Author/Authors :
METLEK, Sedat Süleyman Demirel Üniversitesi - Mühendislik Fakültesi - Bilgisayar Mühendisliği Bölümü, Turkey , METLEK, Sedat Mehmet Akif Ersoy Üniversitesi - Burdur Meslek Yüksekokulu, Turkey , CETİŞLİ, Bayram Süleyman Demirel Üniversitesi - Mühendislik Fakültesi - Bilgisayar Mühendisliği Bölümü, Turkey
Title Of Article :
The Use of Cameras for Real-Time Industrial Object Recognition
Abstract :
In this study, it is shown that the system with a camera runs faster than a color sensor system in real time. For this purpose, interconnected set of four experiments produced by Festo manufactures have been treated as a hardware and software. By using CMOS camera, image processing methods and Bayesian classifier, the real time object recognition has been achieved more realistic and less time consuming. Classification procedures, located at the outlet of the experimental set, were done with PLCs (Programmable Logic Controller). This operation is controlled with computer using parallel port in the new system. An extra electronic driver circuit is designed for the control of electro-mechanical controlled pneumatic arms. A three time faster classification success is obtained by the camera that is put in the place of color sensor.
NaturalLanguageKeyword :
Industrial automation , real , time image processing , Bayesian classifier
JournalTitle :
Journal Of Natural and Applied Sciences
