الفهرس | Only 14 pages are availabe for public view |
Abstract There are many automated inspection and measurement systems available in industries: such as automatic gauging and sorting and computer supported coordinate measuring systems. On the other hand, humans obtain the vast majority of their visual system to improve the view of the target. Therefore in production engineering, computer vision system is one of the important visual systems, which could be developed for measurement and inspection to join another automated inspection and measurement systems, with the power of: 1) Non – contact system. 2) On – line process system. 3) Avoiding human error. 4) Ability for development. This work introduced in this thesis depends upon computer vision system as a tool, image processing as a technique and visual C++ as a programming language to have a prototype of visual system, which is capable of measuring and inspecting some geometrical tolerances, typically straightness, flatness, roundness, angularity, perpendicularity, parallelism and concentricity. The vision system is used to capture an image for the object to be measured or inspected and then the captured image is analyzed by the image processing and least square regression techniques to the geometrical tolerances. A calibration process has been done to calibrate the system for both measurement and inspection using standard block gauges. The system was verified and compared with standard specimens and commercial instrument and the results showed a good accuracy for the introduced system. Key Words: Geometrical Tolerances - Computer vision - Image processing - Calibration. |