本文選題：聯(lián)機(jī)手寫體識(shí)別 + 手寫體數(shù)學(xué)公式��； 參考：《電子科技大學(xué)》2017年碩士論文

【摘要】：在教育行業(yè),為了實(shí)時(shí)追蹤學(xué)生的學(xué)習(xí)軌跡和知識(shí)薄弱環(huán)節(jié),機(jī)器自動(dòng)識(shí)別學(xué)生答題的手寫筆跡成為必要的技術(shù)需求。因此,本文的研究重點(diǎn)為聯(lián)機(jī)數(shù)學(xué)公式手寫體識(shí)別,旨在提出一個(gè)穩(wěn)健可行的解決方案來識(shí)別學(xué)生的手寫數(shù)學(xué)公式筆跡,主要研究?jī)?nèi)容包括如下幾點(diǎn):1、提出了一種融合CNN和DBN的單字符識(shí)別模型應(yīng)用卷積神經(jīng)網(wǎng)絡(luò)(Convolutional Neural Network,CNN)搭建和訓(xùn)練了一個(gè)單字符分類器模型,針對(duì)神經(jīng)網(wǎng)絡(luò)對(duì)對(duì)抗樣本的脆弱性表現(xiàn),同時(shí)創(chuàng)造性的采用了深度信念網(wǎng)絡(luò)(DeepBeliefNetwork, DBN)的解碼重構(gòu)損失作為識(shí)別置信度評(píng)價(jià)模型,最后融合了 CNN和DBN的置信度評(píng)價(jià),增強(qiáng)了對(duì)對(duì)抗樣本的拒識(shí)能力。2、提出了一種基于組合排序的手寫體識(shí)別算法手寫數(shù)學(xué)公式中存在著大量的二義性,二維結(jié)構(gòu)準(zhǔn)確判定存在著相當(dāng)?shù)碾y度,同時(shí)還有許多容易混淆的相似字符,這些情況都增大了機(jī)器自動(dòng)識(shí)別的難度。本文提出了一種基于組合排序的手寫體識(shí)別算法,先把這些不確定的情況都保存下來作為候選,產(chǎn)生候選組合路徑,再基于詞組頻率表,語義模型,識(shí)別置信度等進(jìn)行路徑排序就能保證識(shí)別正確率,可以大大簡(jiǎn)化系統(tǒng)的復(fù)雜度,同時(shí)增大識(shí)別系統(tǒng)的魯棒性。3、提出了一種基于錯(cuò)誤識(shí)別案例的快速學(xué)習(xí)方法識(shí)別出錯(cuò)案例的調(diào)試工作非常繁重,本文提出了一個(gè)基于錯(cuò)誤案例的學(xué)習(xí)方法,可以快速的從我們提供的錯(cuò)誤案例標(biāo)記數(shù)據(jù)中學(xué)習(xí)到新的組合映射知識(shí),并將這些知識(shí)作為系統(tǒng)的組合補(bǔ)充分支,避免了再次出現(xiàn)同樣的識(shí)別錯(cuò)誤�；谏鲜龅姆椒�,實(shí)現(xiàn)了一個(gè)識(shí)別系統(tǒng),實(shí)驗(yàn)結(jié)果表明本文提出的聯(lián)機(jī)數(shù)學(xué)公式手寫體識(shí)別方法具有較高的識(shí)別率及較好的魯棒性。
[Abstract]:In the education industry, in order to track the students' learning track and knowledge weakness in real time, it is necessary to recognize the handwritten handwriting of students' answer questions automatically by machine. Therefore, this paper focuses on on-line mathematical formula handwritten recognition, in order to propose a robust and feasible solution to recognize the handwriting of students' handwritten mathematical formula. The main research contents are as follows: 1. A single character recognition model combining CNN and DBN is proposed. A single character classifier model is built and trained by convolution neural network Convolutional Neural Network. At the same time, it creatively adopts the decoding reconstruction loss of deep belief network (DBN) as the evaluation model of recognition confidence. Finally, it combines the confidence evaluation of CNN and DBN. In this paper, we enhance the ability of rejecting the antagonistic samples. 2. A handwritten recognition algorithm based on combinatorial sorting is proposed. There is a lot of ambiguity in the handwritten mathematical formula, and it is very difficult to determine the 2D structure accurately. At the same time, there are many confusing similar characters, which make automatic recognition more difficult. In this paper, a handwritten recognition algorithm based on combinatorial sorting is proposed. Firstly, these uncertainties are saved as candidates to produce candidate combination paths, and then based on phrase frequency table, semantic model. The correct rate of recognition can be guaranteed by the path sorting of recognition confidence, and the complexity of the system can be greatly simplified. At the same time, the robustness of the recognition system is increased. A fast learning method based on the error recognition case is proposed. The debugging work of identifying the error case is very heavy. In this paper, a learning method based on the error case is proposed. We can quickly learn new combinatorial mapping knowledge from the error case tag data provided by us and use this knowledge as a complementary branch of the system to avoid the same recognition errors. Based on the above method, a recognition system is implemented. The experimental results show that the on-line mathematical formula handwritten recognition method proposed in this paper has a higher recognition rate and better robustness.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 岳品瑩;;2014上半年互聯(lián)網(wǎng)教育綜合實(shí)力排行榜[J];互聯(lián)網(wǎng)周刊;2014年12期

2 楊現(xiàn)民;;信息時(shí)代智慧教育的內(nèi)涵與特征[J];中國(guó)電化教育;2014年01期

3 陳臨強(qiáng);李云霞;沈俊;;聯(lián)機(jī)手寫數(shù)學(xué)公式識(shí)別系統(tǒng)的設(shè)計(jì)和實(shí)現(xiàn)[J];杭州電子科技大學(xué)學(xué)報(bào);2009年02期

4 虞瑾;丁曉青;;聯(lián)機(jī)手寫公式中字符的切分與識(shí)別[J];電視技術(shù);2007年S1期

相關(guān)碩士學(xué)位論文 前2條

1 劉旭升;基于規(guī)則流的初中代數(shù)問楲類人判卷研究與實(shí)現(xiàn)[D];電子科技大學(xué);2016年

2 陳顯玲;初中數(shù)學(xué)錯(cuò)誤模型研究及其在自動(dòng)判卷系統(tǒng)中的應(yīng)用[D];電子科技大學(xué);2016年

，

本文編號(hào)：1883936