Classification model of optical character recognition failures in unrecovered slider serial numbers in hard disk drive manufacturing and image capture processes

Simple item page
dc.contributor.authorChousangsuntorn, Chousak
dc.contributor.authorTongloy, Teerawat
dc.contributor.authorChuwongin, Santhad
dc.contributor.authorBoonsang, Siridech
dc.date.accessioned2026-08-06T10:35:30Z
dc.date.available2026-08-06T10:35:30Z
dc.date.issued2022-01-01
dc.description.abstractIn hard disk drive (HDD) manufacturing processes, there are unrecovered serial number images about 0.01% from the standard optical character recognition (OCR) reading and deep learning approach. We found several failures from two main causes, i.e. manufacturing process and image capture process during standard OCR reading. We proposed classification model used for recognizing the serial number reading failures based on object detection You-Only-Look- Once (YOLO) algorithm and EfficientNet-B0 classification network as well as histogram analysis. The 1000 images captured by digital camera were used for training (600 images) and validation (400 images) the ROI detection model. The other 2100 captured images were used for training and testing classification OCR failure from manufacturing process model. The model testing was performed in 900 images contained 9 causes (classes) of failures. The proposed model reaches F1 score = 0.94.
dc.identifier.citationProceedings of SPIE the International Society for Optical Engineering, 12171, 2022
dc.identifier.doi10.1117/12.2631362
dc.identifier.issn0277786X
dc.identifier.other2-s2.0-85133008643
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/12821
dc.sourceProceedings of SPIE the International Society for Optical Engineering
dc.subjectClassification
dc.subjectEfficientNet
dc.subjectObject detection
dc.subjectOptical character recognition
dc.subjectYOLO
dc.titleClassification model of optical character recognition failures in unrecovered slider serial numbers in hard disk drive manufacturing and image capture processes
dc.typeConference Paper

Files

Collections

Research