驗證碼識別基礎(chǔ)方法及源碼

先說說寫這個的背景最近有朋友在搞一個東西，已經(jīng)做的挺不錯了，最后想再完美一點，于是乎就提議把這種驗證碼給K.O.了，于是乎就K.O.了這個驗證碼。達到單個圖片識別時間小于200ms,500個樣本人工統(tǒng)計正確率為95%。由于本人沒有相關(guān)經(jīng)驗，是摸著石頭過河。本著經(jīng)驗分享的精神，分享一下整個分析的思路。在各位大神面前獻丑了。再看看部分識別結(jié)果8FAE2捫QOQO-4MPQ-4MPOO1--8FAE-.8FAE.jpOT2-YG^H-￥G9H.DD3-MYAV-MYA004-2B7Q-2B7Q.JQ-jipggjpgVjpgpgx3nC況譏G4叱H￡4S83EJW5-X3NC-X3NC.O06-RWTU-FLWT007-GAWQ-GAWOTS-HE45-HE45.J009-83EJ-B.3EJ.jpgJpgUjpqQjpgP9ekcg45B3P3K7F^701O-EKCG-EKC&,jD11-45B3-45Bl.jp012-劌HZ9￥4Zj013-PSKT-PSKT.J014-FWA7-FWA7,P9gpgP9jpgJG2FP4CBxMS巧016-Q4PD-Q4PD.D17-G2FP-G2FP.J01S-4CBX-4CBX.Jfll^-MSTT-MSTT.9jpgP9P9jpg94RY6^vXTKPDyUC.OiO-WRV-'MRV.jp021-&BAV-&BAV.J022-NPNC-NPN024燈KP-XTKP.j024-DVU匚-嘰C,gP9c-jP9pgjpgSBUEF浮V9/3b025-RVWS-RVW5.OQfr-SBUE-SBUE.j027-VKWK-VKWK028-FQ9V-F-Q9V.J029-9J3U-gJ5U.jp」pgpg■jpgpgg7^76PV7B85DGU便(13fl-7KYT-7KYT.j051-B7GG-B7GG.jD52-6PVT-&PVT.jCG3-B35D-B35D.J0B4-GUKS-GUIC3.jpgpgpgpgpg是不是看著很眼熟？處理第一步，去背景噪音和二值化對于這一塊，考慮了幾種方法。方法一，統(tǒng)計圖片顏色分布，顏色占有率低的判定為背景噪音。由于背景噪音和前景色區(qū)分并不明顯，嘗試了很多種取景方法都不能很好去除背景噪音，最終放棄了這種方法。方法二，事后在網(wǎng)上稍微查了下，最近比較流行計算灰度后設(shè)定一個閾值進行二值化。其實所謂的灰度圖片原理是根據(jù)人眼對色彩敏感度取了權(quán)值，這個權(quán)值對計算機來說沒有什么意義。稍微想一下就可以發(fā)現(xiàn)，這兩個過程完全可以合并。于是乎我一步完成了去背景噪音和二值化。閾值設(shè)置為RGB三分量之和到500。結(jié)果非常令人滿意。

OMpQ8FAEYG%2^7qDDO.bmpD01.bmpOOZ.bmpC03-.bmpOD4.bmpx3nCG4^HE4S83EJD05.bmp005.bmp007.bmpOO&.bmpOO9.bmpE?45B39丫皿p3KjFJ07D10.bmpD11.bmp012.bmpD13-.bmpD14.bmpJB電/PEG2FP4CBxMSTjD15.bmpD16.bmp017.bmpD18.bmpD19.bmp94RY帚2xtkpdyuc020.bmp021.bmp022.bmpO23-.bmp024.bmpfiyTSSSBUEF汐V9j3b025.bmp026.bmpC27.bmp028.bmp029.bmp7^7B7G06PV7B85DGU建D30.bmpD31.bmpQ3Z.bmpD33-.bmpQ34.bmp549q^U535J4QD35.bmpD36.bmpQ37.bmp038.bmpQ39.bmpBeF^叱8E^9PAP040.bmp041.bmpO4Z.bmpMi.brripO44.bmp處理第二步，制作字符樣本樣本對于計算機來說是非常重要的，因為計算機很難有邏輯思維，就算有邏輯思維也要經(jīng)過長期訓(xùn)練才能讓你滿意。所以要用事先制作好的樣本進行比較。如果你仔細觀察過這些驗證碼會發(fā)現(xiàn)一個bug,幾乎大部分的驗證碼都是使用同樣的字體，于是乎就人工制作了一套字體的樣本。由于上一步已經(jīng)有去除背景噪音的結(jié)果，可以直接利用。制作樣本這一步有點簡單枯燥，還需要細心。可能因為你的一個不細心會導(dǎo)致某個符號的識別率偏低。在這500個樣本中，只發(fā)現(xiàn)了31個字符。幸虧是某部門的某人員還考慮到了易錯的字符，例如，1和I，0和O等。要不然這個某部門要背負更多的罵名。處理第三步，匹配單個匹配用了最簡單最原始的二值比較，不過匹配的是匹配率而不是匹配數(shù)。我定義了相關(guān)的計分原則。大原則是“該有的有了加分，該有的沒了減分，不該有的有了適度減分，可達區(qū)域外的不算分”。由于一些符號的部分區(qū)域匹配結(jié)果跟另一些符號的完整匹配結(jié)果相似，需要把單個匹配在一個擴大的區(qū)域擇優(yōu)。在一定的圍，找到一個最佳匹配，這個最佳匹配就是當前位置對應(yīng)的符號。完成了一次最佳匹配，可以把匹配位置向右推進一大步，若找不到合適的最佳匹配就向右推進一小步。處理第四步，優(yōu)化和調(diào)整任何一個算法都是需要優(yōu)化和調(diào)整的?，F(xiàn)在要找到最佳參數(shù)配置和最佳代碼組織這一步往往是需要花費最多時間和精力的。處理第五步，驗證結(jié)果這一步呢，純?nèi)肆︱炞C結(jié)果，統(tǒng)計出正確率。思考結(jié)果是出來了，代碼也不多，效果也很理想。搞這一行的，很多時候都想要通用的能否通用，很大程度上在于抽象層次。本方法只是單純的匹配，自然不能通用，但是方法和思想?yún)s是通用的。具體案例具體分析。至于扭曲文字、空心文字等，處理要復(fù)雜的多。網(wǎng)上也有一些使用第三方圖像庫的方法，也許那些方法會比較通用。等有空了有興趣了繼續(xù)搞一下這個主題。源碼至于這個源碼要不要發(fā)布，糾結(jié)了一段時間。網(wǎng)上已經(jīng)有類似的商業(yè)活動了，而且這個識別本身沒有太大難度，再加上某系統(tǒng)天生的bug,此驗證碼本身就相當于沒有設(shè)置，因此發(fā)布此代碼，僅作于學(xué)習(xí)交流。+ViewCode?1234567891011121314151617123456789101112131415161718usingSystem.Collections.Generic；usingSystem.Drawing；usingSystem.IO；usingSystem.IO.Compression；namespaceCrackl2306Captcha{publicclassCracker{List<CharInfo>words_=newList<Charlnfo>()；publicCracker。{0x00,0x00,0x04,0x00,0x97,0x2f,0xe1,0x58,varbytes=newbyte[]{0x1f,0x8b,0x08,0x00,0x00,0xc5,0x58,0xd9,0x92,0x13,0x31,0x0c,0x94,0x9e,0x93,0x0c,0xe0,0x91,0x9b,0x82,0x62,0x0b,0x58,0xee,0xff,0xff,0x10,0x00,0x61,0xd8,0xcc,0xc8,0xea,0x96,0x6c,0x8f,0x13,0x48,0xel,0xaa,0x4d,0x46,0x96,0x6d,0xb5,0x8e,0x96,0x67,0x73,0x7f,0x3b,0x09,0x0e,0x25,0x41,0x49,0xa3,0xae,0xd7,0x5b,0xa9,0xa8,0xd5,0xb4,0x76,0x02,0x6a,0x5c,0x52,0x94,0x54,0xed,0x18,0x5a,0x7f,0x18,0x00,0x00,TOC\o"1-5"\h\z0x84, 0x07, 0x1b, 0x80, 0x4a, 0x9a, 0x08, 0x35, 0xb8, 0x81,0x50,0xe7,0xad,0xbe,0xc4,0x8e,0xb1, 0x4f, 0x2d, 0x5f, 0xba, 0x80, 0xbb, 0xfd, 0x9a, 0xad,0x19,0x36,0xe5,0xad,0x87,0xf1,0x10, 0xc0, 0x8d, 0xc6, 0x50, 0x40, 0x52, 0xf8, 0xb3, 0x98,0x2c,0xd6,0xec,0x59,0xe7,0x0d,0x3e,0x0f,0x93,0x3e,0x1d,0x02,0x7a,0x18,0x8f,0xb6,0xc7,0x46,0x4e,0x01,0xa3,0x96,0xdc,0x3a,0x20,0x77,0xbf,0x2c,0x24,0xe4,0x80,0xa9,0x20,0x14,0xe5,0x2d,0xb5,0x68,0xc9,0x55,0x89,0x23,0x96,0x82,0xaa,0xba,0x58,0xa6,0x03,0x38,0x71,0x4b,0x29,0xd2,0x47,0x80,0xe3,0x84,0x91,0xf4,0x78,0x43,0x64,0x41,0x7b,0x73,0x99,0x80,0x42,0x48,0x00,0xde,0x00,0x12,0x88,0x80,0xdb,0x51,0x4a,0x49,0x84,0x43,0xf6,0x51,0x90,0x27,0x21,0xc9,0xf8,0xac,0x00,0x4d,0xcd,0x46,0x09,0x9d,0x15,0x78,0xe0,0x00,0x1e,0x44,0x2a,0x51,0x8c,0xbc,0xd3,0xa3,0x68,0x8a,0xd5,0x3a,0x20,0x79,0xba,0x4d,0x71,0x4c,0x0b,0x91,0x98,0x90,0x7b,0x2a,0x42,0xc5,0x78,0x7a,0xfc,0xd5,0x1b,0x4b,0x09,0xa7,0x27,0x99,0x38,0x05,0x01,0xc2,0x80,0x39,0x9c,0x67,0xbb,0x4e,0x7f,0x6c,0x33,0xdd,0xed,0x87,0x55,0xda,0x5d,0xb5,0x56,0x33,0xc6,0xf9,0xea,0x60,0x64,0xcf,0xa7,0x41,0xe0,0x5c,0x1c,0xc4,0xb2,0x25,0xa3,0x89,0x88,0x8d,0x16,0x00,0xb5,0xed,0xa5,0x22,0x9d,0x52,0x41,0x53,0x8d,0x92,0x7f,0x31,0x51,0x3f,0xa8,0x00,0x85,0x8a,0x71,0x10,0x92,0x78,0xc4,0x59,0x08,0x39,0x69,0xa9,0x38,0x41,0x48,0xf7,0x40,0x5a,0x03,0xd5,0x3a,0xf5,0xe5,0x9d,0x33,0x66,0xc3,0xd7,0x1f,0xef,0x94,0xa0,0x53,0xea,0xf4,0x15,0xb2,0x1c,0x40,0x2d,0xcf,0xaf,0xce,0xe9,0xd4,0x7a,0x89,0x09,0xe6,0xdd,0xdb,0x0e,0xb8,0x58,0xa7,0x60,0x37,0xfd,0xf2,0xfa,0x2c,0x4e,20212223242526272829303132333435363738394041424344454647484950515253545556575859606162630x51,0x87,0x0d,0xfc,0x16,0x72,6465666768697071727374757677787980818283848586878889909192939495969798991001011021031041051061070x2a,0x5f,0xc0,0x80,0xf0,0x54,0xa7,0xde,0xfc,0x15,0x8b,0x9a,0x36,0x3a,0x2c,0x62,0xfc,0xd4,0x8c,0x31,0xb7,0xea,0xd7,0x26,0xc4,0xaf,0x75,0xea,0xdb,0x8b,0xff,0x9b,0x9b,0x50,0x7e,0xfe,0x15,0xab,0x17,0x2f,0x96,0x96,0xbd,0xaa,0x87,0xdd,0x77,0xa3,0x77,0xd3,0x85,0xf0,0xe0,0x58,0xd5,0xf6,0x8c,0xcd,0xc4,0x63,0x52,0x12,0x48,0x46,0x0f,0x93,0x5a,0xe3,0xea,0x24,0x67,0x73,0x63,0xa0,0xdf,0xdf,0x3d,0x67,0xf6,0xa9,0xfc,0xed,0x08,0xe3,0x82,0x57,0x08,0x35,0x47,0x68,0x9c,0x01,0x40,0x87,0x8b,0xbd,0x0c,0xb3,0xf4,0xe1,0x72,0xd7,0x54,0x62,0xfd,0x40,0xed,0x99,0xa6,0x7e,0x2b,0xe4,0xb4,0xc4,0x62,0x0d,0x79,0xae,0x1b,0xd7,0xf4,0x09,0xb7,0xe1,0x7c,0x44,0x09,0x9a,0xda, 0xff, 0x52, 0x6a, 0x3c, 0xe1, 0xc8, 0xd7, 0xbd, 0xbb,0xbe,0x37,0xfc,0xd6,0xd5,0x4e,0x3c, 0x40, 0x2a, 0x4b, 0x39, 0x1a, 0xbd, 0x2a, 0xcd, 0xc1,0x18,0x59,0x40,0x62,0x78,0xec,0x63, 0x19, 0x72, 0xf0, 0xcf, 0xf8, 0x38, 0xfa, 0x42, 0x3a,0xc8,0x02,0xec,0x5b,0xeb,0x8d,0xae,0xf1,0x45,0xdd,0x32,0x98,0x35,0x3c,0x9f,0xa6,0x3d,0xce,0x13,0xce,0x94,0x38,0x87,0x00,0x8d,0x85,0xc4,0x70,0x17,0x26,0x0e,0xa6,0x1e,0x16,0xcb,0xbf,0x52,0xdf,0x29,0x63,0xc4,0xf6,0x8c,0x35,0xba,0xf2,0xf9,0x1f,0xbf,0x73,0x1f,0x91,0x1b,0x9e,0x24,0x5e,0x63,0x22,0x82,0x23,0x05,0x19,0xb9,0x71,0x73,0xdc,0xcf,0x05,0x88,0x94,0x71,0xdb,0xdd,0x48,0x10,0xd5,0x55,0xb3,0x52,0xc3,0x1b,0x01,0x94,0x13,0x74,0x94,0x3a,0x80,0x2f,0x39,0xe2,0x75,0x0e,0xf2,0xc6,0x18,0xdc,0x46,0xfc,0xf3,0xea,0x14,0x80, 0xc1, 0xce, 0x24, 0xee, 0x72, 0xed, 0x94, 0xaf, 0xfb,0xa9,0xaa,0x4a,0xe0,0xd4,0x22,0xc6, 0xf0, 0x57, 0x1d, 0x8e, 0xd2, 0x90, 0xc6, 0x0c, 0xd3,0x9a,0x53,0xfb,0xd6,0xb7,0xdd,0x14, 0xd4, 0xbd, 0x41, 0xa7, 0x80, 0x7b, 0x23, 0xfe, 0x34,0x56,0x0d,0x96,0x46,0x02,0xfe,0xfd,0xb2,0x00,0x5f,0x01,0x9c,0xa0,0x32,0x39,0xd7,0x90,0xc2,0x6c,0xc7,0x4e,0x68,0x88,0x7d,0x9f,0x9b,0xcf,0xa7,0xbe,0xa0,0xfc,0x18,0x7d,0x07,0x5b,0xa9,0xbe,0x56,

1080xlf,0x67,0x1a,0x4a,0x91,0x9c,0x04,0x38,0x53,0x6b,1090x70,0x68,0x8f,0xea,0xf4,0x34,1100x87,0x7f,0x6e,0x82,0xc3,0xc1,0xab,0x40,0xc4,0x50,1110x13,0x0e,0x33,0x5d,0x67,0x7d,1120x01,0x1f,0xdb,0xc0,0x7f,0xed,0x87,0x7f,0xbc,0x0f,1130x75,0xe0,0xa5,0xba,0xc0,0x84,1140x3d,0x24,0x04,0xe0,0xf1,0x16,0x41,0x3b,0x74,0xd2,1150x52,0xc5,0xf8,0x7c,0x12,0xfb,1160xe4,0x37,0x5b,0xfb,0x57,0x11,0xa1,0x18,0x00,0x00,117};118using(varstream=newMemoryStream(bytes))119using(vargzip=newGZipStream(stream.120CompressionMode.Decompress))121using(varreader=newBinaryReader(gzip))122{123while(true)124{125charch=126reader.ReadChar();127if(ch=='\0')128break;129intwidth=130reader.ReadByte();131intheight二132reader.ReadByte();133134bool[,]map=new135bool[width,height]；136for(inti二0；i<width；137i++)138for(intj二0；j<139height；j++)140map[i,j]141=reader.ReadBoolean()；142words_.Add(new143CharInfo(ch,map))；144}145}146}147148publicstringRead(Bitmapbmp)149{150varresult二string.Empty；151varwidth=bmp.Width；

152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195varheight二bmp.Height；vartable=ToTable(bmp)；varnext二SearchNext(table,-1);while(next<width-7){varmatched=Match(table,next)；if(matched.Rate>0.6){result+=matched.Char;next二matched.X+10;}else{next+=1;}}returnresult；}privatebool[,]ToTable(Bitmapbmp){vartable=newbool[bmp.Width,bmp.Height]；for(inti二0；i<bmp.Width；i++)for(intj=0；j<bmp.Height；j++){varcolor=bmp.GetPixel(i,j)；table[i,j]=(color.R+color.G+color.B<500)；}returntable；}privateintSearchNext(bool[,]table,intstart){varwidth=table.GetLength(0)；varheight二table.GetLength(1)；for(start++；start<width；start++)for(intj=0；j<height；j++)if(table[start,j])returnstart；

196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231returnstart；target,intx0,sourceHeight)bool[,]target.privatedoubleFixedMatch(bool[,]source,bool[,]inty0)target,intx0,sourceHeight)bool[,]target.{doubletotal=0;doublecount二0；inttargetWidth=target.GetLength(0)；inttargetHeight二target.GetLength(l)；intsourceWidth=source.GetLength(0)；intsourceHeight二source.GetLength(l)；intx,y；for(inti二0；i<targetWidth；i++){x=i+x0；if(x<0||x>=sourceWidth)continue；for(intj=0；j<targetHeight；j++){y=j+y0；if(y<0IIy>=continue；if(target[i,j]){total++；if(source[x,y])count++；elsecount——；}elseif(source[x,y])count一二0.55；}}returncount/total；}privateMatchedCharScopeMatch(bool[,]source,intstart