Digital Archive Stamp Detection and Extraction

Archives contain valuable historical information and must be properly preserved. However, traditional archival materials are vulnerable to damage from water, fire, and mold, making long-term storage difficult. To address this issue, digital archives have been established for management. As a result, effective storage, detection, extraction, and utilization of archive information has become a focus of attention. This paper focuses on the feature extraction of archival stamp images, proposing a network structure of stamp extraction based on generative adversarial network for texture feature extraction of stamp images. This method aims to extract more refined texture features, improving the accuracy of stamp text recognition. An improved stamp text recognition method is proposed using PP-OCR, which can recognize text for multiple shape seals. This method effectively solves the problem of deep learning models being unable to recognize text due to the bending and tilting of the ring-shaped text in the stamp. Overall, this research aims to enhance the preservation and utilization of archival materials by improving feature extraction and text recognition methods.