A deep learning-based model for recognising vertically oriented texts in natural scenes

People with impairment encounter problems when trying to recognize the texts within their surroundings. Much recent research focuses on recognizing texts with horizontal, arbitrary, and curved orientations in natural scenes. However, there is a lack of research on recognizing vertically oriented scene texts. This thesis presents a comprehensive literature survey of the existing texts recognition methods in natural scene images. It also discusses the design and development of the proposed vertical text recognition model, the testing and evaluation to prove its performance and potential in recognizing vertically oriented texts in natural scene images.