深度學習演算法支援博物館藏品盤點之實驗：以國立臺灣歷史博物館為例

現代博物館主要的4項功能，包括典藏、研究、展示及教育，其中保存及管理文物藏品則是典藏的重點工作。文物藏品作為博物館展示及研究的核心物件，可謂是見證歷史及文化的最佳物證。對各博物館而言，擁有的藏品數量相當多，而作為保護文化資產的重要場域，其藏品庫房則是博物館建築的重心之一，除了部分被取用或借出的藏品之外，大部分藏品都保存在庫房。其中，藏品管理中的文物盤點始終是一項既繁瑣又浩大的工程，因此適當的導入智慧科技技術，作為輔助人力管理，提升典藏管理、藏品管理的效率及品質，以達智慧管理（庫房智慧化）之目標。本研究利用目前著名之物件偵測演算法（Scaled-YOLOv4與YOLOv5），並加上自行定義文物圖片訓練資料，獲得我們所需的文物辨識模型，進行文物辨識與盤點，以期提高內部管理及盤點效率。

關鍵字

文物物件偵測；科技盤點；深度學習； Scaled-YOLOv4 ；損失函數

並列摘要

The four main functions of modern museums are collection, research, display, and education. The preservation and management of cultural artifact collections are the key tasks involved in collection. As the core objects of display and research at museums, cultural artifact collections are the greatest physical evidence of the endurance of history and culture. Each museum houses multiple collections. Because museums serve as the main venues for the protection of cultural assets, collection storage is one of the main functions of museum buildings. Most of a museum's collections, except for cultural artifacts that are on display or are being loaned, are kept in storage. Inventory of cultural artifacts is a cumbersome but crucial task in museum collection management. Smart technology can be used to improve the efficiency and quality of collection management and achieve intelligent management (smart storage). In this study, two popular object detection algorithms (Scaled-YOLOv4 and YOLOv5) were trained on self-defined cultural artifact images to develop a model for cultural artifact recognition and inventory, which may be used to improve the efficiency of museum inventory management.

並列關鍵字

Cultural Artifact Detection ； Technology Inventory ； Deep Learning ； Scaled-YOLOv4 ； Loss Function

參考文獻

顏上晴（2011）。〈博物館程序標準之研探〉。《博物館學季刊》，25（3），43-63。doi:10.6686/MuseQ.201107_25(3).0002

Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2020). Focal loss for dense object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42, 318-327. doi:10.1109/TPAMI.2018.2858826

Padilla, R., Passos, W. L., Dias, T. L. B., Netto, S. L., & da Silva, E. A. B. (2021). A comparative analysis of object detection metrics with a companion open-source toolkit. Electronics, 10, 279. doi:10.3390/electronics10030279

Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In IEEE Computer Society (Ed.), 29th IEEE Conference on Computer Vision and Pattern Recognition (pp. 779-788). Los Alamitos, CA: IEEE Computer Society. doi:10.1109/CVPR.2016.91

Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2021). Scaled-yolov4: Scaling cross stage partial network. In IEEE Computer Society (Ed.), 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 13024-13033). Los Alamitos, CA: IEEE Computer Society. doi:10.1109/CVPR46437.2021.01283

主題瀏覽