A bilingual text detection in natural images using heuristic and unsupervised learning

Digital images are being produced in a massive number every day. A component that may exist in digital images is text. Textual information can be extracted and used in a variety of fields. Noise, blur, distortions, occlusion, font variation, alignments, and orientation are among the main challenges for text detection in natural images. Despite many advances in text detection algorithms, there is not yet a single algorithm that addresses all of the above problems successfully. Furthermore, most of the proposed algorithms can only detect horizontal texts, and a very small fraction of them consider the Farsi language. In this paper, a method is proposed for detecting multi-orientated texts in both the Farsi and English languages. We define seven geometric features to distinguish text components from the background, and propose a new contrast enhancement method for text detection algorithms. Our experimental results indicate that the proposed method achieves a high performance in text detection on natural images.