Overview of Scene Text Detection and Recognition

With the advent of the intelligent age, different natural scenes contain rich text information, resulting in a wide range of applications. Although many algorithms for scene text detection and recognition have been proposed, they still face great challenges, such as Different shapes, multilingual recognition, severe blurring, irregular aspect ratio, etc. This paper first summarizes the detection, recognition and end-to-end algorithms proposed in recent years, focuses on the current mainstream deep learning algorithm, and then introduces some common data sets in the experiment. Compared with the algorithm under the benchmark data set, some experimental results with good robustness are given. Finally, the future research field that hope to achieve further development is prospected.