Glove defect detection using image processing / Farid Zuhri Kamal

Glove defect detection using image processing is a convenient method to identify failure in glove production industry. This project is designed to identify the defective gloves in the manufacturing line, to help reduce human failure. The glove defect detection can detect three cases which are normal, torn and empty link based on the region of interest (ROI) and the area. The methods used are blob and morphology algorithm to convert the original image to binary image and eliminate noise. A bounding box is obtained to calculate the area of pixel square, in which the resulting area of the normal glove is greater than torn glove. This method is capable of improving and helping the glove industry to enhance their product quality and grow their business.