Writer self-adapting method based on increment quadratic discriminant function for discriminating subspace

Abstract

The invention provides writer self-adapting method based on an increment quadratic discriminant function for a discriminating subspace, and the method comprises the following steps: dynamically updating an MQDF (modified quadratic discriminant function) recognition model by utilizing an increment sample of the writing style of a special user; and leading the updated MQDF recognition model to adapt to the writing style of the special user, thus achieving the effect of improving the recognition rate of the special user. By creatively combining an increment learning algorithm, and an MQDF sorting algorithm and an ILDA (Increment Linear Discriminant Analysis) algorithm, the invention provides an ILDAIMQDF algorithm, and applies the ILDAIMQDF in the field of self-adaption of writers based on handwritten Chinese character recognition, thus solving the problem that the handwritten Chinese character recognition engine has low recognition rate on the special user. By utilizing the method, the handwritten Chinese character recognition engine can adapt to the handwriting style of the specific user, so as to improve the recognition accuracy.

Description

technical field [0001] The invention belongs to the technical field of using computer processing equipment to recognize handwritten character images, in particular to a writer's self-adaptive method based on an incremental quadratic discrimination function of a discrimination subspace. technical background [0002] Handwritten Chinese character recognition generally refers to the recognition technology that the user writes Chinese characters through a handwriting input device (such as a tablet, touch screen, mouse, etc.), and at the same time, the computer converts the Chinese character writing track collected by the handwriting input device into the corresponding Chinese character machine internal code. Traditional handwriting recognition technology usually adopts single-character input method, that is, writing a Chinese character to recognize a Chinese character. The recognition engine used has nothing to do with the user, that is, the recognition engine is trained through...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the deficiency that the traditional recognition engine cannot adapt to the writing style of a specific user, and provide a method that allows the recognition engine to dynamically adapt to the writing style of a specific user so as to improve the recognition rate

Images