Open AccessSorting Speckled Granites (Nehbandan) and Measuring Their Surface Veining using Machine Vision
Author(s) -
Hossein Kardan Moghaddam*,
Amir Rajaei,
Mohamad Reza Maraki
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.c4774.098319
Subject(s) - sorting , texture (cosmology) , automation , artificial intelligence , computer science , quality (philosophy) , surface (topology) , computer vision , quartz , geology , image (mathematics) , pattern recognition (psychology) , mineralogy , mathematics , engineering , mechanical engineering , physics , algorithm , paleontology , geometry , quantum mechanics
Quality control and the appearance evaluation of stones are quite challenging in the industry today. The similar appearance of different stones containing the same minerals may result in economic losses, and if the customers fail to identify the type of slates delivered to them correctly, disagreements may arise between the buyers and granite vendors. This study is an attempt toward the automation of surface quality assessment of the Nehbandan(Iran) speckled granite and measurement of the surface patterns under fixed conditions using image processing techniques in order to classify the granite tiles based on their type and amount of impurities and veins. The experimental tests comparing the presented approach with the texture descriptors in the introduced dataset prove the efficiency of the proposed method and its applications under controlled conditions, including the classification of speckled granite tiles with different image resolutions.