Study Of Image-based Aggregate Morphlogical Characteristics And Mixture Gradation Analysis Techniques

The performance of asphalt pavement is affected by many factors.The shape,angularity and texture characteristics of the mineral aggragate particles affect such performances as the durability,workability,shear and fatigue of the pavement.The measurement of the morphological characteristics of mineral aggregates is the foundation for guarantee the performance of the pavement.During the production process of the asphalt mixture,the gradation of the mineral aggragete is sometimes deviated,which will results in the degradation of the asphalt mixture performance,and the measurement of the mineral mixture gradation is of importance.During the construction of the asphalt pavement,the measurement of the asphalt mixture gradation is an important part in the quality control process.The testing measurement methods used currently are time comsuming and the delay of the measurement results will affect the performance of the asphalt pavement.In this paper,digital image processing techniques are used to extract and analyze the morphological characteristics of the mineral aggregate particles,real time acquire and appraise of the gradation of the mineral mixture and the gradation analysis of the asphalt mixture.The paper studies the digital image processing techniques and its application in the measurement of the particle material,uses the contrast constraints histogram transform to enchance the grayscale image of the asphalt mixture and uses the binary morphological filtering to carry on the edge detection of the aggragate particle image,then,selects the pointer methods used in C#image processing to develop the image processing and analysis application.Using self-designed image acquisition platform,this paper collects the particle images of the limestone aggragates with different sizes and different type and studies the morphological characteristics distributions of the average particle size,roundness,angularity and texture.The paper designs the wear tests to wear the coarse aggregate particles of the limestone with 500 and 1000 times,then acquires the particle images and carries on particle analysis to study the wear effects on the coarse aggregate roundness,angularity and texture characteristics.Further,this paper forms the rutting specimen by using the weared coarse aggregates and studies the wear effects to the high temperature rutting of the asphalt mixture.This paper designs the asphalt mixture cold hopper simulation device,selects the E2V line scan camera,Solios image acquisition card,Nikon lens,light source and the corresponding image acquisition software to build the mineral aggregate mixture online image acquisition platform.The paper studies the effects of the light source on imaging and selects the infrared laser as the auxiliary light source.The paper studies the effects of line period of the line scan camera on imaging and determines the best line period.Finally,the paper uses the double buffer technique to achieve the realtime acquisition and prcessing of the mineral aggregate mixture images.The paper designs the asphalt pavement core sample rotating device and builds the asphalt pavement core sample image acquisition platform based on the Basler L402K line scan camera.The paper determines the line period through experiments to acquire the side images of the core sample with no distortion and realizes continuous image acquisition and storage with Matrox Mil 8.0.The paper develops the asphalt mixture image analysis software by using Visual C#.NET 2008 and achieves such processes as grayscale image obtaining,threshold segmentation,binary morphological filtering,edge tracing and particle analysis,and ultimately realizes the gradation measurement and analysis of the asphalt mixture.Mineral filler plays an important role in the asphalt mixture.The paper provides an image analysis based method,which is named color table method,for the mineral filler content analysis of the asphalt mixture.The method uses the image analysis techniques and acquires the asphalt mortar images with different asphalt and different mineral with different content,then gets the normalized histogram parameters and stores to the color table for the following search and matching by histogram analysis.In the actual measurement process,the asphalt type and the mineral filler type are known,the cross section image of the asphalt mixture are obtained by using a digital camera,and the asphalt mortar parts image of the asphalt mixture can be obtained by using the slice transformation to the asphalt mixture image,then the normalized histogram parameters can be get by histogram analysis to the asphalt mortar parts image.Then,through comparative analysis to the normalized histogram obtained and the histogram stored in the color table by using the histogram correlation method,the mineral filler contents in the asphalt mixture could be desided.From the measurement and matching result of the asphalt mortar image with the filler bitumen ratios of 0.8 and 1.6 shows that the color table method can measure the mineral filler contents in the asphalt mixture and with the color table more perferct,the powerful of the color table is.