In the production process of instant noodles, it is necessary to detect the varieties on the same production line and pack each variety separately. Traditional human detection requires intensive labor, but provides low efficiency. Therefore, we applied the machine vision technology to the production line to increase productivity while reducing false or missed detection rates.
The system uses HIS color space to extract the color feature value of the instant noodle bags, and makes classified marker for the pixels with different colors to determine the types of instant noodles. The software system modules include: image acquisition module, image processing module, learning module and other functional modules. The conveyor belt delivers instant noodles to the visual detection section. The optical sensor receives the signal for arrival of instant noodles and then trigger the industrial camera to capture and send instant noodles images to IPC. The image processing module determines the types of instant noodles, and issues corresponding signals for PLC to perform actions.
(ii). System Features
· Statistical pattern recognition
· Fast and precise detection
· Low missed and false detection rates