Smart pens to assist fibre optic sensors research: Evaluating OCR tools

Reproducibility is one of the big challenges in research. Lab notebooks have been used to record data, observations and relevant remarks of the research processes. Smart pens are devices that record audio, handwriting notes thanks to micro patterned paper, and generate pdf files and audio enriched notes (pencasts). The handwriting notes can then be processed using optical character recognition (OCR) software to generate digital documents allowing the user to archive and access these notes in an easier way. However, OCR for handwriting is still a challenge in the computer vision research area. In this paper, we report the evaluation results of different OCR tools when processing handwriting notes written by 7 participants focusing on the main elements and technical vocabulary identified in fibre optic sensors research.