Real-time Measurement System Of The Heavy Truck Platform Deformation

A measurement system of heavy duty truck frame deformation which is composed of a LED target,a lens and a CMOS camera is designed to be used for the deformation detection of missile launch vehicle launch frame.In the process of missile launch vehicle transportation to the launch pad,the vehicle body will produce certain small deformation under the influence of the load and long distance,as well as change the balance and relative position of the vehicle which finally affect the performance of the heavy vehicle platform.In this paper,we designed a monitoring system based on the principle of visual measurement of heavy vehicle platform deformation to measure the deformation of heavy vehicle platform in real time.The imaging system is built on the vehicle body,so the change of the body deformation in the vertical plane is obtained by monitoring the change of the imaging position of the LED target on the CMOS,the measured results are output to the host computer system in real time.The experimental results show that the system,with the advantages of high precision and good real-time performance,can accurately measure the variation of the target point with respect to the image system.In order to guarantee the precision of system,in this paper a calibration system which is consist of console,stepper motor and target is designed to calibrate the measurement system.Besides,a new algorithm is proposed in this paper to solve the problem of multi-objective light spot problem with relative position information in complex background.Considering the relationship between multi-target spot images,the algorithm of "first division and ending" is proposed.The algorithm uses the filtered image as input information,locate all the suspect spots and store the spot position information in the linked list,And then output the four spot center coordinates information according to the relative position of the spot information,the remaining data of the linked list is the noise signal that is misjudged as a spot.The experimental results show that the algorithm can effectively extract the position of the four center coordinates of the coordinates in the complex background.Theoretical analysis shows that the algorithm can be extended to all multi-target spot center extraction with fixed position information.