A novel approach for assemblage of historical artefacts using the Levenshtein distance and feedback loop

Broken museum artefacts are often incomplete and lack distinctive visual features that allow for clear identification. This makes it difficult for researchers to gain a more in-depth understanding of the past.

The article presents an original method, Skul-Mont, for reassembling broken two-dimensional elements with the conversion of their contours into abstract words. The method utilizes linguistic measures to search for word fragments matching, and utilizes a feedback loop mechanism to combine automated search with expert domain knowledge. The article discusses an example of arranging a broken relief from the Afriasyab Museum of Samarkand.

The feedback loop mechanism used in the experiment allows automatic solution determination based on only information about the contours of the elements. Experts can verify possible solutions, for example, by evaluating the continuity of the texture of individual fragments of the relief. Clusters of elements were identified that could be enlarged and amalgamated, based on the outcomes of subsequent iterations. Ultimately, a complete representation of the broken relief was obtained. The results confirm the efficacy of the methodology in arranging fragmented 2D entities and its potential to automate the search for feasible methods of assembling components. The method described in the article with a feedback loop can be considered as a valuable tool to aid museum employees and archaeologists in their efforts to assemble fragmented historical artefacts.