Study On Geometric Feature Detection Of Geometric Characteristics Of Road Aggregates Based On Binocular Vision

With the continuous development of China’s traffic cause and people’s requirements on road traffic travel constantly improve,China has put forward higher requirements on the quality of road construction.Asphalt concrete pavement and cement concrete pavement is the main form of the highway,has played an important role in the road construction in our country,the geometric properties of the aggregate is mainly refers to the particle size and shape of the aggregate and asphalt concrete aggregate gradation of geometrical properties of asphalt concrete pavement design and road performance,durability,etc.Has a great influence.In this paper,binocular vision is used to detect relevant parameters of geometric features of concrete aggregates,and a binocular vision based geometric features detection platform of concrete aggregates is developed.In addition,practical application is also carried out,and the detection results are compared with those of traditional detection methods to verify the effectiveness of the method.This paper puts forward that the aggregate model of binocular vision recovery method,analyzed the theory of binocular vision model and realization method,studied the key calculation methods of binocular vision,combined with image acquisition when light conditions,external interference and aggregate nature of itself,for aggregate geometric feature detection image acquisition method to reduce the error caused by collecting images,using the method of calibration,the distortion correction,three-dimensional correction,image processing method for processing the digital images of object to be tested,aggregate model is proposed for the binocular vision recovery method.The 3d model of aggregate surface is recovered by binocular vision,which lays a foundation for further extraction of geometric features of aggregates.Paper puts forward the aggregate particle size detection method based on binocular vision,the physical model of binocular vision,through the camera model and parameter calculation calibration to determine the number of measuring point distribution per unit length,by using the methods of image enhancement,threshold segmentation,aggregate of digital image feature extracting information,combined with the 3 d model to calculate the particle size of aggregate,and compares the traditional detection methods of test results,verify the validity of the model and method.Paper puts forward the aggregate geometric feature detection method based on binocular vision,the use of binocular vision and digital image processing techniques to deal with 3 d model of aggregate,the aggregate creating 3 d model,through the image enhancement,image segmentation,image morphology method to aggregate digital image processing,extraction of the geometrical characteristics of the aggregate information.The testing platform was built.Based on the basic principle of binocular vision and the extraction method of geometric features of aggregates,the construction of the detection platform of geometric features of aggregates based on binocular vision is designed,including the design and construction of the collection device of aggregates image and the design of image information processing algorithm and software.Detection platform for practical engineering application,using the design of aggregate geometric feature detection platform has carried on the field test on the geometrical characteristics of aggregate,restore the surface of the aggregate three victor model,extract the geometric feature of aggregate information,and test results are compared with the traditional method,on the geometrical characteristics of aggregate information is analyzed and evaluated.