Computational Ceramicology
This work provides a computational aid for field archaeologists to identify potsherds more efficiently, though it is incremental as it builds on existing image recognition techniques with domain-specific adaptations.
The researchers tackled the problem of identifying potsherds in archaeology by developing two machine-learning tools based on images: one using a novel deep-learning architecture for fracture outlines and another using standard methods for decorative features, achieving results that address data scarcity and class imbalance through synthetic data and specialized training loss.
Field archeologists are called upon to identify potsherds, for which purpose they rely on their experience and on reference works. We have developed two complementary machine-learning tools to propose identifications based on images captured on site. One method relies on the shape of the fracture outline of a sherd; the other is based on decorative features. For the outline-identification tool, a novel deep-learning architecture was employed, one that integrates shape information from points along the inner and outer surfaces. The decoration classifier is based on relatively standard architectures used in image recognition. In both cases, training the classifiers required tackling challenges that arise when working with real-world archeological data: paucity of labeled data; extreme imbalance between instances of the different categories; and the need to avoid neglecting rare classes and to take note of minute distinguishing features of some classes. The scarcity of training data was overcome by using synthetically-produced virtual potsherds and by employing multiple data-augmentation techniques. A novel form of training loss allowed us to overcome the problems caused by under-populated classes and non-homogeneous distribution of discriminative features.