Research On Vision Measurement Technology Of Laser De-icing System

In recent years,winter ice-snow damage comes up frequently in our country.The breakdown accident of power transmission lines and bridge cables caused by ice and snow accretion is also increasing,which has brought serious negative impact on human production and life.With the development and industrialization of laser technology,it can be used in the field of de-icing and anti-icing,which can provide technical support for reducing ice and snow disaster.As a novel and non-contact de-icing technology,laser de-icing has great application potential.In order to make laser deicing technology to be applied,it is necessary to solve the problems of icing identification,measurement and monitoring in the process of remote laser de-icing.This details focuses on the vision measurement and monitoring system of laser de-icing.The details are as follows:(1)A machine vision measurement system based on a zoom surveillance camera was designed.Aiming at the application requirements of laser de-icing,and combining the advantages of PTZ(Pan/Tilt/Zoom)camera in the field of intelligent monitoring,the hardware structure of laser de-icing vision measurement and monitoring system was designed.In terms of software implementation,the user interface was constructed with MFC as the main framework,and the vision processing algorithm was designed by integrating the OpenCV vision library and the Hikvision camera SDK development environment.(2)The icing identification algorithm based on image processing technology was studied.Firstly,the localization of the icing image algorithm and median filtering algorithm with a template size of 5×5 were carried out to enhance the features of interest in the icing image.Then an ice target recognition algorithm was designed based on HSI(Hue/Saturation/Intensity)color space.The difference of H-S component was selected as the first feature,the mean and variance of H,S and I components were selected as the second feature,which can effectively segment the foreground and background regions in the icing image and realize the icing identification.Finally,an icing binary image outline extraction algorithm based on the Canny operator was designed to obtain uniform and continuous edge information of ice target for icing binary image,which lays a foundation for subsequent vision measurements.(3)Based on PTZ camera and image processing technology,a laser triangulation ranging method with adjustable range was designed.Firstly,the basic principle of laser triangulation rangefinder was analyzed.Aiming at the limitation of the blind area and maximum distance in the direct laser triangulation method,a laser triangulation ranging model with an adjustable measurement range was designed.Then,the secondary development of the PTZ camera and the image processing algorithm of laser ranging was studied.The spot detection is realized by the color threshold segmentation algorithm and roundness detection algorithm,and the round fitting algorithm was used to accurately locate the center of the spot.Finally,the laser ranging experiment was carried out,under the condition that the zoom value of the camera is zoom =1 times,the target under test within30 m range was measured.The results show that the measurement error is less than 5 mm within 5 m,no more than 10 mm within 11 m,and no more than 54 mm within 28 m distance,and the system measurement error increases with the increase of distance.By comparing and analyzing the measurement results under the camera zoom of 1/3/5 times,it is found that the resolution of the system under the same zoom value decreases with the increase of distance,and the measurement error decreases with the increase of the zoom value.(4)Combined with the function of laser triangulation ranging,the laser remote de-icing target size measurement scheme based on zoom PTZ camera was designed and implemented.Firstly,the causes of the collected image distortion and its correction methods were studied.Then,the modified model of pixel equivalent of zoom camera was deduced,and a new method for calculating the size of ice target using the pixel equivalent of PTZ camera was proposed.In order to meet the real-time requirements of the system,a cache model based on image queue was designed to realize the cooperative work of image acquisition and processing.Finally,through the outdoor icing simulation experiments,three local ice regions with 11 cm and 28.5 cm ice edge lengths and 13.5 cm ice edge width were selected for fixed-point measurement.By comparing and analyzing the measurement data under different viewing distances(10?18 m)and different camera zoom values(1?18 times),the results show that the measurement error increases with the increase of distance and decreases with the increase of magnification values,and the measurement error of the system was within 2 mm.In this details,according to the laser de-icing vision measurement requirements of large outdoor targets,the machine vision monitoring function of the icing state,distance and target size was designed and realized,which is helpful to promote the laser remote de-icing technology towards practical application as soon as possible.