文档介绍：Ｈybrｉｄ Genetic Aｌgorｉthm Ｂａsｅd Iｍagｅ Ｅnｈaｎｃement
Tｅchｎｏlogy
Abｓtraｃt:Ｉn ｉmａgｅ enhanｃeｍenｔ, Tubbｓ prｏｐoｓed ａ normａｌｉzeｄ ｉｎcoｍpletｅ Bｅｔa funcｔｉon ｔo rｅｐreｓeｎt seｖeral kinｄs of comｍｏnly ｕsed non-lｉｎeａr ｔransｆｏrm ｆｕnｃtiｏｎs to do ｔhｅ reｓeaｒcｈ on image ｅnhaｎｃｅｍenｔ. Ｂｕt hｏw to ｄeｆinｅ tｈe cｏｅｆｆｉｃｉeｎｔs of ｔｈe Ｂeｔa fuｎｃtioｎ ｉs sｔill ａ proｂｌem. We ｐropoｓｅｄ ａ Ｈyｂｒid Genetiｃ Ａlgoritｈｍ whicｈ ｃomｂiｎｅｓ tｈe Dｉｆfｅｒential Evoｌuｔｉｏｎ to ｔhe Geｎｅtic Alｇoｒiｔhｍ ｉn tｈｅ iｍａｇe enhaｎceｍｅnt prｏｃess and ｕtilｉze ｔhe quiｃｋlｙ ｓｅarching abiｌiｔｙ ｏf tｈe algｏritｈm to carry ｏｕt the aｄaptｉｖe mｕtａtioｎ and ｓｅaｒcheｓ. Fｉnalｌｙ we ｕse ｔhｅ Sｉmuｌａtｉoｎ eｘｐeｒｉｍent ｔo provｅ ｔhｅ ｅｆfectiveｎess oｆ thｅ meｔhｏd.
Keｙwoｒds：Imａｇe enhanｃｅment; Hybｒid Genetic Algoｒiｔhm; aｄaｐtivｅ eｎhaｎcemｅｎt
I． INTＲOＤUＣＴIOＮ
Ｉｎ tｈe imagｅ formａtion, tｒansfeｒ ｏr conｖeｒｓｉｏｎ prｏceｓs, due ｔo othｅr obｊeｃｔiｖe ｆａｃtorｓ sucｈ aｓ syｓtem nｏise, inadeｑuatｅ oｒ exｃesｓiｖｅ ｅxpｏｓure, ｒｅlａtｉｖe motiｏn and so ｔhe imｐａｃｔ will gｅt ｔｈe imaｇe ofｔｅｎ ａ difｆeｒenｃe ｂｅｔween the originａｌ imaｇｅ (refeｒreｄ ｔo as ｄｅgraｄed oｒ ｄegｒａｄed) Dｅｇrａｄed ｉmage is ｕsｕally blurrｅｄ ｏｒ after tｈe exｔractiｏn ｏｆ ｉnｆｏrmaｔion tｈrougｈ thｅ machiｎe to rｅduce or ｅven ｗｒｏｎg, ｉｔ mｕst tａｋe ｓｏme meaｓuｒｅｓ ｆor iｔs iｍprovｅｍenｔ.
Imagｅ enhanｃemenｔ teｃｈｎｏｌogｙ ｉs pｒopｏseｄ iｎ thiｓ sense, anｄ the pｕｒｐoｓe iｓ ｔｏ impｒoｖｅ ｔhe image quaｌiｔy． Ｆuｚｚy Imaｇｅ Ｅnｈaｎｃｅｍｅnｔ ｓｉtuａtion accordｉng ｔo tｈｅ ｉmage usｉng ａ ｖarｉety ｏf sｐecｉaｌ tｅchｎｉcal ｈｉghlights some of the iｎfoｒｍａｔion in tｈｅ ｉｍａge, redｕce oｒ eｌｉminａｔe the irrelｅvanｔ inｆoｒmａｔｉon， tｏ eｍpｈａsize the imaｇe oｆ tｈｅ ｗhole ｏr the pｕrposｅ oｆ local feaｔｕres. Imaｇe eｎｈａｎcement mｅｔｈｏd ｉｓ ｓtｉlｌ no uｎiｆied theorｙ, imａｇｅ eｎhaｎcemenｔ ｔecｈniqｕes cａn be diviｄed inｔo ｔｈree categｏrieｓ： point oｐeraｔiｏns, and ｓpaｔｉaｌ ｆｒequenｃｙ
enhancｅｍeｎt mｅthｏdｓ Enｈａncｅｍent Act. Tｈis ｐaper prｅsｅnts an aｕｔｏｍａtｉc adjuｓｔmenｔ aｃcoｒdinｇ to ｔhe ｉmage ｃｈａrａｃteｒistｉｃｓ of ａdaptive ｉｍａge enhａncemｅnt ｍethoｄ tｈaｔ ｃallｅｄ hybｒiｄ geneｔｉc aｌgｏrithm. Ｉt coｍbines the dｉffeｒｅｎｔiaｌ ｅvolution algoｒｉthｍ ｏｆ aｄapｔｉ