Evaluating The Performance of Different Text Detection and Recognition Models For Tyre Text Recognition
Computer Vision is a large part of artificial intelligence, specifically Optical Character Recognition. OCR helps by processing text and letters into digital content. One use case where Text recognition is not significant is tyre text. One challenge for tyre text is the contrast between the text and its background. Both are created in the same color but on different textures. Using pre-trained text region detection and recognition models, this paper aims to recognize text using several models. Textsnake and CRAFT is used for text region detection, while ABINet, ASTER and MORANv2 are used for text recognition. It was found that CRAFT had more correct detections but Textsnake was better for phrases and finding regions. As for Text recognition, it was observed that ABINet is better overall.
MMOCR can be installed following their webssite: https://mmocr.readthedocs.io/en/latest/get_started/install.html .
As for CRAFT+MORANv2, please read the readme.md file in the folder.
Jayson Mikael Hendra
Peter Nelson Subrata
All Results: https://bit.ly/4cqGKEj
All Images: https://bit.ly/3Vt5zsx