【摘要】：隨著科技的發(fā)展,人們生活和工作產(chǎn)生大量的手寫體字符信息,考慮到這些字符所要表達信息的安全性和隱私性,讓機器實現(xiàn)快速、準(zhǔn)確的手寫體字符自動識別技術(shù)勢在必行。手寫體字符識別方法主要是光學(xué)字符識別,但因其識別率低、成本高等問題,還未能廣泛推廣使用。目前包括模板匹配、神經(jīng)網(wǎng)絡(luò)和支持向量機等模式識別的方法已經(jīng)投入到手寫體字符的識別研究。本文針對傳統(tǒng)的字符識別方法實時性差、成本高等問題,提出采用極限學(xué)習(xí)機算法實現(xiàn)手寫體字符識別。論文首先對模式識別的定義、基本組成系統(tǒng)和基本方法進行了介紹和討論,引出了利用神經(jīng)網(wǎng)絡(luò)進行模式識別的方法,對神經(jīng)網(wǎng)絡(luò)的工作原理和特點進行了分析和研究;然后提出用極限學(xué)習(xí)機實現(xiàn)手寫體字符識別方法,針對原始極限學(xué)習(xí)機存在的結(jié)構(gòu)風(fēng)險和經(jīng)驗風(fēng)險不平衡這一問題,提出使用正則極限學(xué)習(xí)機和傅里葉變換優(yōu)化極限學(xué)習(xí)機實現(xiàn)手寫體字符識別。設(shè)計基于BP神經(jīng)網(wǎng)絡(luò)、極限學(xué)習(xí)機、正則極限學(xué)習(xí)機和傅里葉變換優(yōu)化極限學(xué)習(xí)機四種算法實現(xiàn)手寫體字符識別的方法,包括預(yù)處理、特征選擇和降維等具體過程。手寫體字符識別算法仿真的訓(xùn)練樣本為MINIST樣本庫的10000個數(shù)字樣本,測試樣本數(shù)量為1000個,除采用BP神經(jīng)網(wǎng)絡(luò)、極限學(xué)習(xí)機、正則極限學(xué)習(xí)機和傅里葉變換優(yōu)化極限學(xué)習(xí)機四種算法實現(xiàn)手寫體字符的識別結(jié)果外,還設(shè)計實驗分析隱含層神經(jīng)元個數(shù)對仿真結(jié)果的影響。通過對算法仿真結(jié)果的對比分析,BP網(wǎng)絡(luò)作為最經(jīng)典的神經(jīng)網(wǎng)絡(luò)算法,在手寫體數(shù)字識別結(jié)果的精度上達到了較高的水準(zhǔn)。極限學(xué)習(xí)機算法較BP神經(jīng)網(wǎng)絡(luò)在訓(xùn)練時間上表現(xiàn)出極大的優(yōu)勢,但是識別精度低于BP神經(jīng)網(wǎng)絡(luò)�；跇O限學(xué)習(xí)機的兩種優(yōu)化算法,即正則極限學(xué)習(xí)機和傅里葉變換優(yōu)化極限學(xué)習(xí)機,與原始極限學(xué)習(xí)機相比提高了算法的泛化能力,提高了手寫體數(shù)字字符的識別精度。
[Abstract]:With the development of science and technology, people produce a large amount of handwritten character information in life and work. Considering the security and privacy of the information expressed by these characters, it is imperative for the machine to realize rapid and accurate automatic recognition of handwritten characters. The main method of handwritten character recognition is optical character recognition, but because of its low recognition rate and high cost, it has not been widely used. At present, pattern recognition methods, such as template matching, neural network and support vector machine (SVM), have been put into the research of handwritten character recognition. Aiming at the problems of poor real time and high cost of traditional character recognition methods, this paper proposes to use extreme learning machine algorithm to realize handwritten character recognition. Firstly, the definition, basic composition system and basic method of pattern recognition are introduced and discussed, the method of pattern recognition using neural network is introduced, and the working principle and characteristics of neural network are analyzed and studied. Then, a method of handwritten character recognition based on extreme learning machine is put forward, aiming at the imbalance between structural risk and empirical risk of original extreme learning machine. This paper presents a new method to realize handwritten character recognition by using regular limit learning machine and Fourier transform optimization learning machine. Based on BP neural network, extreme learning machine, regular ultimate learning machine and Fourier transform optimization extreme learning machine, this paper designs four algorithms to realize handwritten character recognition, including preprocessing, feature selection and dimensionality reduction. The training sample of handwritten character recognition algorithm simulation is 10, 000 digital samples of MINIST sample database, and the number of test samples is 1000. In addition to the recognition results of handwritten characters, four algorithms of regular limit learning machine and Fourier transform optimized limit learning machine are designed to analyze the effect of the number of hidden layer neurons on the simulation results. Through the comparison and analysis of the simulation results of the algorithm, the BP neural network, as the most classical neural network algorithm, has reached a high level in the accuracy of handwritten digital recognition results. Compared with BP neural network, the algorithm of extreme learning machine shows great superiority in training time, but the recognition accuracy is lower than that of BP neural network. Two optimization algorithms based on LLM, namely regular LLM and Fourier transform LLM, improve the generalization ability of the algorithm and the recognition accuracy of handwritten numeric characters compared with the original LLM.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.4;TP18

【參考文獻】

相關(guān)期刊論文 前10條

1 肖�？�;葛廣英;姚坤;尹紅敏;;基于HALCON的噴碼字符識別技術(shù)的研究與實現(xiàn)[J];現(xiàn)代電子技術(shù);2015年15期

2 王杰;畢浩洋;;一種基于粒子群優(yōu)化的極限學(xué)習(xí)機[J];鄭州大學(xué)學(xué)報(理學(xué)版);2013年01期

3 吳登國;李曉明;;基于極限學(xué)習(xí)機的配電網(wǎng)重構(gòu)[J];電力自動化設(shè)備;2013年02期

4 席偉;;基于判別函數(shù)算法的圖像分類器設(shè)計[J];電腦知識與技術(shù);2012年33期

5 毛力;王運濤;劉興陽;李朝鋒;;基于改進極限學(xué)習(xí)機的短期電力負(fù)荷預(yù)測方法[J];電力系統(tǒng)保護與控制;2012年20期

6 張東娟;丁煜函;劉國海;梅從立;;基于改進極限學(xué)習(xí)機的軟測量建模方法[J];計算機工程與應(yīng)用;2012年20期

7 樊振宇;;BP神經(jīng)網(wǎng)絡(luò)模型與學(xué)習(xí)算法[J];軟件導(dǎo)刊;2011年07期

8 趙志宇;常健;;模式識別概述及其應(yīng)用[J];信息與電腦(理論版);2010年10期

9 田娟;鄭郁正;;模板匹配技術(shù)在圖像識別中的應(yīng)用[J];傳感器與微系統(tǒng);2008年01期

10 岳曉峰;焦圣喜;韓立強;李洪洲;;模式識別中的光字符識別技術(shù)及應(yīng)用綜述[J];河北工業(yè)科技;2006年05期

相關(guān)博士學(xué)位論文 前1條

1 張艷麗;高維數(shù)據(jù)下的判別分析及模型選擇方法[D];山東大學(xué);2015年

相關(guān)碩士學(xué)位論文 前10條

1 鄒亞R，

本文編號：2144199