Text Detection with Convolutional Neural Networks

M. Delakis, C. Garcia, Orange Labs, 2008

Text detection is an important preliminary step before text can be recognized in unconstrained image environments. We present an approach based on convolutional neural networks to detect and localize horizontal text lines from raw color pixels. The network learns to extra ct and combine its own set of features through learning instead of using hand-crafted ones. Learning was also used in order to precisely localize the text lines by simply training the network to reject badly-cut text and without any use of tedious knowledge-based post-processing. Although the network was trained with synthetic examples, experimental results demonstrated that it can outperform other methods on the real-world test set of ICDAR’03.