發(fā)布時(shí)間：2018-06-13 18:13

  本文選題：煤礦智能視頻監(jiān)控系統(tǒng) + 去霧除塵和同步去噪�。� 參考：《中國(guó)礦業(yè)大學(xué)》2013年博士論文

【摘要】：目前我國(guó)大多數(shù)的煤礦視頻監(jiān)控系統(tǒng)還主要停留在人工監(jiān)控階段，智能化煤礦視頻監(jiān)控系統(tǒng)是發(fā)展的必然趨勢(shì)。它可以自動(dòng)采集獲得視頻監(jiān)控圖像序列，進(jìn)行實(shí)時(shí)運(yùn)動(dòng)目標(biāo)檢測(cè)、識(shí)別和跟蹤，通過(guò)理解分析圖像畫面主動(dòng)發(fā)現(xiàn)違規(guī)行為、可疑目標(biāo)和潛在危險(xiǎn)，以快速合理的方式發(fā)出警報(bào)，指導(dǎo)啟動(dòng)相應(yīng)的聯(lián)動(dòng)控制措施。煤礦智能視頻監(jiān)控系統(tǒng)的實(shí)現(xiàn)，需要綜合運(yùn)用圖像處理、機(jī)器學(xué)習(xí)和計(jì)算機(jī)視覺等領(lǐng)域中的多項(xiàng)技術(shù)，本文對(duì)其中的四類關(guān)鍵技術(shù)進(jìn)行研究，具體工作包括： 為了對(duì)伴有隨機(jī)噪聲的煤礦霧塵圖像進(jìn)行清晰化處理，提出一種基于DCPBF的去霧除塵和同步去噪算法。推導(dǎo)建立煤礦霧塵降質(zhì)圖像退化模型；設(shè)計(jì)基于暗原色先驗(yàn)知識(shí)的環(huán)境光、粗略透射率估計(jì)方法與步驟；采用聯(lián)合雙邊濾波快速獲得精細(xì)透射率圖；依據(jù)圖像退化模型構(gòu)建正則化目標(biāo)函數(shù)，求取轉(zhuǎn)換圖像并進(jìn)行高斯雙邊濾波，獲得去霧除塵圖像且同步實(shí)現(xiàn)噪聲的有效去除。 針對(duì)相對(duì)靜止的煤礦視頻監(jiān)控環(huán)境背景，采用背景減除法進(jìn)行運(yùn)動(dòng)目標(biāo)檢測(cè)。提出基于聚類技術(shù)的自適應(yīng)背景建模與更新方法，利用改進(jìn)的FCM算法對(duì)像素灰度取值進(jìn)行聚類，自適應(yīng)選取不同個(gè)數(shù)的聚類構(gòu)建各像素背景模型，隨場(chǎng)景變化進(jìn)行聚類修改、添加和刪除完成背景更新。聯(lián)合背景差分信息、三幀差分信息和空間鄰域信息進(jìn)行前景檢測(cè)，通過(guò)改進(jìn)的OTSU方法自動(dòng)設(shè)置差分閾值。提出結(jié)合像素亮度和紋理特征的運(yùn)動(dòng)陰影檢測(cè)方法，依據(jù)在陰影覆蓋前后的灰度圖像中，像素具有亮度值相關(guān)性和紋理特征值不變性，實(shí)現(xiàn)運(yùn)動(dòng)陰影的檢測(cè)與去除。 將單目標(biāo)跟蹤看作為目標(biāo)和背景的在線分類問(wèn)題，選用線性SVM作為分類工具，提出一種添加樣本約簡(jiǎn)機(jī)制的FLSVMIL方法實(shí)現(xiàn)分類器在線更新，并提出基于FLSVMIL的單目標(biāo)跟蹤算法。由于可能受到無(wú)效歷史信息的干擾，并且難以處理樣本集非線性可分的問(wèn)題，提出基于LSVMSE的單目標(biāo)跟蹤算法，采用集成分類器進(jìn)行運(yùn)動(dòng)目標(biāo)跟蹤。 根據(jù)煤礦智能視頻監(jiān)控系統(tǒng)中多目標(biāo)跟蹤的任務(wù)需求，提出基于UKF-MHT的多目標(biāo)跟蹤算法。設(shè)計(jì)算法的基本框架，確定關(guān)鍵步驟的處理方法，其中包括跟蹤門設(shè)置、目標(biāo)預(yù)測(cè)值與觀測(cè)值的數(shù)據(jù)匹配、航跡評(píng)價(jià)與刪除、航跡聚類和m-best假設(shè)的產(chǎn)生以及目標(biāo)狀態(tài)的預(yù)測(cè)更新。在自適應(yīng)跟蹤修正階段，針對(duì)由目標(biāo)短暫丟失、粘連和分裂可能引起的三類跟蹤錯(cuò)誤，，設(shè)計(jì)具體的判別策略和修正方法。
[Abstract]:At present, most coal mine video surveillance systems in China are still mainly in the stage of manual monitoring, and intelligent coal mine video surveillance system is an inevitable trend of development. It can automatically capture and obtain video surveillance image sequences, detect, identify and track moving objects in real time, actively discover irregularities, suspicious targets and potential dangers through understanding and analyzing images, and issue warnings in a rapid and reasonable manner. Guide to initiate the corresponding linkage control measures. In order to realize the intelligent video surveillance system in coal mine, it is necessary to comprehensively apply many techniques in the fields of image processing, machine learning and computer vision. In this paper, four kinds of key technologies are studied. The main work includes: in order to clear the dust image with random noise, a DCPBF based de-fogging and simultaneous de-noising algorithm is proposed. The degradation model of degraded image of coal mine fog dust is established, the environmental light and rough transmittance estimation method and steps based on the prior knowledge of dark primary color are designed, and the fine transmittance map is obtained quickly by combined bilateral filtering. According to the image degradation model, the regularization objective function is constructed, the converted image is obtained and Gao Si bilateral filtering is carried out, and the de-fogging and dedusting image is obtained, and the effective noise removal is realized synchronously. The background subtraction method is used to detect moving targets against the background of relatively static coal mine video surveillance environment. An adaptive background modeling and updating method based on clustering technology is proposed. The improved FCM algorithm is used to cluster the pixel grayscale, and the background model of each pixel is constructed by selecting different numbers of clustering adaptively, and the background model is modified with the change of scene. Add and delete complete background updates. The background differential information, the three frame differential information and the spatial neighborhood information are combined to detect the foreground, and the differential threshold is automatically set by the improved Otsu method. A method of moving shadow detection combined with pixel luminance and texture features is proposed. According to the correlation of luminance values and the invariance of texture feature values in gray images before and after shadow coverage the detection and removal of moving shadows are realized. Considering single target tracking as an online classification problem of target and background, linear SVM is used as a classification tool. A FLSVMIL method with sample reduction mechanism is proposed to update the classifier online, and a single target tracking algorithm based on FLSVMIL is proposed. Because of the disturbance of invalid historical information and the difficulty of dealing with the nonlinear separable problem of the sample set, a single target tracking algorithm based on LSVMSE is proposed, and an integrated classifier is used for moving target tracking. According to the task requirement of multi-target tracking in intelligent video surveillance system of coal mine, a multi-target tracking algorithm based on UKF-MHT is proposed. Design the basic framework of the algorithm, determine the key steps of the processing methods, including tracking gate setting, target prediction and observation data matching, track evaluation and deletion, Track clustering, m-best hypothesis generation and target state prediction update. In the stage of adaptive tracking correction, a specific discriminant strategy and correction method are designed for three kinds of tracking errors which may be caused by the transient loss, adhesion and splitting of the target.
【學(xué)位授予單位】：中國(guó)礦業(yè)大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2013
【分類號(hào)】：TD76;TP273

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 孫抗;汪渤;周志強(qiáng);鄭智輝;;基于雙邊濾波的實(shí)時(shí)圖像去霧技術(shù)研究[J];北京理工大學(xué)學(xué)報(bào);2011年07期

2 張立;孟相如;馬志強(qiáng);周華;;邊界偏轉(zhuǎn)覆蓋增量支持向量機(jī)[J];北京郵電大學(xué)學(xué)報(bào);2010年04期

3 閆愛云;李海朋;李晶皎;王驕;;視頻運(yùn)動(dòng)目標(biāo)提取的實(shí)現(xiàn)[J];東北大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年11期

4 湯軍;孫偉;;彈道目標(biāo)跟蹤的自適應(yīng)多維分配相關(guān)算法[J];彈道學(xué)報(bào);2011年02期

5 郭烈;張明恒;李琳輝;趙一兵;;一種基于支持向量機(jī)的行人識(shí)別方法研究[J];大連理工大學(xué)學(xué)報(bào);2011年04期

6 熊子源;徐振海;張亮;吳迪軍;肖順平;;基于聚類算法的最優(yōu)子陣劃分方法研究[J];電子學(xué)報(bào);2011年11期

7 甘明剛;陳杰;劉勁;王亞楠;;一種基于三幀差分和邊緣信息的運(yùn)動(dòng)目標(biāo)檢測(cè)方法[J];電子與信息學(xué)報(bào);2010年04期

8 陳香蘋;李生紅;蘇波;金波;;基于圖像噪聲分析的計(jì)算機(jī)生成圖像檢測(cè)算法[J];光電子.激光;2010年02期

9 劉貴喜;馬濤;陳石磊;;應(yīng)用最小偏度采樣的UPF算法[J];光學(xué)精密工程;2008年04期

10 陳龍;郭寶龍;畢娟;朱娟娟;;基于聯(lián)合雙邊濾波的單幅圖像去霧算法[J];北京郵電大學(xué)學(xué)報(bào);2012年04期

相關(guān)博士學(xué)位論文 前10條

1 許志遠(yuǎn);霧天降質(zhì)圖像增強(qiáng)方法研究及DSP實(shí)現(xiàn)[D];大連海事大學(xué);2010年

2 湯義;智能交通系統(tǒng)中基于視頻的行人檢測(cè)與跟蹤方法的研究[D];華南理工大學(xué);2010年

3 劉獻(xiàn)如;視頻圖像序列目標(biāo)跟蹤算法及其應(yīng)用研究[D];中南大學(xué);2011年

4 常甜甜;支持向量機(jī)學(xué)習(xí)算法若干問(wèn)題的研究[D];西安電子科技大學(xué);2010年

5 胡學(xué)友;霧天降質(zhì)圖像的增強(qiáng)復(fù)原算法研究[D];安徽大學(xué);2011年

6 周磊;基于注意機(jī)制的煤礦監(jiān)控圖像知覺編組研究[D];中國(guó)礦業(yè)大學(xué);2010年

7 厲丹;視頻目標(biāo)檢測(cè)與跟蹤算法及其在煤礦中應(yīng)用的研究[D];中國(guó)礦業(yè)大學(xué);2011年

8 王愛平;視頻目標(biāo)跟蹤技術(shù)研究[D];國(guó)防科學(xué)技術(shù)大學(xué);2011年

9 姚志均;目標(biāo)跟蹤系統(tǒng)中的魯棒性研究[D];華中科技大學(xué);2012年

10 曲昭偉;混合交通視頻檢測(cè)算法研究[D];吉林大學(xué);2009年

本文編號(hào)：2014947