Hand writing number identification method based on kernel function

Abstract

A hand-written number identification method based on nuclear function, including character extraction and identification of the hand-written number, at the phase of the extraction, firstly map the low-dimension character to the high-dimension space via nonlinear mapping function to make the problem of the character space become linear dividable or near linear dividable and further uses the nuclear function to extract the characters; at the character identification phase, use simple k-NN distance classifier and shortest distance classifier to identify.

Description

technical field [0001] The invention relates to a recognition method used in the technical field of pattern recognition, in particular to a kernel function-based handwritten digit recognition method. Background technique [0002] The learning method based on kernel function is a new method for machine learning. It has opened up a new land in the field of machine learning and has been successfully applied to pattern recognition, such as target recognition, text classification, time series prediction, etc., and has achieved very good results in both supervised and unsupervised learning. application effect, and fully demonstrated the superior performance of this method. The basic idea of ​​the kernel learning method is that for linearly inseparable data in the input space, the data X in the input space is first mapped to a high-dimensional feature space (the dimension can be infinite) F through a nonlinear mapping, so that in the feature The problem in the space becomes linea...

AI Technical Summary

Problems solved by technology

Although the features extracted by Kernel Principal Component Analysis (KPCA) are very descriptive of the original signal, the extracted features do not have the best discriminative ability. Deficiency

Images