Digitization of historical hydrological records - Benchmark dataset

This dataset is the 100-page benchmark used to evaluate the digitization of historical daily discharge records from the Italian hydrological yearbooks (Annali Idrologici), compiled and published by the Servizio Idrografico e Mareografico Nazionale (SIMN). It comprises 100 scanned page images in GIF format, each reproducing one printed table of daily river flow values (in m³/s) for a single gauging station and calendar year. Each image is paired with its manually digitized transcription, which serves as the ground-truth reference for every value in the table. Transcriptions are provided as spreadsheet files, with one file per page.