Vehicle shape and position detection system and method based on contour scanning

Abstract

The invention relates to a vehicle form and position detection system and method based on contour scanning. The system comprises a vehicle number identification module, a speed measuring module, a laser contourgraph, a gantry, and a processing system. In addition, the method comprises the following steps that: a laser contourgraph performs contour scanning on a vehicle running on a track and calculates a 3D point cloud image through a laser image and a vehicle speed; vehicle information is matched in a database based on a vehicle number to obtain a standard 3D point cloud map; and after comparison of the point cloud images, the appearance size and external parts of the vehicle are detected. According to the invention, with a laser scanning vehicle, adaptability to the hostile environment is enhanced and the strong anti-interference ability is high. The provided method is a non-contact detection method not affecting the running of equipment. With the 3D point cloud calculation, all sizes can be measured and the sensor is convenient to mount. With the three-dimensional matching technology, the recognition accuracy is improved. The overall size of the vehicle and external parts of thevehicle can be detected simultaneously, so that the cost is saved; and the operation capability is optimized by the algorithm and the work efficiency is improved.

Description

technical field [0001] The invention relates to the field of rail vehicle detection, in particular to a vehicle shape and position detection system and method based on contour scanning. Background technique [0002] When the vehicle runs on the track for a long time, first, there may be protruding deformation of the car body or hanging of the roof cable, which makes the shape of the car body exceed the specified size, and the overrun part is easy to collide with the surrounding objects and connect; the second is Some general parts may become deformed, loose, broken or missing, causing some equipment not to work properly. Both of the above situations will bring hidden dangers to the operation of the vehicle. Therefore, after the vehicle has been running for a period of time, it is necessary to regularly carry out vehicle limit detection and conventional component detection, which provides a safety guarantee for the operation of the vehicle. [0003] For limit detection, the...

AI Technical Summary

Problems solved by technology

This detection method installs a detection baffle on the inside of the limit door. If the vehicle exceeds the limit, it will come into contact with the baffle, and the detection is realized by detecting whether there is a contact collision. This detection method has the following defects: (1) limited by the mechanical structure, One type of detection baffle corresponds to one type of vehicle. When measuring different types of vehicles, it is necessary to replace the baffle or build limit gates for different models, which increases the complexity of the detection process or increases the construction cost; (2) vehicles that exceed the limit will Scratching with the baffle, not only the baffle needs to be replaced after a period of time, but the replacement time is long, which affects the use time of the equipment, and may cause certain damage to the vehicle and increase other additional costs

Due to the inherent characteristics of 2D image analysis technology, vehicle shape detection based on 2D image analysis has the following defects: (1) Locomotives that have been used for a long time will have attachments such as rust, sand and dust, and it is difficult to obtain them even after cleaning. Effective photos make the matching recognition rate very low or unrecognizable during image analysis, and the measurement based on matching recognition cannot be carried out; (2) Measuring three-dimensional objects with 2D images has a high impact on the position and angle of the camera (3) The measurement technology implemented by 2D image analysis has requirements for the measurement direction (objects perpendicular to the lens direction become vertical projections in the picture, and the vertical length cannot be measured), It is impossible to take into account all directions through the combination of cameras; (4) Although the basic algorithm of 2D image analysis is relatively mature, the particularity of the vehicle shape detection environment, the complexity of vehicle shape and the diversity of vehicle types, the transformation of 3D vehicle shape into After the two-dimensional image, the practical results prove that it is difficult to form an effective and effective overall technical solution for the system

Images