas metálicas mediante visión artificial : proyecto cameva
Publication date :
2008
Event name :
XIII Congreso Latino Americano de Geologia
Event place :
Lima, Peru
Audience :
International
Main work title :
Actos del XIII Congreso Latino Americano de Geologia
Available on ORBi :
since 09 June 2011

Statistics


Number of views
294 (3 by ULiège)
Number of downloads
365 (3 by ULiège)

Bibliography


Similar publications



Contact ORBi