Publication:
Malaysian Coins Recognition Using Machine Learning Methods

dc.citedby1
dc.contributor.authorZainuddin N.N.en_US
dc.contributor.authorAzhari M.S.N.B.N.en_US
dc.contributor.authorHashim W.en_US
dc.contributor.authorAlkahtani A.A.en_US
dc.contributor.authorMustafa A.S.en_US
dc.contributor.authorAlkawsi G.en_US
dc.contributor.authorNoman F.en_US
dc.contributor.authorid57337300700en_US
dc.contributor.authorid57337300800en_US
dc.contributor.authorid11440260100en_US
dc.contributor.authorid55646765500en_US
dc.contributor.authorid57218103026en_US
dc.contributor.authorid57191982354en_US
dc.contributor.authorid55327881300en_US
dc.date.accessioned2023-05-29T09:05:58Z
dc.date.available2023-05-29T09:05:58Z
dc.date.issued2021
dc.descriptionAdaptive boosting; Discriminant analysis; Machine learning; Nearest neighbor search; Neural networks; Classifieds; Coin recognition; Complex Processes; Daily lives; Images processing; Machine learning approaches; Machine learning methods; Malaysians; Object categories; Training model; Image processingen_US
dc.description.abstractCoins have become essential to our daily lives. As a legal tender, identifying, validating, classifying, and sorting coins is a complex process. In essence, automated coin detectors should identify deteriorated or older coins and distinguish between genuine and fake coins. Nonetheless, we identified limited literature examining different Machine Learning (ML) approaches for detecting Malaysian coins. This study investigates machine learning approaches and identifies the most efficient and accurate for Malaysian coin recognition. The model was trained on 311 images of coins and classified into four object categories: 5, 10, 20, and 50 cents. Six classifiers are used to test the training model. For the Grey-Level Co-occurrence Matrix (GLCM) feature extraction, AdaBoost classifiers were the most accurate, whereas K-Nearest Neighbors (KNN) classifiers were the least accurate. Moreover, the Artificial Neural Networks (ANN) classifier had the highest accuracy in the Histogram of Oriented Gradients (HOG) feature, while the Linear Discriminant Analysis (LDA) classifier had the lowest. The study findings and future directions are discussed. � 2021 IEEE.en_US
dc.description.natureFinalen_US
dc.identifier.doi10.1109/AiDAS53897.2021.9574175
dc.identifier.scopus2-s2.0-85118967730
dc.identifier.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85118967730&doi=10.1109%2fAiDAS53897.2021.9574175&partnerID=40&md5=c4bc42d0dbcf31137e8b8dacaef88792
dc.identifier.urihttps://irepository.uniten.edu.my/handle/123456789/26000
dc.publisherInstitute of Electrical and Electronics Engineers Inc.en_US
dc.sourceScopus
dc.sourcetitle2021 2nd International Conference on Artificial Intelligence and Data Sciences, AiDAS 2021
dc.titleMalaysian Coins Recognition Using Machine Learning Methodsen_US
dc.typeConference Paperen_US
dspace.entity.typePublication
Files
Collections