基於圖像內容特徵匹配之陶瓷碎片重建研究

Content based Feature Matching for Ceramic Fragment Reconstruction

古涵仁(Han-Jen Ku)

指導教授 ： 丁肇隆

共同指導教授 ： 張瑞益(Ray-I Chang)

國立臺灣大學/工學院/工程科學及海洋工程學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.01055

摘要

歷史陶瓷文物經過日積月累的自然變動，在考古挖掘時，通常會有大量的碎片散落在不同處，考古學家在找尋碎片相關特性或藉由經驗拼接的過程，經常會消耗數十年的時間。因此本論文提出一個基於圖像內容特徵之陶瓷碎片重建系統，自動拼接碎片，最終產出拼接建議結果，給予使用者參考，使相關工作者能利用此系統找出匹配的陶瓷碎片標本。本系統主要由三個模組所構成：拼接碎片前處理、產生候選匹配組合、挑選最佳候選匹配組合。經由萃取碎片上的圖像內容特徵，以曲線匹配找出候選匹配組合後，使用二進位粒子群最佳化演算法挑選最佳候選匹配組合。實驗顯示，本研究提出之方法相較於過去相關研究，不僅提供更高的準確度，同時具備更低的最佳化時間耗損，並能適用於多種資料碎片，解決單一特徵不能解決之問題。

關鍵字

碎片重建 ； 特徵萃取 ； 曲線匹配 ； 二進位粒子群最佳化演算法

並列摘要

Archaeological ceramic relics are changed by times, huge amount of fragments will be found at different places during archaeological excavation. Archaeological fragments relation founding and reconstruction requires huge time costing and expert knowledge. Our research proposes a content based feature matching for ceramic fragment reconstruction system (CFRS) which automatically reconstruct fragments and generate a suggestion for user. Related workers can find out matching ceramic fragments through this system. CFRS includes three main module: fragments preprocessing, produces matching candidates and select matching candidates. The features will be first extracted from the images and matching candidates will be picked up through curve matching. After that, Binary Particle Swarm Optimization (BPSO) is applied to select optimized matching candidates. Experiment results indicates the proposed method not only attends higher precision but also reduce the optimization time in our knowledge. Support different data types and solve problems that single feature cannot solve.

並列關鍵字

Fragment reconstruction ； Feature extraction ； Curve matching ； Binary Particle Swarm Optimization (BPSO)

參考文獻

[1] 邱斯嘉, "太平洋史前Lapita陶器：線上數位資料庫的建置與其對後期深化研究的效益," 考古人類學刊, pp. 123-157, 2011.

[3] C. S. Belenguer and E. V. Vidal, "Archaeological fragment characterization and 3D reconstruction based on projective GPU depth maps," in Virtual Systems and Multimedia (VSMM), 2012 18th International Conference on, 2012, pp. 275-282.

[5] G. Oxholm and K. Nishino, "A flexible approach to reassembling thin artifacts of unknown geometry," Journal of Cultural Heritage, vol. 14, pp. 51-61, 2013.

[7] E. Justino, L. S. Oliveira, and C. Freitas, "Reconstructing shredded documents through feature matching," Forensic science international, vol. 160, pp. 140-147, 2006.

[9] Z. Liangjia, Z. Zongtan, and H. Dewen, "Globally Consistent Reconstruction of Ripped-Up Documents," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 30, pp. 1-13, 2008.