Non-contact contact network parameter detection method based on three-dimensional point cloud data

Abstract

The invention discloses a non-contact contact network parameter detection method based on three-dimensional point cloud data. The method comprises the following steps of (1) collecting three-dimensional point cloud data of a contact network, (2) carrying out pre-processing on the point cloud data collected in step (1) and sequentially removing noise point and background point interference in pointcloud, (3) detecting and extracting contact line point cloud data, (4) converting a coordinate system used in collecting the contact network three-dimensional point cloud data into a world coordinatesystem, and (5) calculating the parameters of a contact line and completing the detection of the parameters. According to the method, a detection result of the invention is accurate and reliable, thedetection efficiency is high, the detection is not interfered by weather, a background, object surface reflection and the like, and the detection effect is good.

Description

technical field [0001] The invention relates to the field of high-speed railway catenary detection, in particular to a non-contact catenary parameter detection method based on three-dimensional point cloud data. Background technique [0002] At present, the railway transportation is extremely busy, and the train speed is increasing day by day; in order to ensure good contact between the pantograph and the contact wire and reliable current collection, prolong the service life of the pantograph, and prevent accidents such as "scraping the bow" or "drilling the bow" , it is necessary to regularly detect the guide height and pull-out value of the catenary to eliminate hidden dangers and keep the catenary suspension in good working condition; at present, the detection of catenary parameters is mainly divided into two types: contact type and non-contact type; Compared with contact image detection, non-contact image detection has been more and more widely used in actual detection d...

AI Technical Summary

Problems solved by technology

[0003] At present, there have been some researches on the detection of catenary geometric parameters based on non-contact image processing at home and abroad; for example, Zhan Dong has studied a visual sensor combined with a dual-line array camera and a line-structured light source. Based on the binocular line-array active camera measurement technology, A high-speed dynamic measurement method for catenary geometric parameters; Liu Wenqiang used the genetic particle filter algorithm and the Kalman filter to realize the tracking and positioning of the detection points and the detection value correction of the catenary combination parameters; the above two-dimensional images of opportunities , detect catenary parameters by extracting and locating targets in the image, but this type of detection method based on two-dimensional images is susceptible to interference from weather, two-dimensional image exposure, object surface reflection, etc., affecting the final detection effect ;Due to the depth information, 3D point cloud data can better overcome the above problems; however, the continuous characteristics of 3D point cloud data can also avoid the deviation of local point coordinates caused by external disturbances such as vibration, which will affect the final calculation results

Images