Evaluation Of Sliding Resistance Of Pavement Based On Machine Vision

The road surface must have good anti-slip performance to provide enough friction for the vehicle to ensure safe and smooth running of the vehicle.A good anti-slip performance can ensure the safety of driving,and choosing a reasonable anti-sliding performance index for the road is not only the needs of the project managers,but also the quality of the project itself.Traditional detection methods based on friction coefficient and construction depth are susceptible to test conditions.Studies have shown that anti-sliding performance is closely related to the three-dimensional topographical characteristics of pavements.Studying the three-dimensional topographical characteristics of road surface and analyzing its relationship with anti-sliding performance of road surface not only provides new ideas for the detection of anti-sliding performance of road surface,but also has guiding significance for the selection of daily maintenance and road building materials.In order to objectively and truly reflect the three-dimensional topographical structural features of the road surface,this paper introduces the machine vision theory into the three-dimensional surface reconstruction of the road surface,,designs and develops an experimental platform based on binocular stereo vision,and constructs an asphalt pavement reconstruction system based on binocular stereo vision.Based on the existing stereo matching algorithm,a normalized cross-correlation stereo matching algorithm based on credibility is proposed to recover the three-dimensional shape of road surface.Validate the results using the Croma 8106 3D coordinate system.Based on the relationship between three-dimensional surface appearance and skid resistance performance,this paper introduces the fractal theory.Fractal geometry is the most effective method to describe complex and irregular natural scenery.In this paper,the three-dimensional topography reconstruction of the four different grading specimens of the SMA vehicle smashed board is performed.The fractal dimension of the three-dimensional model of the road surface is extracted by the carpet covering method to describe its structural characteristics,and the average structure measured by the sanding method is obtained.The mean structural depth MTD measured by the sand deposition method and the measured pendulum value BPN measured by the pendulum instrument are respectively subjected to regression analysis with the three-dimensionalfractal dimension of the road surface to obtain an empirical formula.Finally,a verification experiment is performed.The results show that the three-dimensional pavement reconstruction system and the proposed stereo matching algorithm can recover the three-dimensional shape of the pavement with high accuracy.The three-dimensional texture model can accurately reflect the topographic texture features of asphalt pavement.The calculated three-dimensional fractal dimension has a good correlation with the measured pendulum value and structural depth,which can better characterize the anti-sliding performance of the asphalt pavement.The three-dimensional fractal dimension of the SMA pavement that meets the anti-slip performance requirements is given in combination with the current specification.The recommended value is 2.088<FD<2.813.