Pavement crack detection method, device and equipment of aerial photography image

Abstract

The invention provides a pavement crack detection method, device and equipment of an aerial photography image. The method comprises that a deep high-dimension feature of a pavement area is extracted from the pavement aerial photography image, and a high-dimension feature graph is obtained according to the deep high-dimension feature; on the basis of the deep high-dimension feature of the pavementarea, positive and negative samples of the high-dimension feature graph are screened to distinguish pavement crack targets from a pavement background; and the pavement crack targets are classified andpositioned according to coordinates to obtain classification information and coordinate information of the pavement crack targets. The method, device and equipment can be applied to pavement crack detection of the aerial photography image with high-altitude motion or complex scene, and compared with routine crack detection algorithms, the method is more suitable for image collection of an unmanned airborne system, and the crack detection robustness of the aerial photography image is higher.

Description

technical field [0001] Embodiments of the present invention relate to the fields of deep learning and pattern recognition, and in particular to a method, device and equipment for detecting road surface cracks in aerial images. Background technique [0002] At present, one of the main damage forms of highway pavement is pavement crack, among which, the crack types of highway in my country are mainly transverse and longitudinal cracks. If the cracks can be found in the early stage and track their development, the maintenance cost of the road surface will be greatly reduced, and the driving safety of the expressway can be guaranteed at the same time. Therefore, it is extremely important to conduct regular surveys and maintenance on the pavement condition of the highway. [0003] The pavement crack detection method has been developed from the initial manual detection method; with the application of image processing technology, the vehicle-mounted acquisition device and image pr...

AI Technical Summary

Problems solved by technology

However, these existing methods are basically based on the detection and development of the image of the vehicle acquisition device, and cannot be applied to aerial images with more interference and noise.

Images