Method for collecting road inspection facility data based on AI image recognition technology

Abstract

The invention provides a method for collecting road inspection facility data based on an AI image recognition technology, and the method comprises the following steps: obtaining the type of a target collection object based on the relation between the target collection object in a high-precision map and a road type and a road region, and setting different collection plans according to different types of the target collection object; obtaining the spatial position and the spatial shape of the target acquisition object based on the unmanned aerial vehicle remote sensing image, and performing surface vision acquisition through AI acquisition equipment to obtain video stream data information; and performing cross validation on the acquired spatial position, spatial shape and video stream data information of the target acquisition object and the acquisition plan, and acquiring a target road inspection facility through an AI image recognition technology. The AI image recognition technology and the Beidou positioning technology are combined for real-time collection object recognition, recognition results are textualized and normalized, processed result data are small in size, dependence on the bandwidth of a mobile digital network is low, and the data can be automatically uploaded to a background server in real time.

Description

technical field [0001] The invention relates to the technical field of road inspection, in particular to a method for collecting road inspection facility data based on AI image recognition technology. Background technique [0002] The new generation of technological innovation represented by unmanned driving has greatly improved the quality and current situation requirements of spatial data represented by high-precision maps, and the requirements for the accuracy and update frequency of spatial data collection have been raised several levels. [0003] The collection of spatial data relies heavily on the scale effect, and it is necessary to systematically answer the cost question on how to provide a national new infrastructure map. At present, the data collection represented by the map field and the office still uses manual collection as the main method, and the large-scale dependence on manual collection has become a constraint for the spatial information industry. [0004]...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for collecting road inspection facility data based on AI image recognition technology, which can solve the problems of high cost, poor real-time performance and poor standardization in the existing road inspection mainly through manual training to identify and formulate targets. technical problem

Images