Intelligent Identification of Road Layers and Elimination of Interference Sources

Abstract

The patent of the present invention provides a method for intelligent identification of road layers and elimination of interference sources. The steps are as follows: 1. Data preprocessing, including background denoising, zero line setting, gain and sliding average for the original data of ground penetrating radar; 2. Identify and eliminate interference sources, establish an interference source sample library, and realize automatic identification and effective elimination of interference sources through machine learning; 3. Horizontal tracking and information extraction, centering on the reference track seed point, establish a three-level window operator , track and judge sequentially until the obtained layer coordinate information meets the requirements. The core of this method is the "three-level window operator" layer tracking algorithm, which can realize the precise positioning and identification of the tracked road layer.

Description

technical field [0001] The invention belongs to the technical field of ground penetrating radar detection applications, and relates to a method for intelligent identification of road layers and elimination of interference sources. The method includes data preprocessing, identification and elimination of interference sources, layer tracking and information extraction. The core is the "three-level window operator" layer tracking algorithm, which can solve the problems of low recognition accuracy, inaccurate positioning, and "cross-layer" existing in the existing road layer recognition and tracking algorithms. Background technique [0002] With the rapid development of my country's economy, the highway mileage is also increasing. By the end of 2018, the country's existing highway mileage is 4.7735 million kilometers. The quality inspection of new roads, the aging of existing roads, and road surface damage caused by natural or human factors all require high-precision detection o...

AI Technical Summary

Problems solved by technology

At present, the methods that can be used for road layer thickness detection mainly include artificial destructive detection methods (coring method and drilling method, etc.), geophysical nondestructive detection methods, etc. The destructive detection method has high accuracy but low detection efficiency, ground penetrating radar Although the non-destructive detection method has the advantages of high detection efficiency, the existing road layer recognition and tracking algorithms still have low recognition accuracy, inaccurate positioning, and "cross-layer" phenomena.

Images