Abstract:
Aggregate size and shape measurements are extremely important issues in
mining and construction industry because of it directly affect the performance of
aggregate products, also there is a prime need of textural analysis in many fields
including geological and geotechnical studies. Traditional methods are time consuming
and complex. In the present research, we applied DIP (Digital Image Processing)
techniques for grain size analysis. Mainly, there are four sections which are unattached
particles/fragment analysis, Attached particles/fragment Analysis, Moving
particles/fragment Analysis and Colour, texture based classification. In unattached
particles analysis, particles were spread without contacting each other and then analysis
done. In attached particles analysis, watershed transformation was applied to
distinguish particles and then analysis was done. Moving particle analysis were
performed by acquiring a video of free falling particles and generating contact-less flow
of particles using video processing techniques. Colour and Texture based classification
was done by separating the RGB (red, green, blue) bands and calculating mean, standard
deviation and smoothness and then k-rnean classification were performed. Finally results
from Image processing methods were compared with the conventional methods. The
method developed by the research was successfully applied in aggregate and sediment
analysis.
Citation:
Siriwardana, M.U.V.A., Deshapriya, N.L., Abeysinghe, A.M.K.B., Weerawarnakula, S., & Premasiri, R. (2013). Digital image processing technique for particle size, shape and mineralogical, textural analysis. In P.V.A. Hemalal (Ed.), Proceedings of the 7th National Conference on Earth Resources Management (pp. 16-20). Department of Earth Resources Engineering, University of Moratuwa.