Offline Pashto Characters Dataset for OCR Systems

In computer vision and artificial intelligence, text recognition analysis based on images play a key role in the retrieving process. Enabling machine learning technique to recognize handwritten characters of specific language requires standard dataset. Acceptable character datasets are available many languages including English, Arabic, more. However, lack for Pashto hinders application suitable algorithm recognizing useful insights. order address this issue, study presents first image dataset (HPCID) scientific research work. This consists fourteen thousand, seven hundred, eighty-four samples—336 samples each 44 Such collected an A4-sized paper from different students Department University Peshawar, Khyber Pakhtunkhwa, Pakistan. On total, 336 faculty members contributed developing proposed database accumulation phase. contains multisize, multifont, multistyle varying structures.