Text in a Wild and Its Challenges

Abstract

This chapter discusses the challenges encountered during recognition of wild text image. The challenges have been defined in detail with reference to the recent years’ publications in this field. Since decades, the interest shift of researchers has been witnessed from recognition of optical characters to recognition of characters appeared in natural images. Scene text recognition is a challenging problem due to the text having variations in font styles, size, alignment, orientation, reflection, illumination change, blurriness, and complex background. Among cursive scripts, Arabic scene text recognition is contemplated as a more challenging problem due to joined writing, same character variations, large number of ligatures and number of baselines, etc. The study on Latin and Chinese script-based scene text recognition system can be found but Arabic and Arabic-like scene text recognition problem is yet to be addressed in detail. The issues pertaining to text localization and feature extraction are also highlighted. Moreover, this chapter emphasized the importance of having benchmark cursive scene text dataset.