基于Matlab的红外图像处理算法研究

目录摘要.................................................................................................................................IAbstract.........................................................................................................................II第1章绪论..................................................................................................................11.1研究背景和意义...............................................................................................11.2国内外研究现状及发展趋势...........................................................................21.3论文的主要内容安排.......................................................................................4第2章红外图像的介绍..............................................................................................62.1红外热成像技术的介绍及其特征...................................................................62.2红外图像产生原理以及红外图像的特点.......................................................62.3红外图像的直方图介绍...................................................................................72.4红外图像的性质分析（对比度、分辨率、噪声）.......................................82.5本章小结...........................................................................................................9第3章红外图像增强算法研究................................................................................103.1红外图像增强处理总体介绍.........................................................................103.2空间域增强.....................................................................................................113.2.1空域变换、对比度增强.......................................................................123.2.1.1分段线性变换.............................................................................123.2.1.2图像反转.....................................................................................153.2.1.3对数变换.....................................................................................163.2.1.4直方图均衡化.............................................................................173.2.1.5直方图规定化.............................................................................203.2.2空域滤波增强.......................................................................................223.2.2.1均值平滑滤波.............................................................................223.2.2.2中值平滑滤波.............................................................................243.2.2.3线性锐化滤波............................................................................253.2.2.4非线性锐化滤波........................................................................273.3频域增强.........................................................................................................293.3.1低通滤波...............................................................................................293.3.2高通滤波..............................................................................................303.4其他常用的及改进后的红外图像增强处理算法.........................................313.5本章小结.........................................................................................................32第4章结论................................................................................................................33参考文献......................................................................................................................34致谢......................................................................................................................35I摘要随着现代红外技术的快速发展，民用红外监控系统及军用红外探测系统得到了广泛的应用。由于红外摄像器件本身及探测环境影响，成像效果并不理想。在实际应用中，需要对获得的红外图像进行必要的增强处理，使其更适于人眼观察。同时利用对红外目标的检测、分割来确定不易辨认的目标物，为后续的识别与智能控制等奠定基础。本文首先介绍了红外成像机理，在分析了红外图像特点的基础上，比较了几种经典红外图像增强算法。根据红外成像具有图像模糊、噪声大等特点，提出了用直方图均衡增强灰度、用中值平滑滤波消除噪声的红外图像处理算法。实验结果表明，该算法能够增强图像目标、有效的抑制噪声，具有较好的视觉效果。数字图像处理是指将图像信号转换成数5字格式并利用计算机对其进行处理的过程。图像增强是数字图像处理的过程中经常采用的一种方法，它对提高图像质量起着重要的作用。本文先对图像增强的原理进行概述，然后对图像增强的方法分类并给出直方图增强、对比度增强、平滑和锐化等几种常用的增强方法的理论基础，通过Matlab实验得出的实际处理效果来对比各种算法的优缺点，讨论不同的增强算法的技术要点，并对其图像增强方法进行性能评价。关键词：红外图像；直方图增强；对比度增强；平滑；锐化IIAbstractWiththefastdevelopmentoftheinfraredtechnology,infraredmonitorsystemiswidelyusedincivilandmilitaryfields.Becauseofthelimitationsofinfraredmonitorandtheaffectionoftheenvironment,thequalityofinfraredimagingisnotsuchsatisfactory.Inpracticalapplication,imageenhancementtechnologyisbenefittoviewoftheobtainedimage.Meanwhile,undistinguishedtargetcanbeconfirmedbydetectionandsegmentation,providingfoundationsforthefollowingidentificationandintelligentcontrol.Firstly,infraredimagingmechanismisintroduced.Basedonanalyzingthecharacteristicsofinfraredimage,someclassicinfraredimageenhancingalgorithmsarecompared.Accordingtosuchcharacteristicsasblurredimageandhighnoise,anewalgorithmcombiningplatformhistogramequalizationwithenhancedhigh-passfilteringisproposed.Theresultsshowthatthealgorithmcanenhancetheobjectandrestrainnoiseofinfraredimage,achievingbettervisualeffect.Digitalimageprocessingistheproceduresofconvertingimagesignalintodigitalformat,thenusingthecomputertoprocessit.Imageenhancementisdigitalimageprocessingprocessoftenuseamethodtoimproveimagequality,itplay