發(fā)布時間：2018-03-14 15:21

  本文選題：紅外熱像儀　切入點：特征提取　出處：《南京理工大學》2017年碩士論文　論文類型：學位論文

【摘要】：紅外圖像反映了物體之間的溫度差異分布,所以紅外監(jiān)測技術(shù)具有遠距離不接觸和不受惡劣環(huán)境影響的特點,在軍事目標探測與預警、電力設(shè)備在線監(jiān)測和工業(yè)檢測等范圍都得到了較為廣泛的應(yīng)用。但紅外熱像儀大多數(shù)成像范圍窄且分辨率不高,一張圖像往往很難顯示較大目標物體的全部樣貌,給后續(xù)的圖像分析和被測目標的溫度趨勢判斷帶來了不便,而圖像拼接技術(shù)為其提供了一個較為合適的解決方案。論文首先對紅外熱像的拼接技術(shù)進行了研究,著重研究了紅外熱像的特征點提取與圖像配準,在此基礎(chǔ)上提出了相應(yīng)的改進算法。并設(shè)計了與之配套的紅外熱像采集與拼接系統(tǒng),通過實驗驗證了其實用性與可靠性。在圖像預處理的研究中,論文首先分析了紅外成像原理,對紅外圖像的去噪和增強技術(shù)進行了研究,并在此基礎(chǔ)上對紅外圖像進行仿真實驗與結(jié)果分析。在圖像配準的研究中,論文首先對三大配準方式進行了介紹與仿真說明,研究確定了在論文中使用基于特征的圖像配準方案;在特征點的提取中,對經(jīng)典的角點提取算法和局部不變特征提取算法進行了深入的研究,進行了編程實驗仿真與結(jié)果分析,并提出了一種改進的多尺度空間的特征點提取方式,通過編程實驗驗證了其可行性;在圖像的特征匹配中,對待匹配圖像之間的粗匹配和結(jié)合改進的RANSAC算法的特征點的提純方法進行研究,并用編程實驗驗證了其可行性。在圖像融合的研究中,論文對基本的融合方式進行了介紹,來完成圖像拼接的最后一步,并進行編程實驗生成兩幅圖的拼接圖。在軟件系統(tǒng)的設(shè)計中,論文根據(jù)之前對算法的研究成果,設(shè)計了一套基于紅外熱像儀的熱像采集與拼接系統(tǒng)。該系統(tǒng)不僅能完成熱像儀的控制、實時視頻的點溫度檢測和熱像的采集,還能夠選擇相應(yīng)的算法對熱像儀采集的圖像進行處理與拼接,給后續(xù)圖像的分析工作帶來了方便,具有一定的價值和意義。
[Abstract]:Infrared images reflect the distribution of temperature differences between objects, so infrared monitoring technology has the characteristics of remote non-contact and unaffected by adverse environment, in the detection and early warning of military targets, Power equipment on-line monitoring and industrial detection have been widely used. However, most infrared thermal imagers have narrow imaging range and low resolution, so it is difficult for an image to display the full appearance of a large target object. It brings inconvenience to the subsequent image analysis and the temperature trend judgment of the target under test, and the image stitching technology provides a more suitable solution for it. Firstly, the paper studies the technology of infrared thermal image stitching. In this paper, the feature point extraction and image registration of infrared thermal image are emphatically studied, and the corresponding improved algorithm is put forward, and a matching infrared thermal image acquisition and splicing system is designed. The practicability and reliability are verified by experiments. In the research of image preprocessing, firstly, the principle of infrared imaging is analyzed, and the denoising and enhancement techniques of infrared image are studied. On this basis, the infrared image simulation experiments and results analysis. In the image registration research, the paper first introduces the three major registration methods and simulation. In the feature point extraction, the classical corner extraction algorithm and the local invariant feature extraction algorithm are deeply studied, and the programming experiments and results analysis are carried out. An improved feature point extraction method in multi-scale space is proposed, and its feasibility is verified by programming experiments. The methods of coarse matching between matching images and feature points of improved RANSAC algorithm are studied, and the feasibility of this method is verified by programming experiments. In the research of image fusion, the basic fusion methods are introduced in this paper. In the software system design, according to the previous research results of the algorithm, A thermal image acquisition and splicing system based on infrared thermal imager is designed. The system can not only control the thermal imager, detect the point temperature of the real-time video, but also collect the thermal image. It can also select the corresponding algorithm to process and join the images collected by the thermal imager, which brings convenience to the analysis of the subsequent images, and has certain value and significance.
【學位授予單位】：南京理工大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP391.41

【相似文獻】

相關(guān)期刊論文 前10條

1 江勇;余紹霖;許燕;;大屏拼接技術(shù)在巡考領(lǐng)域的應(yīng)用[J];中國公共安全(綜合版);2012年08期

2 譚紹紅;;淺析大屏拼接技術(shù)及應(yīng)用[J];中國公共安全;2013年07期

3 ;大屏幕拼接技術(shù)縱觀[J];電化教育研究;1999年04期

4 張波;莊昆;;超窄邊弧形液晶拼接技術(shù)的研究與應(yīng)用[J];山東煤炭科技;2014年03期

5 盧光義;;大屏幕拼接技術(shù)及其應(yīng)用[J];電腦知識與技術(shù);2009年21期

6 ;大屏拼接技術(shù)的發(fā)展與應(yīng)用[J];中國安防;2013年06期

7 程岳寅;;淺談大屏拼接技術(shù)和產(chǎn)品[J];中國安防;2013年06期

8 吳淑芳;計算機輔助網(wǎng)格拼接技術(shù)[J];長春理工大學學報;2002年02期

9 王改香;;紅外拼接技術(shù)在電力設(shè)備檢測中的應(yīng)用[J];數(shù)字技術(shù)與應(yīng)用;2012年11期

10 張素,王小菊,徐乃平;一種基于小波分解的多圖象拼接技術(shù)[J];西北工業(yè)大學學報;2001年01期

相關(guān)會議論文 前5條

1 倪勇;黃永蒼;陳勇軍;陳旭;閆曉;李彬育;陳駒;;三疊大板梁拼接技術(shù)研究[A];中國鋼結(jié)構(gòu)協(xié)會鍋爐鋼結(jié)構(gòu)分會第十三屆年會論文集[C];2012年

2 程良駿;趙樂\，

本文編號：1611783