資源描述：
《保持圖像亮度的局部直方圖均衡算法.pdf》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在行業(yè)資料-天天文庫(kù)。
1、第5期電子學(xué)報(bào)Vol.34No.52006年5月ACTAELECTRONICASINICAMay2006保持圖像亮度的局部直方圖均衡算法1，2111江巨浪，張佑生，薛峰，胡敏（1.合肥工業(yè)大學(xué)計(jì)算機(jī)與信息學(xué)院，安徽合肥230009；2.安慶師范學(xué)院物理與電氣工程學(xué)院，安徽安慶246011）摘要：現(xiàn)有的局部直方圖均衡算法用于圖像的對(duì)比度增強(qiáng)時(shí)，輸出圖像的亮度與輸入圖像無關(guān).由于圖像增強(qiáng)的很多實(shí)際應(yīng)用需要保持原始圖像的亮度，為此提出一種新的局部直方圖均衡算法.以子塊部分重疊的局部直方圖均衡算法（POSHE）為基礎(chǔ)，改進(jìn)其對(duì)子塊圖像的均衡方式：根據(jù)亮度均值對(duì)子塊圖像進(jìn)行遞歸分解，得到一系列
2、不同灰度范圍的子圖像，然后對(duì)每一個(gè)子圖像在其相應(yīng)的灰度范圍內(nèi)進(jìn)行直方圖均衡，最后合并這些子圖像的均衡結(jié)果.實(shí)驗(yàn)結(jié)果表明，這種局部直方圖均衡新算法既能充分增強(qiáng)圖像局部細(xì)節(jié)，又能保持輸入圖像的亮度，且亮度的保持程度可由用戶進(jìn)行控制.本文算法使所增強(qiáng)的圖像具有更加自然的視覺效果，能更好地滿足消費(fèi)類電子工業(yè)對(duì)圖像對(duì)比度增強(qiáng)的應(yīng)用要求.關(guān)鍵詞：對(duì)比度增強(qiáng)；直方圖均衡；局部直方圖均衡；亮度保持中圖分類號(hào)：TP391.41文獻(xiàn)標(biāo)識(shí)碼：A文章編號(hào)：0372-2112（2006）05-0861-06LocalHistogramEgualizationwithBrightnessPreservation
3、1，2111JIANGJu-lang，ZHANGYou-sheng，XUEFeng，HUMin（1.SchoolofComputerScienceandInformation，hefeiUniversityofTechnology.hefei，Anhui230009，China；2.SchoolofPhysicsandElectricalEngineering，AngingTeacher'sCollege.Anging，Anhui246011，China）Abstract：Asthetraditionalalgorithmsforlocalhistogramegualizationi
4、susedtoenhancethecontrastofanim-age，theoutputbrightnessisalwaysregardlessoftheinputimage.However，thisisnotthedesirablepropertyinmanyap-plicationswherepreservingtheoriginalbrightnessisessential，soanewalgorithmoflocalhistogramegualizationispro-posed.Basedonthealgorithmofpartiallyoverlappedsub-blo
5、ckhistogramegualization（POSHE），wechangethewayofegualizingsub-block：basedonthemean，eachsub-blockisseparatedrecursivelyintoaseriesofsub-imageswithdifferentgrayranges，thenthehistogramsareegualizedovertheirrespectivegrayranges，andfinallytheoutputofthesub-blockiscomposedwiththeresultingegualizedsub-
6、images.Thesimulationresultsshowthat，withthenewalgorithm，thelocalde-tailsontheimagecanbeenhancedeffectively，andtheoriginalbrightnesscanalsobepreservedinthemeantime.Besides，scalabledegreeofbrightnesspreservationmaybeachievedbychoosingthenumberofrecursiveseparation.Theproposedalgorithmcanenhanceth
7、eimageinamorenaturalmannerandhence，isveryusefulformanycommercialpurposessuchasconsumerelectronicswherebrightnesspreservationisreguired.Keywords：contrastenhancement；histogramegualization；localhistogramegualization；brightnesspreserv
