Three-dimensional pavement crack extraction method based on section local extremum and segmentation sparsity

Abstract

The embodiment of the invention provides a three-dimensional pavement crack extraction method based on section local extremum and segmentation sparsity. The method comprises the following steps of obtaining a maximum value point and a minimum value point according to an elevation difference value of two adjacent points in elevation data of a cross section in line scanning three-dimensional data; selecting an extreme value point to delete to obtain an extreme value pair, and deleting an abnormal extreme value point; selecting a plurality of points from the extreme value stable section as sparsepoints, deleting abnormal sparse points, and supplementing confidence points when the length of the missing section is too large; carrying out interpolation and filtering on confidence points in thesection stability confidence point set to obtain a reference contour; an accurate reference contour being obtained. According to the method, the three-dimensional pavement crack information is extracted more accurately, the point with lower elevation is selected as the suspected crack point according to the characteristic of smaller crack proportion, and the crack is obtained after processing, sofalse detection of the crack and missing detection of shallow cracks can be prevented, the influence of different pavement texture backgrounds on crack extraction can be overcome, and the three-dimensional pavement crack information can be extracted more accurately.

Description

technical field [0001] The invention relates to the technical field of line scanning three-dimensional data processing, in particular to a method for extracting three-dimensional pavement cracks based on section local extremum and segmental sparseness. Background technique [0002] With the development of line scanning 3D measurement technology, the 3D measurement system can acquire more and more 3D pavement data, including data of different pavement backgrounds and different types of cracks. [0003] However, in practical engineering applications, the 3D pavement data collected by the vehicle-mounted 3D system contains relatively obvious driving attitude fluctuations and road surface deformation information. These large-scale attitude deformation information bring difficulties to the crack detection task, and are not conducive to the 3D crack depth, Acquisition of three-dimensional information such as width and length. On the other hand, due to system noise and pavement te...

AI Technical Summary

Problems solved by technology

[0008] The embodiment of the present invention provides a three-dimensional pavement crack extraction method based on the local extremum of the section and segmental sparseness, which is used to solve the defects of inaccurate and poor robustness in the extraction of three-dimensional pavement cracks in the prior art, and realize more accurate extraction of three-dimensional pavement cracks. pavement cracks

Images