Handwritten text recognition method and system based on sound signals

Abstract

The invention relates to a handwritten text recognition method and system based on a sound signal. The method comprises the following steps: setting a text recognition range; converting a plurality of handwritten character picture sets of different users into sound signals; entering an initialization stage, outputting a character recognition result based on a random forest model and user confirmation, and obtaining an enhanced recognition set of all characters in a text recognition range; entering a use stage, and finding a sound signal closest to the feature sound signal of the handwritten character of the user in the enhanced recognition set by using a KNN classification algorithm so as to realize text recognition; and performing error correction processing. Compared with the prior art, on the premise that the privacy of the user is guaranteed, the handwritten text input by the user is recognized through the sound signal, non-contact handwritten text input recognition is achieved, recognition of different languages can be achieved in different environments, the application range is wide, reliability is high, and the possibility of misrecognition is reduced.

Description

technical field [0001] The invention relates to the field of handwritten text recognition, in particular to a method and system for handwritten text recognition based on sound signals. Background technique [0002] In recent years, with the consumption upgrade and the gradual popularization of AI, VR, AR and other technologies, the functions of smart devices have become more and more diversified. Wearable smart devices such as smart watches and smart wristbands have been favored by many users because of their portable and practical characteristics. favorite. However, smart watches or other smart wearable devices are limited by a small interactive interface and a small number of interactive interfaces, which makes their interaction with users poor, and it is not convenient for users to type or input text, and handwritten text input is easy to be Misidentification has greatly reduced the usability of some smart wearable devices. In order to solve this dilemma, many device ma...

AI Technical Summary

Problems solved by technology

However, the power consumption of the camera is relatively large, and the camera is easily restricted by the use environment, and the effect is not good at night or in a dimly lit environment, and the camera will also touch the privacy of the user.

[0004] Chinese patent CN201911122436.7 discloses a handwritten text recognition method, which outputs recognized characters by collecting motion signals of the user's handwriting actions. However, this method requires a stylus accessory, and uses movement information such as displacement and rotation collected by the sensor on the stylus accessory. As a motion signal, handwritten text recognition, in the field of smart device interaction, obviously reduces the portability of the device

Images