thundarstrom/indian-anpr-ocr-corpus

Indian License Plate Character Recognition (PARSeq & LMDB) Part of the Edge-AI Traffic & Vehicle Analytics System repository by thundarstrom. Dataset Summary A curated dataset of 18,537 normalized license plate image crops aligned strictly with Indian Motor Vehicle Act alphanumeric formats (^[A-Z]{2}[0-9]{1,2}[A-Z]{1,3}[0-9]{4}$). Formats Provided Raw Crops & Ground Truth: parseq_dataset/ with gt.txt (Tab-delimited: filename \t text).… See the full description on the dataset page: https://huggingface.co/datasets/thundarstrom/indian-anpr-ocr-corpus.