Quantitative Study Of Shape Characteristics For Coarse Aggregate Based On Digital Image Processing

In Shape characteristics of coarse aggregate, the form and angularity are involved.The former is related to particle flat and elongated(F&E)ratio and the latter to the sharpness of the corners in the outline of the particle,which have significant effects on the properties of asphalt pavement. F&E particles are undesirable since they have a tendency to break easily under loading.More angular particles can increase stone-on-stone interlocking and consequently enhance the pavement strength and the resistance to permanent deformation. At present,the test procedures for them are time-consuming and labor-intensive, particularly the test for angularity is subjective.In order to quantify shape characteristics of coarse aggregate in a fast,objective and accurate manner,this dissertation,based on DIP techniques, systematically studies the quantification of shape characteristics. Mainly three parts are included,which are the acquisition,processing and analysis of coarse aggregate images,the selection and calculation of the quantitative indexes of shape characteristics,the test research for the verification of the quantitative indexes selected.In this dissertation,coarse aggregate images are captured by CCD camera and backlighting box, image distortion is corrected using Lens Correction filter of Photoshop CS2,and then Image-Pro Plus is used for image recognition.Eight quantitative indexes are selected for test research after a detail review about quantitative indexes for shape characteristics of coarse aggregate is made,among that, Elongation,Flakiness,Sphericity,Shape Factor are the indexes for form,and the remains including Angularity(R),Angularity(Rmax),Angularity(P),Angularity(Pc) are for angularity.In order to show the macroscopic property of coarse aggregate,volumetric model of coarse aggregate,based on 2d images,is founded,and sieving correlation is done to verify the model with twenty seven coarse aggregate samples,then,weighted mean of eight quantitative indexes are caculated by the program based on EXCEL VBA.Finally,for the sake of verification, the uncompacted voids test of twenty coarse aggregate samples are made, and the experiment results are correlated to that of DIP test.The study showes that Elongation is not suitable for quantifing form,and the indexes, Flakiness,Sphericity,Shape Factor,Angularity(P),Angularity(Pc),magnifying the ends of shape characteristics as much as possible,are better to distinguish all kinds of shape characteristics,among that, Shape Factor is the best index for form and Angularity(P) is the best one for angularity.