Tunnel segment circular seam linearization identification and mileage ring number marking method

Abstract

The invention relates to a tunnel segment circular seam linearization identification and mileage ring number marking method. The method is characterized in that: a group of tunnel images with pattern numbers are input into a trained linearization identification algorithm model for identification, and a linearization mark of a circular seam is output; circular seam mark data is obtained based on the linear mark, wherein the circular seam mark data comprises an image number and relative position information of the circular seam in the corresponding tunnel image; the group of tunnel images are spliced into a whole tunnel image according to pattern numbers; comparison calculation is performed on the circular seam mark data of the corresponding tunnel image and the starting and ending mileage data and the starting ring number data of the whole tunnel image to obtain ring starting and ending mileage data and circular seam mileage data relative to the whole tunnel image; the whole tunnel image, the mileage and the ring number are displayed; and the linearization recognition algorithm model is obtained after a training sample set is input into a convolutional neural network for recognition training, wherein the training sample set comprises an original tunnel image and a linearization mark corresponding to the original tunnel image.

Description

technical field [0001] The invention relates to a method for linear recognition of ring seams of tunnel segments and marking methods for mileage ring numbers. Applicable to the field of tunnel engineering. Background technique [0002] With the development of the urban economy, subway construction has entered a peak period, and more and more subway tunnels have entered the stage of operation and maintenance. Traditional maintenance methods are inefficient through manual investigation and records, and cannot intuitively understand the location of the disease and the overall picture of the tunnel. Therefore, the introduction of mobile 3D laser scanning technology is applied to the safety monitoring of subway tunnels and the establishment of paperless digital files for the entire life cycle of tunnels. [0003] One of the important tasks of the digital archives is to accurately number each segment in the panoramic orthophoto image and calculate its mileage. The premise is to d...

AI Technical Summary

Problems solved by technology

[0005] The tunnel ring seam recognition method disclosed in the above literature is mainly based on the original image processing and threshold segmentation, which has high requirements on the clarity and gray scale range of the original image, and the generalization ability is relatively low. The recognition accuracy is not high, and many recognition errors and omissions need to be manually corrected after the recognition mark, which has limited improvement in work efficiency

Images