本文選題：手寫體識別 + 神經(jīng)網(wǎng)絡(luò)　； 參考：《天津大學(xué)》2016年碩士論文

【摘要】：隨著模式識別技術(shù)在信息科學(xué)中的廣泛應(yīng)用,手寫文本識別也成為了現(xiàn)在的熱點(diǎn)研究內(nèi)容,在我們身邊常用的就有手機(jī)的手寫輸入,車牌自動拍照提取,文檔掃描輸入等,都是要求識別圖像中的文本。文本識別技術(shù)在很多需要自動識別信息的領(lǐng)域有很重要的理論意義和實(shí)用價值。本文研究神經(jīng)網(wǎng)絡(luò)方法用于手寫文本識別。本文論述了手寫文本識別的意義和發(fā)展現(xiàn)狀,討論了字符識別預(yù)處理的灰度化、圖像去噪、二值化、歸一化和字符細(xì)化等基本步驟。本文實(shí)驗(yàn)部分對手寫48×48的彩色字符圖像進(jìn)行灰度化、二值化、歸一化為16×16的黑白圖像,字符特征向量提取用到了逐像素特征提取法。將提取的字符特征轉(zhuǎn)換成神經(jīng)網(wǎng)絡(luò)的輸入向量。選取20組手寫樣本訓(xùn)練BP神經(jīng)網(wǎng)絡(luò),用另外20組樣本測試神經(jīng)網(wǎng)絡(luò)的識別效果。本文采用不同訓(xùn)練函數(shù)的BP算法進(jìn)行訓(xùn)練,比較識別效果和效率,最后采用帶動量項(xiàng)的自適應(yīng)學(xué)習(xí)率訓(xùn)練函數(shù)。本實(shí)驗(yàn)在MATLAB平臺下實(shí)現(xiàn),論文最后對實(shí)驗(yàn)結(jié)果進(jìn)行了分析總結(jié)。本文研究表明,基于BP網(wǎng)絡(luò)的手寫文本識別的正確率較高,有一定抗干擾和噪聲能力,將來可以進(jìn)一步應(yīng)用到實(shí)踐中去。
[Abstract]:With the wide application of pattern recognition technology in information science, handwritten text recognition has become a hot research topic. All are required to recognize the text in the image. Text recognition technology has important theoretical significance and practical value in many fields which need automatic recognition information. In this paper, neural network method is studied for handwritten text recognition. This paper discusses the significance and development of handwritten text recognition, and discusses the basic steps of character recognition preprocessing, such as grayscale, image denoising, binarization, normalization and character thinning. In the experiment part of this paper, the color character image of handwritten 48 脳 48 is grayscale, binary, normalized to 16 脳 16 black and white image. The character feature vector extraction is based on pixel by pixel feature extraction method. The extracted character features are converted into input vectors of neural networks. Twenty groups of handwritten samples were selected to train BP neural network and the other 20 groups of samples were used to test the recognition effect of BP neural network. In this paper, BP algorithm with different training functions is used to compare the recognition effect and efficiency. Finally, the adaptive learning rate training function of the driving term is used. The experiment is implemented on MATLAB platform. Finally, the experimental results are analyzed and summarized. The research shows that the BP neural network based handwritten text recognition has a high accuracy, anti-jamming and noise ability, and can be further applied to practice in the future.
【學(xué)位授予單位】：天津大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TP391.43;TP183

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張德富,殷正坤;人工神經(jīng)網(wǎng)絡(luò)的發(fā)展及其哲理[J];科學(xué)技術(shù)與辯證法;2000年04期

2 陳瓊,鄭啟倫;人工神經(jīng)網(wǎng)絡(luò)系統(tǒng)硬件實(shí)現(xiàn)的研究進(jìn)展[J];電路與系統(tǒng)學(xué)報(bào);2000年01期

3 單潮龍,馬偉明,賁可榮,張磊;BP人工神經(jīng)網(wǎng)絡(luò)的應(yīng)用及其實(shí)現(xiàn)技術(shù)[J];海軍工程大學(xué)學(xué)報(bào);2000年04期

4 羅維平,向陽;人工神經(jīng)網(wǎng)絡(luò)的實(shí)現(xiàn)及其應(yīng)用[J];中南民族學(xué)院學(xué)報(bào)(自然科學(xué)版);2000年S1期

5 孫學(xué)梅;人工神經(jīng)網(wǎng)絡(luò)及其在控制中的應(yīng)用[J];張家口職業(yè)技術(shù)學(xué)院學(xué)報(bào);2000年02期

6 汪鐳,周國興,吳啟迪;人工神經(jīng)網(wǎng)絡(luò)在傳動領(lǐng)域中的應(yīng)用[J];電機(jī)與控制學(xué)報(bào);2001年01期

7 侯晉,陳國松,王鎮(zhèn)浦;人工神經(jīng)網(wǎng)絡(luò)的發(fā)展及在多元校正中的應(yīng)用[J];分析科學(xué)學(xué)報(bào);2001年01期

8 戰(zhàn)凱,朱文斌,黃文斌;人工神經(jīng)網(wǎng)絡(luò)在軍用艦船工程中的應(yīng)用展望[J];青島大學(xué)學(xué)報(bào)(工程技術(shù)版);2001年03期

9 梁艷春,聶義勇;從科學(xué)研究方法論看人工神經(jīng)網(wǎng)絡(luò)研究的發(fā)展[J];吉林大學(xué)學(xué)報(bào)(信息科學(xué)版);2002年01期

10 索興梅,王崇桃;人工神經(jīng)網(wǎng)絡(luò)在農(nóng)業(yè)中的應(yīng)用研究[J];計(jì)算機(jī)與農(nóng)業(yè);2002年02期

相關(guān)會議論文 前10條

1 吳兵;;一種具有語義分布的自構(gòu)造的新人工神經(jīng)網(wǎng)絡(luò)系統(tǒng)及其應(yīng)用[A];1999年中國神經(jīng)網(wǎng)絡(luò)與信號處理學(xué)術(shù)會議論文集[C];1999年

2 劉R，

本文編號：1835305