SuperOCR: A Conversion from Optical Character Recognition to Image Captioning
This work addresses the problem of improving character recognition accuracy and simplifying training data requirements for real-world OCR applications, particularly for embedded systems.
This paper converts optical character recognition (OCR) into an image captioning task, eliminating the need for character detection and bounding box labels during training. The method outperforms existing approaches on license plate and watermeter character recognition tasks.
Optical Character Recognition (OCR) has many real world applications. The existing methods normally detect where the characters are, and then recognize the character for each detected location. Thus the accuracy of characters recognition is impacted by the performance of characters detection. In this paper, we propose a method for recognizing characters without detecting the location of each character. This is done by converting the OCR task into an image captioning task. One advantage of the proposed method is that the labeled bounding boxes for the characters are not needed during training. The experimental results show the proposed method outperforms the existing methods on both the license plate recognition and the watermeter character recognition tasks. The proposed method is also deployed into a low-power (300mW) CNN accelerator chip connected to a Raspberry Pi 3 for on-device applications.