Automated final visual inspection using image processing

• Main Authors: Lumain, Joshua Edward N., Navarrete, Miguelito G., Jr., Racpan, John Andino L.
• Format: text
• Language: English
• Published: Animo Repository 2014
• Subjects: Optical pattern recognition; Image Processing > Digital Techniques; Computer vision > Industrial applications; Communication; Electrical and Electronics
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/10556
• Institution: De La Salle University

The study is about creating an automated final visual inspection using image processing software. An algorithm in NI LABVIEW would be designed and created that would function as a replacement in the current manual visual inspection processes. 3 cameras are used in order to take the images of the whole products. The system analyzes products using edge detection and color matching. Line scores and match scores are obtained from ideal products, these score are then considered the basis of the comparison to new products being analyzed. The system then considers whether or not the product being analyzed. The system then considers whether or not the product passed inspection if they have achieved the minimum score established. Depending on the end decision the system sorts the products if they have passed or failed using a binning system. From the testing and analysis that the group had done, the group concludes that the system had done analysis on the objects examined. A sufficient accuracy rate had been achieved to consider the system more effective than manual visual inspection.