Scene Text Detection and Localization using Fully Convolutional Network

Early methods for text detection in scene images have obtained encouraging achievements over several benchmarks. Nevertheless, they commonly drop when working with challenging situations, although implemented by using deep learning designs. In this research, we introduce a robust yet simple pipeline that produces accurate and fast text detection in natural scene images. The pipeline quickly predicts text of random orientations in full-color images, discarding redundant intermediate stages, with a single neural network. The proposed method experimented on ICDAR 2015 and MSRA-TD500 datasets and an advantage in detecting text from natural scene images efficiently for assisting the visually impaired.