DOME-T: adversarial computer vision attack on deep learning models based on Tchebichef image moments
A modification of a stopping method for text recognition in a video stream with best frame selection
An approach to road scene text recognition with per-frame accumulation and dynamic stopping decision