Pengembangan Sistem Ai Untuk Generasi Pola Batik Modern Berbasis Progressive Growing Of Generative Adversarial Networks (Pggan) Dengan Integrasi Kearifan Lokal Jember

Safii, Restu Imam (2026) Pengembangan Sistem Ai Untuk Generasi Pola Batik Modern Berbasis Progressive Growing Of Generative Adversarial Networks (Pggan) Dengan Integrasi Kearifan Lokal Jember. Undergraduate thesis, Politeknik Negeri Jember.

	Text (Abstract) Abstract.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (110kB)
	Text (Bab 1 Pendahuluan) Bab 1 Pendahuluan.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (191kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (203kB)
	Text (Laporan Lengkap) Laporan Lengkap.pdf - Submitted Version Restricted to Registered users only Download (6MB) \| Request a copy

Abstract

Batik is an Indonesian cultural heritage with high aesthetic and philosophical value, including Jember batik, which is known for its Labako motif. The development of artificial intelligence technology provides opportunities for exploring batik motif design digitally, especially in generating new motif variations while maintaining local wisdom elements. This study aims to develop an artificial intelligence system based on Progressive Growing of Generative Adversarial Networks (pGGAN) to generate modern batik motifs by integrating Jember’s local wisdom, particularly tobacco leaf elements. The system was developed as a web-based application using the Spiral Model, which was carried out in three development iterations. The generative process is also supported by implicit trigonometric functions to form decorative geometric patterns. The results show that the system can generate new batik motifs with a final resolution of 256 × 256 pixels. Image quality evaluation shows that the FID score decreased from 312.45 to 87.34, while the IS score increased from 1.23 to 3.42. These results indicate an improvement in the visual quality and diversity of the generated images. White Box Testing achieved a success rate of 94.55%, with 104 passed test cases out of 110. Cultural expert validation shows that the generated motifs reflect Jember’s local wisdom through leaf elements and decorative patterns, although the tobacco leaf characteristics, color composition, and Labako motif identity still need further improvement

Item Type:

Thesis (Undergraduate)

Contributors: