Four-way Value Space Positioning System Based On Structured Light

As an important component of heat pump air-conditioning equipment,fourway valve has been using artificial way of welding assembly.With the popularization and utilization of air conditioners in recent years,the problems of labor intensity and low production efficiency of traditional manufacturing industries gradually appear and it is difficult to meet the rising consumer demand each year.Although some enterprises also adopt the four-way valve welding equipment produced abroad,there are still shortcomings such as harsh working environment,expensive cost and poor welding quality.In order to realize automatic brazing assembly of four-way valve and improve the welding efficiency and quality,a space positioning method for four-way valve based on structured light is proposed.The surface features of the four-way valve are acquired by computer vision technology and structured light.The coordinate information of the key feature points of the workpiece is obtained based on the structured light three-dimensional measurement principle to assist the welding equipment in positioning.Compared with the machine vision method,the structured light method can avoid the error that the position of the workpiece to be measured is not coplanar in the two-dimensional vision measurement process and improve the measurement accuracy.The experimental results show that for the four-way valve workpiece,the measurement accuracy of the line structure light method is better than the surface structure light,but the surface structure light method has higher measurement efficiency.Structural light measurement error within 0.5mm,which satisfies the actual requirements.The specific research content of this paper is as follows:According to the actual processing environment of the four-way valve and the measuring principle of the visual system,a testing system experimental platform based on linear structured light is designed and constructed.By studying the projection model of the camera and the linear light measurement mathematical model,positioning principle,to develop linear optical system measurement programs.After,researched on the basic principles of Zhang zhengyou calibration method and apply to camera calibrate.In the RGB and HSV color space,the characteristics of the linear light stripe image are studied.The rough position of the fringe pattern is located by threshold segmentation,gray projection and template matching.Analyze and compare the extraction algorithm of stripe center on the surface of the workpiece,choose the gradient center of gravity method to extract at the sub-pixel level,and obtain the exact location of the center.In order to improve the measurement speed,the measurement system based on the surface structure light is studied and realized.By studying the measurement model of the surface light,the stripe template based on the three primary colors is designed.Preprocessing the images by white balance,color purification and region enhancement to enhance the fringe image features,then separating the three-color fringes and using the fast thinning algorithm to refine the stripe center one by one.Compared with the linear light system,the surface light method only needs to capture a single image to obtain the image information of all the characteristic locations,which greatly reduces the calculation amount and further improves the measurement efficiency.Study the camera nonlinear model,analyze the causes and principles of distortion.The sample data was collected and fitted to the light plane by the least squares method.The internal and external parameters and the light plane coefficients were obtained.According to the image data and the calibration parameters of the experiment,analysis and comparison of experimental results.