Classification and detection of symbols in ancient papyri

Abstract

The advent of powerful image processing and machine learning tools has generated opportunities for more automated and efficient paleography. In this work, methods were developed to automate aspects of the transcription process of ancient papyri. First, a technique is proposed that employs color thresholding and contouring for the automatic extraction of symbols from papyri photographs. Second, two symbol classifiers are considered: a support vector machine that intakes Histogram of Oriented Gradients features, and a convolutional neural network. Finally, a novel system is described based on the sliding window approach that limits the number of windows to be considered and uses a combination of a support vector machine and a convolutional neural network to benefit from both accurate and fast detections. The performance of these methods was evaluated on a set of papyri photographs from the Petrie Museum of Egyptian Archaeology.