基於概略共同向量的人臉部分比對

A Study on the RCV-based Partial Face Recognition

陳冠志(Guan-Chih Chen)

指導教授 ： 陳靖國 ； 戴紹國

朝陽科技大學/資訊學院/資訊管理系/碩士(2017年)

摘要

人臉識別在移動裝置和硬體的蓬勃發展下，逐漸地被應用在真實生活中，例如:透過人臉辨識的門禁系統、攝影監視系統、出勤系統以及智慧型手機等等，但是到了今天人臉辨識仍然存在著一些問題尚未完全被解決。例如，在人臉部分被遮蔽情況下的比對就是一個很重要的議題。
在本文中將會使用概略的共同向量(Rough Common Vector)並且提出了人臉部分比對的機制來解決此問題。概略共同向量的特點是減少訓練資料的維度並且提取出重要的資料再作分析，而部分比對的機制可以自動地分析整張人臉影像需要整張影像進行比對還是切割後再比對比較有利，而且以一套適應性的權重投票機制來決定最相似的人臉。本篇將會使用AR人臉資料庫進行人臉部分比對，在人臉有遮蔽物的情況下，進行我們方法的效率評估，實驗結果顯示我們部分比對方法的識別率可以達到87.77%。

關鍵字

概略共同向量 ； 人臉部分比對 ； 人臉識別 ； 主成分分析 ； 線性判別分析

並列摘要

As technology develops with the mobile devices and hardware, face recognition is increasingly used in real-life applications, such as access control systems for face recognition, photographic surveillance systems, attendance systems, smartphone and so on. However, there are some problems that have been solved completely yet. For example, we want to face recognition when the human face is partially masked (partial match) is an important issue. In this paper, first we will discuss the eigenface method of face recognition, and discuss how Rough Common Vector worked. It will be used to solve this problem.
The feature of rough common vector is to reduce the dimension of training data and extract important data, then analysis it. Then chose the better result by using the partial match mechanism, automatically chose the entire human face or part of it, and also using a set of adaptive voting weight schema to decide the most similar human face.
This article will use the AR face database for experiments, when the face with shelter cases, evaluate the accuracy through our method. The accuracy of partial match can be 87.77% in this paper.

並列關鍵字

rough common vector ； face partial match ； face recognition ； PCA ； LDA

參考文獻

[41] Zhao, W., Chellappa, R. and Phillips, P. J. (1999), “Subspace Linear Discriminant Analysis for Face Recognition,” Computer Vision Laboratory, Center for Automation Research, University of Maryland.

[42] Zhao, W., Chellappa, R., Phillips, P. J. and Rosenfeld, A. (2003), “Face Recognition: A Literature Survey,” ACM Computing Surveys, Vol. 35, No. 4, pp.399-458.

[1] Agrawal, R., Gehrke, J., Gunopulos, D. and Raghavan, P. (1998), “Automatic Subspace Clustering of High Dimensional Data for Data Mining Applications,” Proc. ACM SIGMOD Int. Conf. Management of Data, Vol. 27, No. 2, pp. 94-105.

[2] Ahonen, T., Hadid, A. and Pietikainen, M. (2006), “Face Description with Local Binary Patterns: Application to Face Recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 28, No. 12, pp. 2037-2041.

[3] Batur, A. U. and Hayes, M. H. I. I. I. (2001), “Linear Subspaces for Illumination Robust Face Recognition,” Proc. of the IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, Vol. 2, pp. II-296-II-301.