本文選題：高速公路　切入點(diǎn)：視頻圖像　出處：《北京交通大學(xué)》2017年碩士論文

【摘要】：惡劣的路面條件是造成道路交通事故的重要誘因,當(dāng)車輛在雨、雪、霧等惡劣天氣下行駛時(shí),極易發(fā)生車輛打滑側(cè)翻等重特大交通事故。因此,準(zhǔn)確快速地對(duì)路面狀態(tài)進(jìn)行檢測(cè)識(shí)別,并在道路條件惡劣的情況下做出及時(shí)的預(yù)警和反應(yīng),對(duì)高速公路的高效安全運(yùn)營(yíng)具有重大現(xiàn)實(shí)意義。隨著視頻圖像處理技術(shù)的成熟發(fā)展及高速公路全程視頻監(jiān)控系統(tǒng)的普及,監(jiān)控?cái)z像機(jī)已成為主要的交通監(jiān)測(cè)設(shè)施。因此,利用視頻圖像技術(shù)識(shí)別路面狀態(tài)已成為當(dāng)前研究熱點(diǎn)。然而,對(duì)于混合路面狀態(tài)的識(shí)別,以及不同光照條件下的路面狀態(tài)識(shí)別是兩個(gè)亟需重點(diǎn)解決的難題。本論文提出利用視頻圖像技術(shù)檢測(cè)路面狀態(tài),并開發(fā)可滿足不同路段需求的路面狀態(tài)檢測(cè)算法。具體研究?jī)?nèi)容如下:(1)論文首先將路面狀態(tài)分為干燥、潮濕、積雪、結(jié)冰、積水5種狀態(tài),根據(jù)原始圖像大小,制定了圖像分塊原則,提取單一狀態(tài)的圖像塊構(gòu)建了路面狀態(tài)圖像庫(kù),保證了樣本的質(zhì)量和純度。(2)采用三階顏色矩法提取了 9維顏色特征向量,采用灰度共生矩陣法提取能量、熵、相關(guān)性、對(duì)比度4個(gè)紋理特征量,基于該13維圖像特征向量,形成了路面狀態(tài)特征數(shù)據(jù)庫(kù)。(3)提出了基于SVM(支持向量機(jī))的路面狀態(tài)視頻識(shí)別方法,為提升該算法的普適性的識(shí)別精度,采用網(wǎng)格搜索算法優(yōu)化了 SVM中的核函數(shù)因子C和懲罰因子g。(4)建立路面狀態(tài)圖像分類模型,首先,利用參數(shù)尋優(yōu)的SVM分類器對(duì)多組不同樣本量進(jìn)行了訓(xùn)練,得到多組路面狀態(tài)圖像分類模型;接著,對(duì)上述多組分類模型進(jìn)行性能測(cè)試識(shí)別,甄選出分類效果最佳的模型。(5)依托搭建的實(shí)驗(yàn)系統(tǒng)及源于不同采集方式的視頻樣本,對(duì)大量不同環(huán)境、未經(jīng)訓(xùn)練的實(shí)際路面狀態(tài)原始圖像進(jìn)行了分塊識(shí)別驗(yàn)證。實(shí)驗(yàn)結(jié)果表明:基于SVM尋優(yōu)分類器和視頻圖像分塊識(shí)別的方法科學(xué)可行,網(wǎng)格搜索尋優(yōu)算法下的路面狀態(tài)分類模型對(duì)單一狀態(tài)下的樣本識(shí)別準(zhǔn)確率90%以上,對(duì)混和路面樣本的識(shí)別準(zhǔn)確率85%以上。有效解決了混合狀態(tài)路面狀態(tài)和不同光照條件下路面狀態(tài)的識(shí)別難題。
[Abstract]:Bad road conditions is an important cause of road traffic accidents, when the vehicle is in the rain, snow, fog and other inclement weather driving, prone to skid rollover and other serious traffic accidents. Therefore, accurate detection of road conditions, and make early warning and timely response in a bad way the case is of great practical significance to the safe operation of the highway, and highway. With the popularity of mature development of the whole video monitoring system of video image processing technology, surveillance cameras have become the main traffic monitoring facilities. Therefore, the use of video image recognition technology of pavement state has become a hotspot of current research. However, for the identification of mixed pavement state, and the road condition identification under different illumination conditions are two need to focus on resolving the problem. This paper presents the use of video image detection technology Road conditions, and development can meet the needs of different sections of pavement state detection algorithm. The specific contents are as follows: (1) the first road state is divided into dry, wet, snow, ice, water in 5 states, according to the size of the original image, formulated the principle of image block, image block extraction single state construction road condition image library, to ensure the quality and purity of the sample. (2) the extraction of 9 dimensional color feature vector by three order color moment method using gray level co-occurrence matrix method to extract energy, entropy, correlation, contrast 4 texture features, the 13 dimensional image feature vector based on the formation of the State Road feature database. (3) based on SVM (support vector machine) the road condition identification method for video, boosting the algorithm's general recognition accuracy, using the grid search algorithm to optimize the function of nuclear factor C g. and the penalty factor in SVM (4) established State road image classification model, firstly, using SVM classifier parameter optimization was trained on different samples, resulting in multiple sets of pavement state image classification model; then, test the performance of recognition to the group classification model, the selection of the best results of classification model. (5) based on the experimental setup and due to the different acquisition methods of video sample, a large number of different environment, without the state of actual pavement training image block identification. The experimental results show that the optimization method based on SVM classifier and video image block recognition is scientific and feasible, grid search algorithm under the pavement state classification model for sample identification under the single state accuracy rate above 90%, the recognition accuracy of the sample mixture pavement above 85%. Effective solution to the identification of mixed state of pavement state and under different illumination conditions of pavement state Difficult problems.

【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U418.6

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 蔡蕾;張德才;李斌;高蘭達(dá);汪林;劉清彬;;交通氣象埋入式路面狀況檢測(cè)器檢測(cè)方法研究[J];道路交通與安全;2014年06期

2 王麗君;淮永建;彭月橙;;基于葉片圖像多特征融合的觀葉植物種類識(shí)別[J];北京林業(yè)大學(xué)學(xué)報(bào);2015年01期

3 湯筠筠;郭忠印;李長(zhǎng)城;辛欣;韓暉;;基于路面摩擦因數(shù)的冬季典型路面狀態(tài)識(shí)別模型[J];中國(guó)公路學(xué)報(bào);2014年11期

4 張少博;全書海;石英;楊陽(yáng);李云路;程姝;;基于顏色矩的圖像檢索算法研究[J];計(jì)算機(jī)工程;2014年06期

5 齊志權(quán);王寶鋒;裴曉飛;馬國(guó)成;;基于路面特征參數(shù)和制動(dòng)防抱調(diào)節(jié)特征的路面識(shí)別方法[J];汽車工程;2014年03期

6 萬(wàn)劍;趙愷;王維鋒;;基于高維特征和RBF神經(jīng)網(wǎng)絡(luò)的濕滑道路圖像判別方法[J];交通信息與安全;2013年02期

7 范會(huì)敏;王浩;;模式識(shí)別方法概述[J];電子設(shè)計(jì)工程;2012年19期

8 劉新宇;黃德啟;;基于SVM分類器的道路濕滑圖像分類方法研究[J];武漢理工大學(xué)學(xué)報(bào)(交通科學(xué)與工程版);2011年04期

9 龔永罡;湯世平;;面向大數(shù)據(jù)的SVM參數(shù)尋優(yōu)方法[J];計(jì)算機(jī)仿真;2010年09期

10 李虹;馮彥輝;林君;;基于改進(jìn)BP神經(jīng)網(wǎng)絡(luò)的路面狀態(tài)識(shí)別研究[J];微計(jì)算機(jī)信息;2010年02期

相關(guān)博士學(xué)位論文 前2條

1 戴金波;基于視覺(jué)信息的圖像特征提取算法研究[D];吉林大學(xué);2013年

2 李虹;基于機(jī)器視覺(jué)路面狀態(tài)識(shí)別關(guān)鍵技術(shù)研究[D];吉林大學(xué);2009年

相關(guān)碩士學(xué)位論文 前2條

1 孫鑫;基于突變理論的高速公路交通安全風(fēng)險(xiǎn)耦合研究[D];北京交通大學(xué);2015年

2 王龍;圖像紋理特征提取及分類研究[D];中國(guó)海洋大學(xué);2014年

，

本文編號(hào)：1684445