Research On Adaptive Control Method Of Line Structured Light Sensor Parameters

The machine vision based on line structured light measurement system(LSLMS)has the advantages of simple structure,low cost,non-contact,high efficiency and so on.It is widely used in many fields,such as shape measurement,state monitoring,objects recognition and the like.When it measures surface with complex geometric texture or highlighted surface,there are some problems,such as large noise,difficult to ensure the measurement accuracy.Aiming at the above problems,the adaptive control of sensor exposure time is studied in order to obtain high-quality strip image,and then the measurement accuracy of workpiece with different surface characteristics is improved.In this paper,the adaptive control method of sensor parameter in LSLMS is studied.The main work and innovation points include: 1)According to the laser triangulation principle,the LSLMS is built.This system is made up of the control subsystem,vision measurement subsystem and workbench motion subsystem.The hardware of the system is selected and the software system is developed on the Visual Studio 2010 platform;2)The calibration of line structured light sensor based on the reference target is finished,the calculation method of laser plane equation is given,and the motion coordinate system is established.The coordinate transformation matrix from camera coordinate system to motion coordinate system is calculated,and then the transformation from pixel coordinate system to motion coordinate system is realized;3)A center extraction method based on the back propagation(BP)neural network is proposed.The basic principle of stripe center extraction using the neural network,the method that calculates the ideal center points for network training and the network weight tuning algorithm are presented successively,the factors that affect the center extraction are analyzed.The proposed method can meet the requirements of high precision and efficiency at the same time.It overcomes the problems of center point loss and redundant center point that the Steger method has under different exposure times;4)Within a certain exposure time based on the linear relationship between the exposure time and the section average width of laser stripe,an adaptive control method for the exposure time of line structured light sensor is proposed.The calculation method of laser stripe width based on the BP neural network is presented.This method can not only guarantee the measurement accuracy,but also improve the calculation efficiency,so as to ensure the real-time requirements of controlling.It can control the exposure time of the sensor adaptively when it measures surface with complex geometric texture or highlighted surface,so as to obtain high-quality laser stripe image and the average width of stripe section is controlled within the ideal width range.This method has faster adjustment speed,higher extraction accuracy of the center of the adjusted strip image,richer 3D measurement data of the measured workpiece surface.In this paper,a LSLMS is built based on the laser triangulation principle,on which the adaptive control for the exposure time of the line structured light sensor is realized.Experimental results show that the proposed method can control the exposure time of the sensor adaptively.When measuring surface with complex geometric texture or highlighted surface,it can still obtain high-quality laser stripe image,and realize the fast and accurate measurement of the measured objects surface.