Classification of filigree silver with Artificial Neural Networks according to production methods

Year 2024, Volume: 14 Issue: 1, 83 - 87, 30.06.2024

Sabahattin Akgül , Hamit Adin , Hüseyin Ahmetoğlu

https://doi.org/10.36222/ejt.1336397

Abstract

The jewelry industry uses precious stones and metals in various ways while ornaments and jewelry are made. One of the methods used is the filigree method. The most critical factor in the filigree method is human and craftsmanship. However, rapid technological developments make the machine use in filigree mandatory. As a result, filigree products produced by handwork can be created using serial molds in the factory environment. This study aims to classify the molded product filigree silver using artificial neural networks. Filigree products produced by filigree masters and as mold products were compared to distinguish the filigree products. The color of the silver jewelry, the state of the jewelry, the silver setting status, the brass metal used in the silver jewelry, the form of the inner filling motif, the shape of the roof wire, the smoothness of the structure, the proper placement of the inner filling, the symmetrical status of the motifs on the jewelry are trained in the system using Deep Learning, which is an artificial neural networks method through thehe data collected from features such as the use of valuable or worthless stones. The success of classifying filigree jewelry handcrafts or mold products using Deep Learning through artificial neural network methods was evaluated. As a result of the study, the classification with deep learning was conducted successfully.

Keywords

Artificial Neural Networks, Deep Learning, Filigree method, Jewelry

References

• [1] Türe A, Savaşçın MY. Birth of jewelery Goldaş publications 2000.
• [2] Öztemel E. Artificial neural networks Papatya publications April 2012.
• [3] Deng L, Yu D. Deep Learning: Methods and Applications, vol. 7. 2013.
• [4] LeCun Y, Bengio Y, Hinton G. Deep learning Nature İnternational journel of science pages 436–444 (28 May 2015)
• [5] Goodfellow I. “Chapter06 Deep Feedforward Networks,” Deep Learning Book, no. 1, pp. 169–229, 2015.
• [6] Buduma N, Locascio N. Fundamentals of Deep Learning, vol. 521. 2015.
• [7] Ahmetoğlu H, Daş R. Classification of Attack Types from Big Data Sets with Deep Learning 2019 International Artificial Intelligence and Data Processing Symposium (IDAP)