An algorithm for context‐aware handwriting recognition and pencraft analysis

Handwriting recognition imposes a specific challenge on image analysis and preprocessing. This paper suggests an improvement over some commonly used algorithms for optical character recognition, by adding context awareness and self-adjustment of the model parameters, based on individual writing style and pencraft characteristics. Discussed analytical steps can be used along different OCR implementations and complement character recognition analysis. Suggested approach can be used also with block-character inputs, where additional meta information is generated from content analysis of the inputs and search of expression/style of writing patterns.