Hana: A Handwritten Name Database for Offline Handwritten Text Recognition?

Methods for linking individuals across historical data sets, typically in combination with AI based transcription models, are developing rapidly. Probably the single most important identifier is personal names. However, names prone to enumeration and errors although modern methods designed handle such challenges these sources of critical should be minimized. For this purpose, improved large-scale databases crucial components. This paper describes provides documentation HANA, a newly constructed database which consists more than 1.1 million images handwritten word-groups. The collection names, containing 105 thousand unique total 3.3 examples. In addition, we present benchmark results deep learning models that automatically can transcribe from scanned documents. Focusing mainly on due its vital role linking, hope foster sophisticated, accurate, robust text recognition through making challenging publicly available. source, process, image-processing procedures involved extracting general forms.