作者: Sri-Kaushik Pavani , Ekta Prashnani
DOI:
关键词:
摘要: A system and method using a text extraction application for identifying words with multiple orientations from an image are described. The receives input image, generates progressively blurred images, detects blobs in the outputs ellipses over blobs, word orients normalizes first version of word, inverted performs OCR on confidence scores associated word.