Research and Application of Online Quality Detection System Based on 3D Vision in Rectangular Steel Production Line

In order to solve the disadvantages of low efficiency, high labor costs, low precision and low intelligence level, the on-line quality detection system based on three-dimensional vision is designed for the rectangular cold-formed hollow steel. The system applies the machine vision technology, the image processing technology and the automatic control technology. It uses the high-performance PLC controller that can control to acquire the images of four smart sensors and achieve the communication between software and hardware of system. The image processing software is designed to complete the global calibration of four smart sensors thus to realize the three-dimensional reconstruction of point cloud images. Then the measured data of rectangular steel can be obtained by processing the point cloud image of three-dimensional reconstruction and be real-time displayed. Finally, the entire system can be achieved to automated on-line detection. The practical application shows that the system has the advantages of simple operation, short detected time, high precision and stability. It can be applied to detect any steel types of 180*180mm~ 500*500mm in this production line, and improve the intellectualization of the detected process. Meanwhile, It can greatly reducing labor costs and ensuring the production quality.