Study On Paving Separation Detection Method Of High Performance Asphalt Mixture Based On Image

Asphalt pavement,as an important form of road,is widely used in the field of public transportation in China.During the construction process of asphalt pavement,the segregation of asphalt mixture is one of the important factors that cause early damage to the pavement.The commonly used traditional segregation detection methods currently have problems such as time-consuming,labor-intensive,and low efficiency.Currently,there is a lack of fast,efficient,inexpensive,and stable methods for detecting the segregation of asphalt mixtures.With the development of digital image processing technology,its efficient and non-destructive advantages provide a new research direction for the field of highway construction and quality inspection.This article analyzes the mechanism of segregation in asphalt mixture materials and proposes a segregation detection method based on digital image processing technology.At the same time,a segregation detection system for asphalt pavement paving is developed based on the GUI design platform of Matlab software.The main research content of this article is as follows:Analyze the grading characteristics of high-performance asphalt mixtures and their impact on the segregation of asphalt mixtures through the Bailey method.Study the mechanism of material segregation in various stages of asphalt mixture production and construction,analyze and summarize the advantages and disadvantages of commonly used segregation detection methods,and determine the research direction of segregation detection based on digital image technology.Using digital image technology to detect asphalt pavement segregation,the asphalt pavement image is first preprocessed to obtain a suitable image preprocessing method for asphalt pavement.Secondly,a three-dimensional model of asphalt pavement surface texture is established using the shading restoration method,and the image texture depth of the asphalt pavement three-dimensional model is calculated.The correlation analysis between the image texture depth and the corresponding construction depth obtained from the sand laying method test is conducted.The research results indicate that there is a good correlation between the two,and based on the construction of depth segregation evaluation standards,a segregation evaluation standard for image texture depth is proposed.On this basis,the U-Net deep learning model is trained to segment images of asphalt mixture particles and analyze the uniformity of the distribution of coarse and fine aggregates in asphalt mixture.Finally,a segregation detection method for asphalt mixture paving was obtained by combining texture depth and uniformity.Based on the segregation detection method studied in this article,a segregation detection system for asphalt pavement paving is developed through the GUI design platform of Matlab software,achieving efficient non-destructive detection of segregation.Compared with current commonly used testing methods,it has advantages such as high efficiency,low cost,convenient use,and non-destructive testing.