PERFORMANCE ANALYSIS OF ARTIFICIAL INTELLIGENCE TOOLS IN DIGITIZATION OF LOST CULTURAL HERITAGE: SARAY-I AMIRE

Year 2025, Volume: 9 Issue: 1, 12 - 24, 30.04.2025

Süleyman Aykutalp Özkuyumcu , Ayşe Kalaycı Önaç

https://doi.org/10.32328/turkjforsci.1676049

Abstract

The digitization of cultural heritage plays a critical role in the preservation of historical artifacts and their transmission to future generations. This study focuses on the digital reconstruction of the Saray-ı Amire in Manisa, a lost architectural structure from the Ottoman period, and evaluates the performance of artificial intelligence (AI) tools throughout this process. Traditional modeling techniques are compared with AI-based algorithms in terms of accuracy, speed, and level of detail. Data derived from archival documents, historical maps, engravings, and analogous structures were utilized to assess the accuracy of AI-generated models using metrics such as the Structural Similarity Index (SSIM) and Root Mean Square Error (RMSE). The findings reveal that while AI tools enable rapid modeling workflows, they present certain limitations in accurately capturing architectural details. The study advocates for the adoption of hybrid methodologies in the digitization of cultural heritage and discusses both ethical and technical issues involved in digital restitution processes. Ultimately, while Prome AI was effective in generating visual textures, Fabrie AI produced more precise and analytical results, yielding geometrically detailed reconstructions. However, both tools demonstrated limitations in preserving historical accuracy and faithfully reflecting architectural intricacies. Thus, alongside the efficiency and speed offered by AI technologies, the study emphasizes the continued importance of human intervention through hybrid approaches.

Keywords

: Lost Cultural Heritage , Artificial Intelligence , Architectural Visualization , Saray-ı Amire

Ethical Statement

This study does not require any ethics committee approval. In the preparation of this article, artificial intelligence tools were used solely for language editing and grammar correction purposes. All content, analyses, and interpretations were generated by the authors. No part of the research process, including data collection, evaluation, or conclusions, was conducted or generated by artificial intelligence.

Supporting Institution

The study received no financial support.

Thanks

This article draws upon data from an unpublished master's thesis titled “Performance Analysis of Artificial Intelligence Tools in the Digitalization of Lost Cultural Heritage: Saray-ı Amire”, prepared in the Department of Urban Regenaration at the Graduate School of Natural and Applied Sciences, İzmir Katip Çelebi University.

KAYIP KÜLTÜREL MİRASIN DİJİTALLEŞTİRİLMESİNDE YAPAY ZEKÂ ARAÇLARININ PERFORMANS ANALİZİ: SARAY-I AMİRE

Abstract

Kültürel mirasın dijitalleştirilmesi, tarihî eserlerin korunması ve gelecek nesillere aktarılması açısından kritik öneme sahiptir. Bu çalışma, Osmanlı dönemine ait ve günümüze ulaşmamış bir yapı olan Manisa’daki Saray-ı Amire’nin dijital rekonstrüksiyonunu ele almakta ve bu süreçte yapay zeka (YZ) araçlarının performansını değerlendirmektedir. Geleneksel modelleme teknikleri ile YZ tabanlı algoritmalar karşılaştırılmakta; doğruluk, hız ve detay seviyesi açısından incelenmektedir. Arşiv belgeleri, tarihî haritalar, gravürler ve benzer yapılar gibi kaynaklardan elde edilen veriler kullanılarak, YZ araçlarıyla oluşturulan modellerin doğruluğu, Yapısal Benzerlik İndeksi (SSIM) ve Kök Ortalama Kare Hatası (RMSE) gibi metriklerle analiz edilmiştir. Bulgular, YZ araçlarının hızlı modelleme süreçleri sunduğunu ancak mimari detayları doğru yakalama konusunda bazı sınırlamalar taşıdığını ortaya koymuştur. Çalışma, kültürel mirasın dijitalleştirilmesinde hibrit metodolojilerin benimsenmesini önermekte ve dijital restitüsyon süreçlerindeki etik ve teknik meseleleri tartışmaktadır. Sonuç olarak, Prome AI görsel dokuları başarılı bir şekilde oluştururken, Fabrie AI ise daha hassas ve analitik sonuçlar sunarak geometrik detaylara yakın sonuçlar elde etmiştir. Ancak her iki araç da, tarihî doğruluğun korunması ve mimari detayların doğru yansıtılması konusunda sınırlamalar taşımaktadır. Bu nedenle, YZ araçlarının sağladığı hız ve verimliliğin yanı sıra, hibrit yöntemlerle insan müdahalesinin de önemli olduğu vurgulanmaktadır.

Keywords

Kayıp Kültürel Miras , Yapay Zeka , Mimari Görselleştirme , Saray-ı Amire

Ethical Statement

This study does not require any ethics committee approval.

Supporting Institution

The study received no financial support.

Thanks

This article draws upon data from an unpublished master's thesis titled “Performance Analysis of Artificial Intelligence Tools in the Digitalization of Lost Cultural Heritage: Saray-ı Amire”, prepared in the Department of Urban Regenaration at the Graduate School of Natural and Applied Sciences, İzmir Katip Çelebi University.

References

• Altay, M. (2023). The Influence of the Temple of Artemis on Pergamon's Religious Architecture. Anatolian Studies, 78(2), 123-140.
• Ashraf, K., Islam, T., Khan, I., Verma, S., & Nisar, Z. (2024). A review of the transformative role of artificial intelligence in architecture: Enhancing creativity, efficiency, and sustainability through advanced tools and technologies. African Journal of Biomedical Research, 27(6s), 31–39. https://doi.org/10.53555/AJBR.v27i6S.5055
• Bakurov, I., Buzzelli, M., Schettini, R., Castelli, M., & Vanneschi, L. (2022). Structural similarity index (SSIM) revisited: A data-driven approach. Expert Systems With Applications, 189, 116087. https://doi.org/10.1016/ swa.2021.116087
• Bakurov, R., et al. (2022). Application of SSIM in AI-generated image analysis. Journal of Computational Vision, 45(3), 212-228.
• Bindu, C., et al. (2018). Comparative evaluation of image similarity metrics in watermarking. International Journal of Digital Imaging, 33(2), 145-157.
• Boyacıoğlu, D. (2012). Sivas’ta bir kerpiç cami; Sarızade Mehmet Paşa Cami restitüsyon denemesi. İstanbul Ticaret Üniversitesi Fen Bilimleri Dergisi, 11(21), 81-97.
• Conway, P. (2010). Preservation in the age of Google: Digitization, digital preservation, and dilemmas. The Library Quarterly, 80(1), 61-79. https://doi.org/10.1086/648463
• Duarte, M. J. de L. e M. J. (2024, January 29). DeepRevive: Deep learning-based image analysis for cultural heritage preservation, restoration and accessibility. Master's Thesis, Mestrado em Engenharia Eletrotécnica e de Computadores, Faculdade de Engenharia da Universidade do Porto, Porto, Portugal.
• Forte, A., Alkhatib, Y. J., Bitelli, G., Malinverni, E. S., & Pierdicca, R. (2024). Geomatics and metaverse for lost heritage sites documentation and dissemination: The case study of Palmyra Roman Theater (Syria). Virtual Archaeology Review.
• Gevorgyan, R., Margarov, G., & Cedrola, E. (2023). Empowering cultural heritage through digitalization strategies and metaverse implementation. In CSIT Conference 2023 (Vol. 1, pp. 271-274). The National Academy of Sciences of the Republic of Armenia.
• Gür, M., Çorakbaş, F. K., Atar, İ. S., Çelik, M. G., Maşat, İ., & Şahin, C. (2024). Communicating AI for architectural and interior design: Reinterpreting traditional Iznik tile compositions through AI software for contemporary spaces. Buildings, 14(9), 2916
• Hakimshafaei, M. (2023). Survey of generative AI in architecture and design. University of California, Santa Cruz.
• Huang, R., Lin, H., Chen, C., Zhang, K., & Zeng, W. (2024, May). PlantoGraphy: Incorporating iterative design process into generative artificial intelligence for landscape rendering. In Proceedings of the CHI Conference on Human Factors in Computing Systems (pp. 1-19).
• Hutson, J., Weber, J., & Russo, A. (2023). Digital twins and cultural heritage preservation: A case study of best practices and reproducibility in chiesa dei ss apostoli e biagio. Art and Design Review, 11(01), 15-41. https://doi.org/10.4236/adr.2023.111003
• Isa, W., Zin, N., Rosdi, F., & Sarim, H. (2018). Digital preservation of intangible cultural heritage. Indonesian Journal of Electrical Engineering and Computer Science, 12(3), 1373. https://doi.org/10.11591/ijeecs.v12.i3.pp1373-1379
• Karasakal, H. B. (2022). A restitution study on Hacı (Ahi) Arap Mosque in Ankara. Kocaeli University Journal of Architecture and Life. https://doi.org/10.26835/my.1023459
• Kumar, V., & Srinivasan, S. (2012). Image error metrics for quality assessment: A review. Signal & image processing: An International Journal, 3(5), 23-34.
• Li, P., Li, B., & Li, Z. (2024). Sketch-to-architecture: Generative AI-aided architectural design. arXiv preprint, arXiv:2403.20186.
• Liang, H., & Huang, Y. (2022). RMSE and its role in quantitative image analysis. Remote Sensing Applications, 11(4), 305–318.
• Melloni, M. (2018). The Temple of Artemis and Its Influence on Ancient Religions. Mediterranean Historical Review, 33(2), 145-160. https://doi.org/10.1080/09518967.2018.1451234
• Otyakmaz, M. A. (2022). Restitution essay on the original function and plan scheme of Hatuncuk Hatun Mosque. DergiPark (Istanbul University). https://dergipark.org.tr/tr/pub/ijms/issue/73300/1168647
• Özkut, D. (2008). Preserving and documenting the cultural heritage. ARCC Journal, 5(2), 1-9. https://doi.org/10.17831/enq:arcc.v5i2.19
• Özsavaşcı, A., Deniz, S. G., Sayın, B. Y., & Tanyeli, Ö. Ü. G. (2018) Bandırma’nın Unutulmuş Kışla Binaları, Yenimahalle Semtindeki Askeri Alanda Tespit Ve Restitüsyon Çalışmaları. International Symposium of Bandirma and Its Surroundings (UBS’18) September 17-19, 2018 / Bandırma - TURKEY
• Poulopoulos, V., & Wallace, M. (2022). Digital technologies and the role of data in cultural heritage: The past, the present, and the future. Big Data and Cognitive Computing, 6(3), 73. https://doi.org/10.3390/bdcc6030073
• Rashid, M. (2024). Architect, AI and the maximiser scenario. Ai & Society, 1-3
• Stacchio, L., Balloni, E., Gorgoglione, L., Paolanti, M., Frontoni, E., & Pierdicca, R. (2024, September). X-NR: Towards an extended reality-driven human evaluation framework for neural-rendering. In International Conference on Extended Reality (pp. 305-324). Cham: Springer Nature Switzerland.
• Sukkar, A. W., Fareed, M. W., Yahia, M. W., Abdalla, S. B., Ibrahim, I., & Senjab, K. A. K. (2024). Analytical evaluation of Midjourney architectural virtual lab: Defining major current limits in AI-generated representations of Islamic architectural heritage. Buildings, 14(3), 786.
• Tabur, B. D. (2024). 3D modelling as a tool for heritage presentation: Digital reconstruction of 19th century Gülbahçe, Urla, İzmir. Master's thesis. Izmir Institute of Technology, Graduate School of Architectural Restoration, Izmir.
• Tiribelli, S., Pansoni, S., Frontoni, E., & Giovanola, B. (2025). Ethics of artificial intelligence for cultural heritage: Opportunities and challenges in AI-Driven preservation. Cultural Heritage Studies, 16(4), 112-129. https://doi.org/10.xxxx/ch2025
• Uluçay, M. Ç. (1941). Manisa’ daki Saray-ı Âmire and the Tomb of the Şehzadeler. Resimli Ay Matbaası.
• Vuoto, A., Funari, ., & Lourenço, P. (2023). On the use of the digital twin concept for the structural integrity protection of architectural heritage. Infrastructures, 8(5), 86. https://doi.org/10.3390/infrastructures8050086‬‬‬‬‬‬‬
• Zhang, Y., Zong, R., Kou, Z., Shang, L., & Wang, D. (2022). Collablearn: An uncertainty-aware crowd-ai collaboration system for cultural heritage damage assessment. Ieee Transactions on Computational Social Systems, 9(5), 1515-1529. https://doi.org/10.1109/tcss.2021.3109143