Loading...

Automation of Vision Measurement Machine to Develop Parts Profile Dimensional Measurement Algorithm based on Machine Vision and Image Processing Technique Algorithms

Hosseini, Arian | 2022

91 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55899 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Khodaygan, Saeed
  7. Abstract:
  8. Automated dimensional inspection is commonly expensive because of the requirement for high-precision measurement devices. To perform a precision measurement, the technician must be highly skilled and fully understands the operation of the equipment. Moreover, automation of the mentioned process to reduce dimensional measurement time is a complicated task due to restrictions of precise equipment such as CMM. With the expansion of the use of cameras in the industry, the measurement method with the help of machine vision systems is one of the cost-effective methods that can be achieved with the development of a suitable image processing algorithm to achieve acceptable accuracy compared to the traditional methods in the dimensional measurement of manufactured parts. This project proposed a semi-intelligent automatic method, with the help of automation and a camera, to reduce the complex parts measuring time (generally for reverse engineering) and to minimize the human error factor. To accomplish this project, a manual vision measuring machine in the metrology laboratory was upgraded to the automatic version by installing stepper motors, position detection sensors, and an object detection camera. Then, by developing two algorithms, based on image processing, the measurement process will be done automatically. At first, the added camera detects the position of objects placed on the table. Secondly, the table movement path will be generated based on the detected contours of objects. Then, an industrial camera captures partial images of objects. Finally, through several image processing procedures and a calibration model, contours from partial images are combined into one global image. This process results in a point cloud consisting of every point from the 2D contour of objects, which can be directly used for automatic measurement with the 10 μm resolution. On the other hand, a GUI is developed for the Windows operating system to integrate upgraded VMM and developed algorithms. To verify the proposed method's accuracy, metric rectangle block gauge, angle block gauge, and Whitworth thread gauge were measured. Furthermore, to qualify the method for a real-world problem the 40A26 ANSI sprocket was measured and the data was compared to the standard, revealing a favorable correlation
  9. Keywords:
  10. Image Processing ; Automation ; Dimensional Measurement ; Vision Machine ; Two Dimentional Contour Reconstruction

 Digital Object List

 Bookmark

No TOC