Effect evaluation system and method for sorting stability of convolutional neural network handwriting recognition

Abstract

The invention relates to an effect evaluation system and method for the sorting stability of convolutional neural network handwriting recognition. The effect evaluation system includes a convolutionalneural network handwriting recognition module, a parameter estimation module and an uncertainty calculation module, wherein the convolutional neural network handwriting recognition module takes handwritten data of a user as input data, enables the data to pass through a convolutional neural network (CNN), finally outputs the predicted classification probability and carries out sorting according to the classification; the parameter estimation module takes the predicted probability outputted by the CNN handwriting recognition module as the input to estimate parameters required for calculating the relation uncertainty by adopting a statistical mode; and the uncertainty calculation module takes parameters outputted by the parameter estimation module as input data, and estimates the stabilityof a convolutional neural network handwriting recognition sorting result through designing a fusion probability calculation model for the data uncertainty and the data relation uncertainty.

Description

technical field [0001] The invention belongs to the field of effect evaluation, and in particular relates to a system and method for evaluating the stability of handwriting recognition and sorting by a convolutional neural network. Background technique [0002] Handwriting recognition refers to the process of informatizing the orderly trajectory generated when writing on a handwriting device into the internal code of Chinese characters. In fact, it is a mapping process from the coordinate sequence of the handwritten trajectory to the internal code of Chinese characters. It is the most natural human-computer interaction. , One of the most convenient means. With the popularization of mobile information tools such as smart phones and PDAs, handwriting recognition technology has also entered the era of large-scale application. [0003] The standard for evaluating the quality of a handwritten character is mainly determined by the accuracy of the recommended characters after reco...

AI Technical Summary

Problems solved by technology

[0004] The technical solution of the present invention is to overcome the deficiencies of the prior art, provide a convolutional neural network handwriting recognition and sorting stability evaluation system, and establish a recognition effect that can represent data uncertainty and sorting stability for convolutional neural network handwriting recognition Evaluate the model to solve the problem that the stability of identification ranking is difficult to estimate

Images