Truck carriage volume measurement method based on single image

Abstract

The invention discloses a truck carriage volume measurement method based on a single image. The defects that an existing photoelectric volume measurement system can only be applied to a detection channel with the fixed position, and multiple laser devices, cameras and sensors are needed are mainly overcome. The truck carriage volume measurement method includes the steps of 1, inputting the image of a truck, and calibrating an intrinsic parameter matrix of a camera through end points in the image of the truck in three orthogonal directions; 2, calculating normal vectors of wheels and coordinates of circle centers of the wheels in cooperation with the intrinsic parameter matrix and an elliptic equation of the wheels of the truck, and obtaining a carriage plane equation; 3, calculating extrinsic parameters of the camera through the end points and the coordinates of the circle centers of the wheels; 4, solving space coordinates of top points of a carriage in cooperation with the carriage plane equation and a camera matrix operation equation, and the solving the volume of the carriage. By means of the truck carriage volume measurement method, the volume of the carriage of the truck can be rapidly and accurately measured, the automation degree is high, and the truck carriage volume measurement method can be used for a charging system of an expressway toll-gate.

Description

technical field [0001] The invention belongs to the field of image information processing, in particular to a method for measuring the volume of a truck compartment, which can be used in a billing system of a highway toll station. Background technique [0002] For trucks, the billing system of expressway toll stations needs to measure the volume of different truck compartments separately, and different fees correspond to different compartment volumes. The traditional truck volume measurement mostly adopts photoelectric volume measurement system, which needs to use lasers, cameras, sensors and other equipment at the same time. Among them, the common method is that the lasers on the top and both sides of the detection channel emit strip laser perpendicular to the detection channel, and irradiate the surface of the truck to form a The optical profile image is used to extract the three-dimensional profile data of the truck by using the laser image collected by the CCD camera tha...

AI Technical Summary

Problems solved by technology

Henry P and others can use the depth image to build a 3D model of the object more accurately. See Henry P, Krainin M, Herbst E, et al. Using Kinect-style depth cameras for dense 3D modeling of indoor environments [J]. The International Journal of Robotics Research, 2012, 31(5): 647-663, the depth image not only contains the color information but also contains the depth information of each pixel, but the cost is very high, the operation is inconvenient, and real-time measurement cannot be performed, which makes the application range of this method restricted

Kholgade et al. used virtual reality technology to transform the existing 3D models of similar objects to achieve 3D matching between the model and a single uncalibrated image. The 3D model of the object can be recovered from a single image. See Kholgade, Natasha, et al.3D object manipulation in a single photograph using stock 3D models[J].ACM Transaction on Graphics(TOG),2014,33(4):127, however, because the picture is not calibrated, the model has scale blur and can only be used in 3D Scene roaming, not suitable for object volume measurement

Considering that the truck wheel can provide a single spatial circle feature, the single-circle feature positioning based on monocular vision can provide the plane information of the wheel, which is of great significance to the calculation of the volume of the car. Zheng et al. conducted a theoretical derivation on the single-circle positioning problem. Zheng Yinqiang ,Ma Wenjuan,Liu Yuncai.Another way of looking at monocular circle pose estimation[C].Proceedings of International Conference on Image Processing,2008:86-92,Given the geometrical sense that the space circle pose has two solutions Complete explanation, but no solution to eliminate spurious solutions

Images