Date of Award
5-2024
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Electrical Engineering and Computer Science
First Advisor
Debasis Mitra
Second Advisor
Xianqi Li
Third Advisor
Eraldo Ribeiro
Fourth Advisor
Brian Lail
Abstract
This research introduces ASR-net(Ancient Script Recognition), a groundbreaking system that automatically digitizes ancient Indus seals by converting them into coded text, similar to Optical Character Recognition for modern languages. ASR-net, with an 95% success rate in identifying individual symbols, aims to address the crucial need for automated techniques in deciphering the enigmatic Indus script. Initially Yolov3 is utilized to create the bounding boxes around each graphemes present in the Indus Valley Seal. In addition to that we created M-net(Mahadevan) model to encode the graphemes. Beyond digitization, the paper proposes a new research challenge called the Motif Identification Problem (MIP) related to recurring patterns (motifs) on Indus seals that appear to have specific functions within certain periods of the civilization. Despite challenges in applying deep learning to MIP, The database was created to store the ImageID, Image, the list of encoded graphemes present in that particular image followed by the Motif on the IVC Seal in the structured format.
Recommended Citation
Atturu, Deva Munikanta Reddy, "Deep Learning in Indus Valley Script Digitization" (2024). Theses and Dissertations. 1416.
https://repository.fit.edu/etd/1416