本文選題：紅外成像 + 細節(jié)增強　； 參考：《華中科技大學》2016年碩士論文

【摘要】：紅外成像技術是一種采集環(huán)境中的紅外輻射信號并將其以二維圖像形式表現(xiàn)出來的信號處理技術。由于紅外線具有抗干擾能力強、隱蔽性好、保密性高等優(yōu)勢,紅外成像技術在醫(yī)療檢測、工業(yè)探傷以及軍事目標偵察等領域有著廣泛的應用。但是目前的紅外圖像普遍存在對比度低的缺點,不僅嚴重影響其本身的圖像質量,還不利于后續(xù)的圖像目標識別與追蹤等,因此必須對其進行增強處理。然而傳統(tǒng)的紅外圖像增強技術盡管可以提升圖像的對比度與信噪比,但也會丟失圖像中的細節(jié),并且算法的場景適應性差,對紅外成像技術的應用產(chǎn)生極大的限制。因此,本文從紅外圖像的基本特征出發(fā),通過分析現(xiàn)有紅外圖像增強算法存在的優(yōu)缺點,提出一種基于引導濾波器的紅外圖像自適應細節(jié)增強算法。該算法總共包含四個步驟:首先采用引導濾波器將原始圖像分成包含低頻背景的基本層和包含高頻細節(jié)的細節(jié)層;然后采用基于圖像統(tǒng)計特性的自適應直方圖投影算法處理基本層,以實現(xiàn)圖像對比度的改善,并提升算法的場景適應性;同時采用基于用自適應增益校正處理細節(jié)層,以增強圖像細節(jié)并抑制圖像噪聲;最后利用經(jīng)過處理的基本層和細節(jié)層構建最終的輸出圖像。多場景下真實紅外圖像數(shù)據(jù)的實驗結果表明,與現(xiàn)有的同類算法相比,本文算法能有效處理多種場景下的紅外圖像,并且輸出圖像的對比度好、細節(jié)豐富,具有良好的圖像質量。算法的主觀和客觀評價標準結果均表明其在紅外圖像動態(tài)范圍壓縮、對比度提升與細節(jié)保留方面有良好的處理性能以及出色的場景適應性。
[Abstract]:Infrared imaging technology is a kind of signal processing technology which collects infrared radiation signals in environment and displays them as two-dimensional images. Infrared imaging technology has been widely used in the fields of medical detection, industrial flaw detection and military target reconnaissance because of its advantages of strong anti-interference ability, good concealment, high confidentiality and so on. However, the current infrared images generally have the shortcomings of low contrast, which not only seriously affect their own image quality, but also is not conducive to the subsequent image recognition and tracking, so it must be enhanced processing. However, although the traditional infrared image enhancement technology can improve the contrast and signal-to-noise ratio of the image, it will also lose the details of the image, and the algorithm has poor adaptability to the scene, which greatly limits the application of the infrared imaging technology. Therefore, based on the basic characteristics of infrared image and the advantages and disadvantages of existing infrared image enhancement algorithms, an adaptive detail enhancement algorithm for infrared image based on guided filter is proposed in this paper. The algorithm consists of four steps: firstly, the original image is divided into basic layer with low frequency background and detail layer with high frequency details by using a bootstrap filter; Then the adaptive histogram projection algorithm based on the statistical characteristics of the image is used to process the basic layer to improve the image contrast and enhance the scene adaptability of the algorithm. Finally, the final output image is constructed by using the processed basic layer and the detail layer. The experimental results of real infrared image data in multi-scene show that the proposed algorithm can deal with infrared images in many scenarios effectively, and the output image has good contrast and rich details. Good image quality. The subjective and objective evaluation results show that the algorithm has good processing performance and excellent scene adaptability in infrared image dynamic range compression, contrast enhancement and detail retention.
【學位授予單位】：華中科技大學
【學位級別】：碩士
【學位授予年份】：2016
【分類號】：TP391.41

【參考文獻】

相關期刊論文 前8條

1 賈宏光;吳澤鵬;朱明超;宣明;劉慧;;基于廣義線性運算和雙邊濾波的紅外圖像增強[J];光學精密工程;2013年12期

2 朱道廣;隋修寶;朱才高;劉寧;陳錢;;基于多尺度的高動態(tài)紅外圖像增強算法[J];紅外技術;2013年08期

3 劉斌;王霞;金偉其;陳艷;劉崇亮;劉秀;;Infrared image detail enhancement based on local adaptive gamma correction[J];Chinese Optics Letters;2012年02期

4 金偉其;劉斌;范永杰;王霞;徐超;陳艷;劉崇亮;夏潤秋;;紅外圖像細節(jié)增強技術研究進展[J];紅外與激光工程;2011年12期

5 劉斌;金偉其;王嶺雪;陳艷;劉崇亮;劉秀;;基于空域和頻域處理的紅外圖像細節(jié)增強算法[J];紅外技術;2011年08期

6 范永杰;金偉其;劉斌;劉崇亮;;FLIR公司熱成像細節(jié)增強DDE技術的分析[J];紅外技術;2010年03期

7 毛玉星;孫才新;徐澤宇;;一種紅外圖像細節(jié)增強算法[J];計算機工程與應用;2009年31期

8 宋巖峰;邵曉鵬;徐軍;;基于雙平臺直方圖的紅外圖像增強算法[J];紅外與激光工程;2008年02期

相關碩士學位論文 前3條

1 劉婷婷;紅外圖像細節(jié)增強算法與實現(xiàn)的研究[D];電子科技大學;2014年

2 朱才高;紅外圖像增強算法研究及其DSP實現(xiàn)[D];南京理工大學;2014年

3 涂志強;一種自適應雙平臺直方圖紅外圖像增強算法及其FPGA實現(xiàn)[D];華中科技大學;2012年

，

本文編號：1838564