臉部辨識已經廣泛的被應用在很多地方,多數的臉部辨識方法是聚焦在二維影像的探討,我們提出一個創新的方法結合二維臉部特徵資訊及三維的臉部深度資訊,接著我們使用賈伯小波強化二維影像的區域特徵向量在不同角度及方向的特徵資訊,然後我們結合二維臉部特徵資訊及三維臉部深度資訊,使用主成份分析法來取得二維臉部及三維臉部深度的特徵向量資訊,在辨識方面,我們提出植基於差異進化之多層類神經網路於臉部辨識應用。經實驗結果證明我們所提出的方法可以有效的辨識不同臉部表情及不同臉部姿勢的人臉。另外,我們也比較不同特徵數量的臉部資訊及測試在不同隱藏層數量,實驗結果證明我們提出的差異演化法擁有較穩定的辨識效果,也證明了結合二維臉部特徵及三維臉部表面特徵向量的方法,可以有效的提升臉部辨識率的效果。
Face recognition is a widely in many applications. Most of the approaches to face recognition have focused on the use of two dimensional images. We present an innovative method that combines two-dimensional texture and three-dimensional (3D) images surface feature vectors. Next, we use Gabor wavelets extracting local features at different scales and orientations by two-dimensional facial images. Next, we combine the texture with the three-dimensional (3D) images surface feature vectors based on principal component analysis (PCA) to obtain feature vectors from gray and facial surface images. We also propose a differential evolution (DE) algorithm for face recognition based on multilayer neural networks as an identification model. In ours experimental results demonstrate for the recognition different face poses and facial expressions method was efficiency. In addition, our work compared with different number of facial images feature vector and hidden nodes the experimental results demonstrated that the proposed differential evolution algorithm have a stable recognition rate. The texture and shape features can improve the efficiency of face recognition.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。