【摘要】：近年來,對于人臉圖像方面的研究日益增多,人臉檢測、人臉特征點(diǎn)定位、人臉跟蹤及識別發(fā)展迅速,國內(nèi)外的許多學(xué)者已經(jīng)研究出了相對有效的方法和技術(shù),在應(yīng)用于人的正臉方向時,效果顯著且準(zhǔn)確率較高。對人臉進(jìn)行姿態(tài)估計,能夠為人臉圖像的進(jìn)一步研究提供技術(shù)支持。由于本課題是在具體的研究項目上所提出的,為了能夠?qū)⒁曨l中感興趣的人臉用拍攝的普通人臉進(jìn)行替換,使獲得的新視頻替換效果顯著且自然,本文主要是針對人臉左右偏轉(zhuǎn)在㧟30°到30°,上下俯仰在㧟30°到30°的情況,通過訓(xùn)練分類器對圖像或者視頻中人臉不同朝向的離散角度進(jìn)行估計。本文采用將主動形狀模型(Active Shape Models,ASM)特征點(diǎn)檢測方法與隨機(jī)森林分類算法相結(jié)合的方式,首先利用自行設(shè)計的樣本采集裝置采集人臉樣本,以用于隨機(jī)森林分類器的訓(xùn)練。用ASM算法檢測得到68個特征點(diǎn)并對其進(jìn)行歸一化處理,通過實驗對比,最終利用選取的7個關(guān)鍵點(diǎn)與其余特征點(diǎn)之間的距離作為特征來訓(xùn)練分類器。由于所得距離特征個數(shù)較多,即7×67(忽略點(diǎn)到自身的距離)=469,包含了較多的冗余信息,因此通過主成分分析算法(Principal Component Analysis,PCA)進(jìn)行擇優(yōu)選取,可以降低約90%的數(shù)據(jù)量。利用隨機(jī)森林?jǐn)?shù)據(jù)處理的高效性,將獲得的距離特征作為訓(xùn)練隨機(jī)森林的輸入數(shù)據(jù),訓(xùn)練得到不同朝向的分類器,構(gòu)成多層隨機(jī)森林。實驗結(jié)果表明,本文的算法在人臉左右偏轉(zhuǎn)范圍為㧟30°到30°,俯仰范圍為㧟30°到30°時,能夠較為準(zhǔn)確的得到視頻中人臉不同姿態(tài)的角度值。實驗表明,提取灰度特征、Gabor特征訓(xùn)練得到的隨機(jī)森林分類器,其準(zhǔn)確率要低于本文選取距離特征訓(xùn)練所得的分類器;同時與其他姿態(tài)估計算法相比,本文基于多層隨機(jī)森林分類的姿態(tài)估計算法所得到的結(jié)果較為準(zhǔn)確,并且效率較高。
[Abstract]:In recent years, the research on face image is increasing day by day, face detection, face feature point location, face tracking and recognition are developing rapidly. Many scholars at home and abroad have developed relatively effective methods and techniques. When applied to face orientation, the effect is remarkable and the accuracy is high. Face pose estimation can provide technical support for the further study of face image. Because this subject is put forward on the specific research project, in order to be able to replace the interested face in the video with the ordinary face taken, so that the new video replacement effect is remarkable and natural. This paper mainly aims at the situation that the face deflects from-30 擄to 30 擄and pitches up and down from-30 擄to 30 擄, and estimates the discrete angles of different face orientations in image or video by training classifier. In this paper, the active shape model (Active Shape Models,ASM) feature point detection method is combined with the random forest classification algorithm. Firstly, a self-designed sample acquisition device is used to collect face samples for the training of the random forest classifier. 68 feature points are detected by ASM algorithm and normalized. Finally, the distance between the selected seven key points and the other feature points is used as the feature to train the classifier. Due to the large number of distance features, that is, 7 脳 67 (distance from neglect point to itself) = 469, which contains a lot of redundant information, the amount of data can be reduced by 90% by selecting the optimal distance by principal component analysis (Principal Component Analysis,PCA) algorithm. Based on the high efficiency of random forest data processing, the obtained distance feature is used as input data of training random forest, and the classifier with different orientations is trained to form multi-layer random forest. The experimental results show that the proposed algorithm can accurately get the angle values of different face pose when the range of face deflection is -30 擄to 30 擄and pitch range is -30 擄to 30 擄. The experimental results show that the accuracy of the random forest classifier which is obtained by extracting gray feature and Gabor feature training is lower than that obtained by distance feature training in this paper. At the same time, compared with other attitude estimation algorithms, the results obtained by this algorithm based on multi-layer stochastic forest classification are more accurate and efficient.
【學(xué)位授予單位】：河北工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王建璽;李瑋瑤;;基于改進(jìn)Adaboost算法的人臉檢測[J];微處理機(jī);2015年05期

2 張忠義;;基于改進(jìn)LK光流的目標(biāo)跟蹤算法研究[J];信息技術(shù);2015年10期

3 劉袁緣;陳靚影;俞侃;覃杰;陳超原;;基于樹結(jié)構(gòu)分層隨機(jī)森林在非約束環(huán)境下的頭部姿態(tài)估計[J];電子與信息學(xué)報;2015年03期

4 牛曉霞;;基于三維模型的人臉姿態(tài)估計[J];微處理機(jī);2014年06期

5 李榮;劉坤;高文鵬;;基于視覺的目標(biāo)姿態(tài)估計算法[J];黑龍江科技大學(xué)學(xué)報;2014年01期

6 楊占棟;解梅;;基于半動態(tài)外觀模型的人臉識別[J];計算機(jī)工程;2011年24期

7 陳振學(xué);常發(fā)亮;劉春生;徐建光;;基于Adaboost算法和人臉特征三角形的姿態(tài)參數(shù)估計[J];武漢大學(xué)學(xué)報(信息科學(xué)版);2011年10期

8 方匡南;吳見彬;朱建平;謝邦昌;;隨機(jī)森林方法研究綜述[J];統(tǒng)計與信息論壇;2011年03期

9 李盛文;鮑蘇蘇;;基于PCA+AdaBoost算法的人臉識別技術(shù)[J];計算機(jī)工程與應(yīng)用;2010年04期

10 陳曉鋼;陸玲;周書民;劉向陽;;一種新的人臉姿態(tài)估計算法[J];數(shù)據(jù)采集與處理;2009年04期

相關(guān)碩士學(xué)位論文 前1條

1 王小明;彩色圖像序列的人臉姿態(tài)估計和跟蹤研究[D];華東師范大學(xué);2007年

，

本文編號：2419955