It all started rather innocuously. I walked into Dr GT Murthy’s office one fine day, andchanged my life. “Doc” was then the General Manager, Central R&D, of a very largeelectrical company headquartered in Bombay. In his new state-of-the-art electronics center,he had hand-picked some of India’s best engineers (over a hundred already) ever assembledunder one roof. Luckily, he too was originally a Physicist, and that certainly helped me gainsome empathy. Nowadays he is in retirement, but I will always remember him as athoroughly fair, honest and facts-oriented person, who led by example. There were severalthings I absorbed from him that are very much part of my basic engineering persona today.You can certainly look upon this book as an extension of what Doc started many years agoin India … because that’s what it really is! I certainly wouldn’t be here today if I hadn’t metDoc. And in fact, several of the brash, high-flying managers I’ve met in recent years,desperately need some sort of crash course in technology and human values from Doc!
標簽:
開關電源
上傳時間:
2021-11-23
上傳用戶:
車牌定位---VC++源代碼程序
1.24位真彩色->256色灰度圖。
2.預處理:中值濾波。
3.二值化:用一個初始閾值T對圖像A進行二值化得到二值化圖像B。
初始閾值T的確定方法是:選擇閾值T=Gmax-(Gmax-Gmin)/3,Gmax和Gmin分別是最高、最低灰度值。
該閾值對不同牌照有一定的適應性,能夠保證背景基本被置為0,以突出牌照區域。
4.削弱背景干擾。對圖像B做簡單的相鄰像素灰度值相減,得到新的圖像G,即Gi,j=|Pi,j-Pi,j-1|i=0,1,…,439 j=0,1,…,639Gi,0=Pi,0,左邊緣直接賦值,不會影響整體效果。
5.用自定義模板進行中值濾波
區域灰度基本被賦值為0。考慮到文字是由許多短豎線組成,而背景噪聲有一大部分是孤立噪聲,用模板(1,1,1,1,1)T對G進行中值濾波,能夠得到除掉了大部分干擾的圖像C。
6.牌照搜索:利用水平投影法檢測車牌水平位置,利用垂直投影法檢測車牌垂直位置。
7.區域裁剪,截取車牌圖像。
標簽:
1.24
256
圖像
閾值
上傳時間:
2013-11-26
上傳用戶:懶龍1988