人臉辨識近年來已被廣泛的應用,如門禁管理、身份辨識、金融提款業務、犯罪偵測等,在各領域上皆有不同之應用及成效。在取得人臉影像上,發現臉部有可能會遮蔽或是破損的情形,在此情況下即會提高臉部辨識的誤差率。故本研究除了基礎的臉部特徵擷取之外,還希望還原這些被遮蔽或是破損的區域,將人臉影像還原至可能的樣子。本研究先利用Feature-based(特徵基底)方式,擷取出人臉之特徵點並建立人臉資料庫,將人臉影像以及特徵點資訊等資料存入,以方便後續臉部資訊的修補。而在修補部份,首先是分析欲比對之人臉,將未遮罩區域之特徵點找出,這部份的方法與先前完整人臉的特徵擷取方法相同,接著再與資料庫中人臉比對,找尋最適當人臉,修補之。 In recent years, facial recognition has been applied to different fields extensively, such as entrance control, identification recognition, financial withdraw practice, criminal detection, and many more. Facial recognition techniques can be applicable in varies fields and has significant results. In obtaining facial images, if the face image were covered or damaged, the degree of error and inaccuracy will increase significantly. Therefore, the purpose of this study is not only for extracting facial features, but also to recover the damaged areas. Firstly, this study uses Feature-based methodology to detect facial feature points and build up the facial database. By inserting facial images and features points to the database, it will increase the accuracy and convenience in face reconstruction. As for the recovering procedure, the system will analyze the feature points on the uncovered face at first. These steps are the same as the part of extracting facial features. Next steps will be to compare with the faces within the facial database to find the applicable face, and reconstruct it.