本文選題：視頻煙霧檢測(cè) + 前景分割��； 參考：《太原理工大學(xué)》2017年碩士論文

【摘要】：火災(zāi)是一種突發(fā)性高、破壞力強(qiáng)的自然災(zāi)害,生產(chǎn)生活中用火不慎也會(huì)對(duì)人的生命財(cái)產(chǎn)安全造成嚴(yán)重威脅�；馂�(zāi)初期煙霧先于火焰出現(xiàn),煙霧探測(cè)能夠?yàn)槿藛T疏散和撲救火災(zāi)爭(zhēng)取更多寶貴的時(shí)間。因此,火災(zāi)煙霧探測(cè)已成為當(dāng)今社會(huì)迫切需要解決的課題。目前,視頻煙霧檢測(cè)因其高效性、非接觸性、可嵌入性等優(yōu)點(diǎn)而成為新的研究方向。為了提升煙霧視頻檢測(cè)系統(tǒng)的高效性和可靠性,論文圍繞煙霧視頻檢測(cè)技術(shù)的三個(gè)階段——前景目標(biāo)檢測(cè)階段、煙霧特征提取階段、煙霧識(shí)別階段進(jìn)行研究,提出了融合圖像分離和特征分析的煙霧檢測(cè)算法。(1)前景目標(biāo)檢測(cè):針對(duì)前景目標(biāo)檢測(cè),提出了基于ICA和GBVS的煙霧圖像分離檢測(cè)算法,改善了傳統(tǒng)高斯混合模型所需樣本量大且采樣時(shí)間跨度長(zhǎng)這一問題。該算法在煙霧分離階段先利用ICA煙霧前景初步分離煙霧模型得到初步煙霧前景,然后通過GBVS提取圖像多通道、多尺度的底層特征得到煙霧前景顯著區(qū)域,最后根據(jù)顏色和紋理的特征組合進(jìn)行直方圖匹配來識(shí)別煙霧。實(shí)驗(yàn)結(jié)果表明,該算法在煙霧前景分離階段將ICA和GBVS相結(jié)合,有效縮減了煙霧前景區(qū)域的范圍,提取的煙霧區(qū)域小而集中,ROC曲線顯示該算法整體性能表現(xiàn)優(yōu)秀。(2)煙霧特征提取:針對(duì)煙霧特征提取,提出了基于多維紋理分析的煙霧特征提取檢測(cè)算法。傳統(tǒng)煙霧視頻的線性動(dòng)態(tài)系統(tǒng)(LDS)僅采用亮度值作為圖像信息,忽視了其他信息如彩色視頻的RGB信息,并且由于密集采樣導(dǎo)致計(jì)算量較大。該算法首先經(jīng)過煙霧顏色過濾和背景差分預(yù)處理得到煙霧候選區(qū)域,避免出現(xiàn)密集采樣,降低了計(jì)算量;然后在多維圖像塊中新增RGB和HOG特征,增加了圖像塊的維度;最后基于對(duì)多維圖像數(shù)據(jù)的高階分解,分析煙霧視頻的動(dòng)態(tài)紋理特征。由于采用滑動(dòng)時(shí)間窗,可以確定畫面中煙霧的確切位置和煙霧發(fā)生的具體時(shí)間。實(shí)驗(yàn)以檢測(cè)率為評(píng)價(jià)指標(biāo),采用多元比較的方法,結(jié)果表明,該算法提高了動(dòng)態(tài)紋理特征分析的可靠性,同時(shí)計(jì)算量較小。(3)煙霧識(shí)別:針對(duì)煙霧識(shí)別,提出了基于廣義熵模糊神經(jīng)網(wǎng)絡(luò)的煙霧視頻圖像聚類檢測(cè)算法。該算法首先利用基于ICA和GBVS的煙霧圖像分離算法得到煙霧前景,然后提取基于高階線性動(dòng)態(tài)系統(tǒng)(h-LDS)的多維動(dòng)態(tài)紋理特征,最后利用一種廣義熵模糊神經(jīng)網(wǎng)絡(luò)對(duì)特征進(jìn)行訓(xùn)練和分類。實(shí)驗(yàn)結(jié)果表明,基于廣義熵模糊神經(jīng)網(wǎng)絡(luò)的煙霧視頻圖像聚類算法有較高的聚類正確率,且訓(xùn)練誤差較小。
[Abstract]:Fire is a kind of natural disaster with high burst and strong destructive power. The careless use of fire in production and life will also pose a serious threat to the safety of human life and property. Smoke appears before flame in the early stage of fire, and smoke detection can buy more valuable time for evacuation and fire fighting. Therefore, fire smoke detection has become an urgent problem to be solved in today's society. At present, video smoke detection has become a new research direction because of its high efficiency, non-contact, embeddability and other advantages. In order to improve the efficiency and reliability of smoke video detection system, this paper focuses on three stages of smoke video detection technology: foreground target detection stage, smoke feature extraction stage, smoke recognition stage. A smoke detection algorithm based on ICA and GBVS is proposed. (1) foreground target detection: aiming at foreground target detection, a smoke image separation detection algorithm based on ICA and GBVS is proposed. The problem of large sample size and long sampling time span for traditional Gao Si hybrid model is improved. At the stage of smoke separation, the ICA smoke foreground model is used to obtain the initial smoke foreground, and then GBVS is used to extract the multi-channel image, and the multi-scale bottom feature is used to obtain the significant region of smoke foreground. Finally, the smoke is identified by histogram matching according to the combination of color and texture. The experimental results show that the ICA and GBVS are combined in the stage of smoke foreground separation, and the range of smoke foreground region is reduced effectively. The extracted smoke region is small and concentrated ROC curve shows that the whole performance of the algorithm is excellent. (2) smoke feature extraction: for smoke feature extraction, a multi-dimensional texture analysis based smoke feature extraction detection algorithm is proposed. The traditional linear dynamic system of smoke video (LDS) only uses luminance value as image information, neglecting other information such as RGB information of color video, and because of dense sampling, the computation is large. Firstly, smoke candidate regions are obtained by smoke color filtering and background differential preprocessing to avoid dense sampling and reduce computational complexity, then RGB and HOG features are added to multi-dimensional image blocks, and the dimension of image blocks is increased. Finally, based on the higher order decomposition of multidimensional image data, the dynamic texture features of smoke video are analyzed. Because of the sliding time window, the exact location and time of smoke in the screen can be determined. The experiment takes the detection rate as the evaluation index and adopts the method of multivariate comparison. The results show that the algorithm improves the reliability of the dynamic texture feature analysis, and the computation is small. (3) smoke recognition: for smoke recognition, A clustering detection algorithm for smoke video images based on generalized entropy fuzzy neural network is proposed. Firstly, the smoke image separation algorithm based on ICA and GBVS is used to obtain the foreground of smoke, and then the multi-dimensional dynamic texture feature based on high-order linear dynamic system (h-LDS) is extracted. Finally, a generalized entropy fuzzy neural network is used to train and classify the features. The experimental results show that the clustering algorithm based on generalized entropy fuzzy neural network has higher clustering accuracy and less training error.
【學(xué)位授予單位】：太原理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張煒程;李佩;高陳強(qiáng);;森林背景下基于自適應(yīng)區(qū)域生長(zhǎng)法的煙霧檢測(cè)[J];重慶郵電大學(xué)學(xué)報(bào)(自然科學(xué)版);2016年01期

2 賈陽;林高華;王進(jìn)軍;方俊;張永明;;基于顯著性檢測(cè)和高斯混合模型的早期視頻煙霧分割算法[J];計(jì)算機(jī)工程;2016年02期

3 史玉坤;仲貞;張德馨;楊金鋒;;利用多特征判別的煙霧檢測(cè)方法研究[J];信號(hào)處理;2015年10期

4 董蘭芳;余家奎;;基于圖像分離的視頻煙霧檢測(cè)方法[J];計(jì)算機(jī)工程;2015年09期

5 方帥;祁林娟;于磊;;多特征分析的視頻煙霧檢測(cè)方法[J];計(jì)算機(jī)工程與應(yīng)用;2016年13期

6 潘廣貞;元琴;樊彩霞;喬慧芬;王志豪;;基于混合高斯模型和幀差法的吸煙檢測(cè)算法[J];計(jì)算機(jī)工程與設(shè)計(jì);2015年05期

7 錢生;陳宗海;林名強(qiáng);張陳斌;;基于條件隨機(jī)場(chǎng)和圖像分割的顯著性檢測(cè)[J];自動(dòng)化學(xué)報(bào);2015年04期

8 李敏;劉軻;羅惠瓊;王添?xiàng)?趙麒;;基于混合高斯模型的異常檢測(cè)算法改進(jìn)[J];計(jì)算機(jī)應(yīng)用與軟件;2014年06期

9 張燕君;邵明亮;;基于小波紋理特性分析的視頻煙霧檢測(cè)算法研究[J];光學(xué)技術(shù);2013年04期

10 王濤;劉淵;謝振平;;基于自適應(yīng)神經(jīng)模糊推理系統(tǒng)的視頻煙霧檢測(cè)[J];計(jì)算機(jī)工程;2011年23期

相關(guān)博士學(xué)位論文 前2條

1 康鋒;基于視覺特征的早期農(nóng)林火災(zāi)檢測(cè)方法的基礎(chǔ)研究[D];浙江大學(xué);2010年

2 王勇;基于統(tǒng)計(jì)方法的運(yùn)動(dòng)目標(biāo)檢測(cè)與跟蹤技術(shù)研究[D];華中科技大學(xué);2009年

相關(guān)碩士學(xué)位論文 前7條

1 梁何;視覺顯著性在視頻目標(biāo)識(shí)別中的應(yīng)用研究[D];南京郵電大學(xué);2016年

2 胡傳秀;基于融合特征的火焰和煙霧識(shí)別方法研究[D];華東交通大學(xué);2016年

3 馮海永;基于超像素的顯著性區(qū)域檢測(cè)研究[D];江南大學(xué);2016年

4 岳姣姣;基于多特征融合的林火煙霧檢測(cè)算法研究[D];燕山大學(xué);2016年

5 鄧林;視頻煙霧檢測(cè)算法研究[D];電子科技大學(xué);2015年

6 李智明;基于改進(jìn)FastICA算法的混合語音盲分離[D];上海交通大學(xué);2015年

7 游勇;基于視頻動(dòng)態(tài)特征分析的煙霧檢測(cè)算法研究[D];西南交通大學(xué);2013年

，

本文編號(hào)：2113669