Natural scene text detection based on SWT, MSER and candidate classification
Leibin Guan, Jizheng Chu · 2017
This paper presents a novel scene text detection algorithm based on Stroke Width Transform (SWT), Maximally Extremal Regions (MSER) and candidate classification. Firstly, utilize the SWT and MSER to extract the candidate characters at the same time. Secondly, preliminary filtering the candidate connected components based on heuristic rules. Thirdly, using mutual verification and integration to class all candidate into two categories: strong candidates, weak candidates. If the weak candidate has similar properties with strong candidate, then the weak candidate is changed into strong candidate. Finally, the text area is aggregated into text lines by text line aggregation algorithm. The experiment results on public datasets show that the proposed method can detect text lines effectively.