Apparatus for recognition of handwritten Chinese characters

Abstract

An apparatus for recognition of handwritten Chinese characters contains a bus; an input means connected to the bus for receiving input imagery data created from a handwritten Chinese character; a Cellular Neural Networks or Cellular Nonlinear Networks (CNN) based integrated circuit operatively connected to the bus for extracting features out of the input imagery data using pre-trained filter coefficients of a plurality of order convolutional layers stored therein; a memory connected to the bus, the memory being configured for storing weight coefficients of fully-connected (FC) layers; a processing unit connected to the bus for performing computations of FC layers to classify the extracted features from the CNN based integrated circuit to a particular Chinese character in a predefined Chinese character set; and a display unit connected to the bus for displaying the particular Chinese character. Greater than 95% recognition accuracy is achieved using multiple bi-valued 3*3 filter kernelsas pre-trained filter coefficients.

Description

technical field [0001] The present invention relates generally to the field of machine learning, and more particularly, to an apparatus for handwritten Chinese characters (Chinese characters) with a cellular neural network or cellular nonlinear network (CNN) based integrated circuit embedded therein. Background technique [0002] Ideograms are graphic symbols that represent ideas or concepts. Some ideograms are intelligible only through familiarity with prior conventions; others express their meaning through pictorial resemblance to physical objects. [0003] Chinese characters are ideographic characters mainly used for writing Chinese. Even if Chinese students already know the characters, learning to read handwritten Chinese takes practice. There are two main reasons: [0004] 1. Most people's handwriting is not as neat as printed letters or standard computer fonts. Writing obviously takes practice and time; most people simply cannot be bothered. The result is sloppy, ...

AI Technical Summary

Problems solved by technology

This is faster to write, but harder to read

[0005] 2. People write characters in different ways, so even if a student has learned how a person writes, it cannot be assumed that a student understands what someone else wrote

However, learning handwritten Chinese characters remains a challenge

Images