Scene Text detection and Recognition by Using Multi-Level Features Extractions Based on You Only Once Version Five (YOLOv5) and Maximally Stable Extremal Regions (MSERs) with Optical Character Recognition (OCR)
Essam Almousawee, Nidhal K. El Abbadi · Al-Salam Journal for Engineering and Technology · 2022
Textual information within scene images is very important in computer vision applications such asimage retrieval based on its content, Tourist translator, and Navigation systems assistant. This paper presented ascene text recognition system based on YOLO. The main steps of the suggested methodology are: text detectionand localization (using YOLOv5), text segmentation (using Morphological processing as a new method), featuresextraction (using MSERs), character segmentation and word segmentation (using bounding boxes with graph), andfinally character recognition (using OCR). In this work we create a new dataset model that includes most of thetext challenges such as Font type, Font size, Font color, and Font. The proposed system gives higher performancefor detection, localization, and recognition when using dataset containing many challenges, the results were 80%,96%, and 87.6 for precision, recall, and F-score respectively. Comparing with other similar works it was better.The accuracy of (OCR) is more than 99%.