Machine Learning Artificial Character Generation

Abstract

The discussed embodiments address the problems of traditional ECR training by artificially generating handwritten characters in a unique manner according to machine learning techniques that transform handwritten samples according to generation and discriminative rules. The provided solution produces a large number of artificially generated characters that appear to be handwritten. Therefore, compared with the conventional technology, the embodiments of the present invention provide the system character library with more efficient acquisition of additional characters. Furthermore, embodiments are designed to be suitable for machine learning, and as the technique is implemented, the technique becomes increasingly effective. In short, the solutions presented improve computing technology itself, making it possible to reliably recognize electronic Chinese characters.

Description

technical field

[0001] The present invention relates to artificial neural networks and artificially generating handwritten characters, handwritten alphanumerics, and calligraphy. Embodiments discussed below artificially generate handwriting according to machine learning techniques that transform handwriting samples according to generation and discriminative rules, thereby producing human-generated handwriting that appears to be human generated handwriting. Background technique

[0002] Character recognition is an important application of artificial neural networks. Character recognition applications can recognize text in documents as well as symbols and characters in industrial applications. In creating an accurate artificial neural network, the system is trained to recognize characters, a long and tedious process. Typically, actual handwriting samples are repeatedly provided to the system to build up a library of characters that the system can recognize. In addition, the...

Images