Font Size: a A A

The Research On Geometric Features Of Aggregates Based On Digital Image Processing

Posted on:2019-07-08Degree:MasterType:Thesis
Country:ChinaCandidate:X M WuFull Text:PDF
GTID:2382330563995253Subject:Transportation engineering
Abstract/Summary:PDF Full Text Request
As one of the basic units of mixture,the aggregate has an important influence on the engineering performance of the pavement structure.The application of computer technology has led to a tremendous shift in the way that people think about solving problems.The traditional methods of gauge and caliper can no longer meet people’s production needs.It has become a new research trend to study the geometric characteristics of aggregate particles with the aid of computer technology so as to replace the subjective measurement by using the quantitative measurement.This paper divides the geometric characteristics of aggregate into three levels,that is,shape characteristics,angularity and texture,and systematically analyzes the characteristics with the image technology.By analyzig the principle and advantages of CCD,industrial CT and scanner,an improved scanner device is designed to capture the image of aggregate particle in this paper.In order to denoise,three methods are compared.And then,the median filtering method is used to denoise the aggregate images.After the effect and running time of the three image segmentation methods are compared,the OSTU algorithm is selected to extract the aggregates.In the processing of aggregate morphology,an improved morphological algorithm based on the intersecting cortex model(ICM)is designed and realized.Comparing with mathematical morphology algorithms,the improved algorithm is identical in processing effects,but the processing speed of ICM is 3-4 times as faster as the traditional one.In order to solve the problem of internal holes and the incomplete segmentation of aggregate images,a morphological algorithm based on ICM is used to repair the aggregates.In order to obtain the geometric parameters of the aggregate particles themselves,the connected regions of the aggregates are marked and the feature extraction is realized.After determining the area,perimeter,the major axis,short axis and focal length of aggregate’s equivalent ellipse,this paper systematically proposes the geometric parameters of aggregates,including Shape Index SI1 and SI2,Form Factor,Angularity Index AI1 and AI2,Texture Factor.Aggregate Geometric Characteristic Evaluation System(AGCES)is developed by using MATLAB 2014a.In order to study the rationality of the geometric characteristics of aggregates,this paper applies the regular geometry instead of aggregates to test the evaluation system.Finally,different types and sizes of limestone,limestone(mill),basalt,gneiss and diabase are quantitatively analyzed,and compare with the results of the laboratory.The results show that there is a correlation between the geometric characteristics and the laboratory results of aggregates,which is an important way to objectively investigative the geometric characteristics of aggregates.The development of this system can provide an effective evidence for the evaluation of aggregate quality in road design,construction and maintenance.
Keywords/Search Tags:Image collection, Image segmentation, Aggregate, Geometric features
PDF Full Text Request
Related items