Bridge crack detection data set pre-screening and calibration method

Abstract

The invention discloses a bridge crack detection data set pre-screening and calibration method, and aims to reduce the workload of a complicated process of manually screening data of tens of thousands of images and labeling images meeting requirements, and effectively reduce the subjective factor influence of a screener in the screening process. Firstly, a bridge image is acquired, edge detection and threshold segmentation based on a traditional algorithm are performed on the acquired bridge image, contour features are extracted, and a non-crack image is preliminarily judged and abandoned by formulating a screening rule. Thirdly, fine screening work is further carried out manually, a coarse screening result is corrected, and finally a bridge crack detection data set is obtained. According to the method, manpower and calculation time can be saved, and the detection cost for solving corresponding crack detection and other building problems is reduced.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and in particular relates to a bridge crack detection method. Background technique [0002] As of the end of 2018, my country has built more than 90,000 long-span bridges, more than 5,000 super-long-span bridges, and more than 100 super-long-span bridges with a main span of more than 400 meters. Most of the bridges are located in the traffic arteries. Once a collapse accident occurs, it will not only cause huge economic losses, but also cause casualties of innocent people. Therefore, the safety of bridges is related to the national economy and the people's livelihood. In recent years, due to the rapid growth of traffic flow, it has caused great pressure on the operation safety of bridges. In addition, due to the long-term construction of bridges, poor design performance, and harsh natural environments, bridge collapses have occurred frequently in recent years, causing great losses. Exi...

AI Technical Summary

Problems solved by technology

The invention can save manpower and calculation time, and reduce the detection cost of solving construction problems such as corresponding crack detection

Images