Computer vision algorithms for measurement and inspection of external screw threads
| Autor: | E.S. Gadelmawla |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0209 industrial biotechnology
Engineering business.industry Machine vision Applied Mathematics Single measurement Image processing 02 engineering and technology Thread (computing) Condensed Matter Physics 01 natural sciences 010309 optics Screw thread 020901 industrial engineering & automation 0103 physical sciences Maximum difference Computer vision algorithms Computer vision Artificial intelligence Metric system Electrical and Electronic Engineering business Instrumentation |
| Zdroj: | Measurement. 100:36-49 |
| ISSN: | 0263-2241 |
| DOI: | 10.1016/j.measurement.2016.12.034 |
| Popis: | The current methods of measuring screw threads are either time consuming or expensive. In addition, no single measurement method is available and capable of accurately measuring all screw thread features while significantly reducing the measurement time. This paper introduces a vision system for automatic measurement and inspection of most types of screw threads. Many image processing and computer vision algorithms have been developed to analyze the captured images of screw threads and perform the measurement and inspection processes. Most of the common screw thread features (18 features) could be measured by the introduced vision system. The system has been calibrated for both imperial and metric units and was verified by measuring a standard ISO metric thread plug gage and comparing the results of the measurements with the standard values. The results showed that the maximum difference between the standard and measured values was ±5.4 μm, which provide a good accuracy for measurement. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect