【摘要】：智能交通系統(tǒng)已經(jīng)走進(jìn)人們的生活,它廣泛應(yīng)用于收費(fèi)站、停車場(chǎng)等諸多場(chǎng)景之中。車牌識(shí)別作為其中最為重要的部分成為一個(gè)研究熱點(diǎn),許多專家學(xué)者提出了優(yōu)秀的識(shí)別算法。目前車牌識(shí)別技術(shù)已經(jīng)相當(dāng)成熟,對(duì)清晰車牌有著較高的識(shí)別率,但是一旦圖像質(zhì)量有所下降,識(shí)別率將會(huì)大大降低。車牌識(shí)別系統(tǒng)主要分為三大部分:車輛檢測(cè),車牌獲取和字符識(shí)別。本文將對(duì)此展開深入研究。車輛檢測(cè)部分,研究高效率的車輛檢測(cè)算法。本文采用基于卷積神經(jīng)網(wǎng)絡(luò)的車輛檢測(cè)算法,實(shí)現(xiàn)了從原始視頻圖片中自動(dòng)截取并保存車輛圖片,極大的降低了訓(xùn)練樣本的獲取成本。車牌獲取部分,研究了圖像灰度化,直方圖均衡化,去均值以及車牌傾斜校正等預(yù)處理操作,通過(guò)預(yù)處理可以降低干擾因素,突出車牌有用信息,便于后續(xù)的識(shí)別。該車牌獲取器可以方便快捷的從車輛圖片中截取高質(zhì)量的車牌圖片。車牌算法的識(shí)別結(jié)果對(duì)于手動(dòng)標(biāo)點(diǎn)情況過(guò)于敏感,標(biāo)點(diǎn)位置偏差極大的降低了車牌分割和識(shí)別效果。本文研究了兩套標(biāo)點(diǎn)優(yōu)化算法,根據(jù)用戶標(biāo)點(diǎn)和圖像信息,算法自動(dòng)矯正車牌標(biāo)點(diǎn),進(jìn)一步提高車牌的分割效果,最終提高車牌識(shí)別率和識(shí)別結(jié)果穩(wěn)定性。車牌字符識(shí)別部分,研究了多幀字符識(shí)別算法。對(duì)于數(shù)字字母車牌字符,先通過(guò)稀疏自編碼器提取字符的稀疏特征,再由支持向量機(jī)完成識(shí)別工作。對(duì)于中文字符,則通過(guò)費(fèi)希爾判別準(zhǔn)則的字典學(xué)習(xí)提取字符的殘差信息,再利用softmax完成中文字符的識(shí)別。不同于常見(jiàn)的單幀車牌識(shí)別算法,本文利用車牌在監(jiān)控視頻不同幀中的多張圖片共同參與識(shí)別,充分利用多幀圖像間的相對(duì)信息和自身的圖像信息。在單幀車牌識(shí)別的基礎(chǔ)上設(shè)計(jì)兩套多幀識(shí)別算法,分別為結(jié)果融合型多幀識(shí)別算法和特征融合型多幀識(shí)別算法。試驗(yàn)結(jié)果表明多幀識(shí)別對(duì)于較模糊車牌有著更高的識(shí)別率。
[Abstract]:Intelligent Transportation system (its) has come into people's life, it is widely used in many scenes such as toll station, parking lot and so on. As the most important part of license plate recognition, many experts and scholars put forward excellent recognition algorithms. At present, the license plate recognition technology is quite mature, and has a high recognition rate for clear license plate, but once the image quality is reduced, the recognition rate will be greatly reduced. License plate recognition system is mainly divided into three parts: vehicle detection, license plate acquisition and character recognition. This article will carry on the thorough research to this. In the part of vehicle detection, the efficient vehicle detection algorithm is studied. In this paper, the vehicle detection algorithm based on convolution neural network is used to automatically capture and save the vehicle images from the original video images, which greatly reduces the cost of obtaining training samples. In the part of license plate acquisition, the preprocessing operations such as image grayscale, histogram equalization, de-mean and license plate tilt correction are studied. Through the preprocessing, the interference factors can be reduced, the useful information of license plate can be highlighted, and the subsequent recognition can be facilitated. The license plate acquirer can conveniently and quickly intercept the high quality license plate image from the vehicle image. The recognition result of license plate algorithm is too sensitive to manual punctuation, and the deviation of punctuation position greatly reduces the effect of license plate segmentation and recognition. In this paper, two sets of punctuation optimization algorithms are studied. According to the user punctuation and image information, the algorithm automatically corrects the license plate punctuation, further improves the segmentation effect of the license plate, and finally improves the recognition rate and the stability of the recognition result. In the part of character recognition of license plate, multi-frame character recognition algorithm is studied. For the characters of alphabetical license plate, the sparse features are extracted by sparse self-encoder, and then the recognition is completed by support vector machine (SVM). For Chinese characters, the residual information of characters is extracted from the dictionary of Fisher criterion, and the recognition of Chinese characters is completed by softmax. Different from the common single-frame license plate recognition algorithm, this paper uses multiple images in different frames of the surveillance video to participate in the recognition, and makes full use of the relative information between the multi-frame images and their own image information. On the basis of single frame license plate recognition, two sets of multi-frame recognition algorithms are designed, one is result fusion multi-frame recognition algorithm and the other is feature fusion multi-frame recognition algorithm. Experimental results show that multi-frame recognition has a higher recognition rate than fuzzy license plate.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U495;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 程振彪;;我國(guó)應(yīng)更加重視燃料電池汽車發(fā)展(連載三)[J];汽車科技;2017年01期

2 曲景影;孫顯;高鑫;;基于CNN模型的高分辨率遙感圖像目標(biāo)識(shí)別[J];國(guó)外電子測(cè)量技術(shù);2016年08期

3 劉操;鄭宏;黎曦;余典;;基于多通道融合HOG特征的全天候運(yùn)動(dòng)車輛檢測(cè)方法[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2015年08期

4 金立生;王巖;劉景華;王亞麗;鄭義;;基于Adaboost算法的日間前方車輛檢測(cè)[J];吉林大學(xué)學(xué)報(bào)(工學(xué)版);2014年06期

5 謝永祥;董蘭芳;;復(fù)雜背景下基于HSV空間和模板匹配的車牌識(shí)別方法研究[J];圖學(xué)學(xué)報(bào);2014年04期

6 劉洋;王海暉;向云露;盧培磊;;基于改進(jìn)的Adaboost算法和幀差法的車輛檢測(cè)方法[J];華中科技大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年S1期

7 薛丹;孫萬(wàn)蓉;李京京;賈海龍;楊子峰;王政;;一種基于SVM的改進(jìn)車牌識(shí)別算法[J];電子科技;2013年11期

8 咼潤(rùn)華;蘇婷婷;馬曉偉;;BP神經(jīng)網(wǎng)絡(luò)聯(lián)合模板匹配的車牌識(shí)別系統(tǒng)[J];清華大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年09期

9 宋曉琳;王文濤;張偉偉;;基于LBP紋理和改進(jìn)Camshift算子的車輛檢測(cè)與跟蹤[J];湖南大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年08期

10 吳麗麗;余春艷;;基于Sobel算子和Radon變換的車牌傾斜校正方法[J];計(jì)算機(jī)應(yīng)用;2013年S1期

相關(guān)碩士學(xué)位論文 前3條

1 姚欽文;基于卷積神經(jīng)網(wǎng)絡(luò)的車臉識(shí)別研究[D];浙江大學(xué);2016年

2 陳明明;視頻中的車輛檢測(cè)及車型識(shí)別研究[D];南京郵電大學(xué);2013年

3 洪梅;基于多幀字符識(shí)別的車牌識(shí)別系統(tǒng)的研究[D];南京理工大學(xué);2012年

，

本文編號(hào)：2204194