Handwritten form identification system of BP neural network based on dynamic sample selection strategy

Abstract

The invention provides a handwritten form identification system of BP neural network based on a dynamic sample selection strategy. Weights of different layers of network neurons are initialized randomly; a gradient descent method is used to optimize the network weight, in first round of iteration, all samples are used to calculate the total gradient, the total gradient is used to update the weights of different layers, and whether a sample serves as a training sample in next round of iteration is determined according to whether the sample is far from a decision boundary, and the training samples selected in the last round are used to calculate the total gradient, update the weights of different layers and select samples for next round of iteration repeatedly till the minimal stop error or the maximal interaction frequency is reached; and the obtained neural network is used to identify an unknown hand-written font sample. Compared with a traditional classification technology, According to sample selection strategy of the invention, the samples are selected dynamically according to the distances to the decision boundary, the amount of training sample is decreased step by step, and an algorithm can effectively solve the problem that training time of the BP network is too long in a big data set.

Description

technical field [0001] The invention relates to the field of pattern recognition, in particular to a BP neural network handwriting recognition method and system based on a dynamic sample selection strategy. Background technique [0002] Optical Character Recognition (OCR) is a practical technology for computers to automatically and quickly identify text on paper and convert it into editable text. Handwritten digit recognition is an important branch of optical character recognition technology. Its research object is how to use computers to automatically recognize Arabic numerals written by people on paper. The main difficulties in handwritten number recognition are: first, there is not much difference in the shape of some numbers, which makes it difficult to distinguish these numbers accurately; It is used by people in all countries and regions, and its writing has obvious regional characteristics. It is difficult to fully achieve a universal digital recognition system with ...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problems that the existing handwritten character recognition technology based on BP neural network has too long training time and low practicability when processing large data sets, the solution of the present invention is to provide a neural network technology based on a dynamic sample selection strategy to Realize handwritten font recognition, use BP neural network to train network weights, and use gradient descent method to optimize the algorithm model

Images