Chinese character model training method, handwritten character recognition method, device, apparatus and medium

Abstract

The invention discloses a Chinese character model training method, a handwritten character recognition method and device, an apparatus and a medium. The Chinese character model training method comprises the following steps: obtaining font image training samples, labeling the font image training samples with a Chinese secondary font library, and dividing the font image training samples into a training set and a test set according to a preset distribution rule; inputting the training set into a convolution loop neural network model, and updating weights and biases in the convolution loop neuralnetwork model by using the back propagation algorithm based on small batch gradient descent to obtain an initial handwritten character recognition model; inputting the test set into the initial handwritten character recognition model, obtaining the recognized Chinese characters corresponding to each handwritten character image, and obtaining the recognition accuracy rate on the basis of the recognized Chinese characters and label Chinese characters; and determining the initial handwritten character recognition model as a target handwritten character recognition model if the recognition accuracy rate is higher than the preset accuracy rate. The target handwritten character recognition model can recognize handwritten characters accurately.

Description

technical field [0001] The invention relates to the field of handwritten character recognition, in particular to a Chinese character model training method, handwritten character recognition method, device, equipment and medium. Background technique [0002] Most traditional Chinese character recognition methods use OCR (Optical Character Recognition, Optical Character Recognition) technology for recognition. Because there are many types of Chinese characters, such as "Song, Kai, Yao and Fangsong", and the structure of some Chinese characters is relatively complicated, such as "魑, Mei", and there are many characters with similar structures in Chinese characters, such as "Shou and Love", so that the accuracy of Chinese character recognition cannot be guaranteed. Standard, simple and standardized sentences can be recognized by using OCR (Optical Character Recognition) technology, but for sentences composed of handwritten words, because everyone's writing habits are different a...

AI Technical Summary

Problems solved by technology

Because there are many types of Chinese characters, such as "Song, Kai, Yao and Fangsong", and the structure of some Chinese characters is relatively complicated, such as "魑, Mei", and there are many characters with similar structures in Chinese characters, such as "Shou and Love", making the accuracy of Chinese character recognition impossible to guarantee

Standard, simple and standardized sentences can be recognized by using OCR (Optical Character Recognition) technology, but for sentences composed of handwritten words, because everyone's writing habits are different and not standard Chinese characters composed of horizontal and vertical strokes, When OCR technology is used for identification, there will be inaccurate identification, which greatly limits the performance of the identification system, resulting in low identification accuracy and unsatisfactory identification effect.

Images