Data augmentation based on multiscale radon transform for seven segment display recognition

Simple item page
dc.contributor.authorPopayorm, Sorawee
dc.contributor.authorTitijaroonroj, Taravichet
dc.contributor.authorPhoka, Thanathorn
dc.contributor.authorMassagram, Wansuree
dc.date.accessioned2026-08-06T10:27:25Z
dc.date.available2026-08-06T10:27:25Z
dc.date.issued2020-01-01
dc.description.abstractTo alleviate the problem of limited data in creating rotation, scale, perspective, and illumination invariant of the neural network training sets, the multiscale Radon transform is proposed in this study to enhance the data augmentation for seven segment display recognition. Resizing, smoothing, and coefficient shifting generate the desired invariant effects for the training model. The accuracy rates from the experiment demonstrate the superiority of the proposed method over other data augmentation techniques with the best overall accuracy performance of 87.05%-outperforming other data augmentation techniques by 6-13%. The convolutional neural network model generated from the proposed multiscale Radon transform data augmentation is suitable for seven segment display recognition and could become beneficial to other type of self-luminous type of images.
dc.identifier.citationKst 2020 2020 12th International Conference on Knowledge and Smart Technology, 47-51, 2020
dc.identifier.doi10.1109/KST48564.2020.9059315
dc.identifier.other2-s2.0-85084091118
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/10633
dc.sourceKst 2020 2020 12th International Conference on Knowledge and Smart Technology
dc.subjectConvolutional neural network
dc.subjectData augmentation
dc.subjectMultiscale Radon transform
dc.subjectSeven segment display recognition
dc.titleData augmentation based on multiscale radon transform for seven segment display recognition
dc.typeConference Paper

Files

Collections

Research