本文選題：人臉檢測 + AdaBoost算法　； 參考：《杭州電子科技大學(xué)》2017年碩士論文

【摘要】：人臉檢測是人臉識別、表情分析、人臉跟蹤等領(lǐng)域的前提和基礎(chǔ),在視頻監(jiān)控領(lǐng)域有著廣泛的應(yīng)用價(jià)值。人臉檢測具有的不易被檢測目標(biāo)發(fā)現(xiàn)的顯著優(yōu)勢,可以目標(biāo)不配合的情況下進(jìn)行監(jiān)控,為智能化的視頻監(jiān)控提供了有力保障。本文研究的內(nèi)容是基于AdaBoost算法的視頻人臉檢測,由于在視頻人臉檢測時(shí)需要考慮眾多非理想因素(如復(fù)雜環(huán)境、多人臉、旋轉(zhuǎn)人臉等問題),而現(xiàn)有的檢測算法往往僅針對其中某種情況有較好的檢測效果。因此,對于視頻人臉檢測時(shí)存在的眾多干擾因素,建立一個(gè)檢測率高、誤檢率低、魯棒性較強(qiáng)的人臉檢測算法仍然是一個(gè)具有挑戰(zhàn)性的問題。由于AdaBoost人臉檢測算法具有法具有檢測效果好,并且基本可以達(dá)到實(shí)時(shí)人臉檢測的目的,因此本文采用該算法進(jìn)行視頻檢測,以下是本文的主要研究內(nèi)容:首先,以AdaBoost算法為基礎(chǔ),并針對于人臉檢測分類器訓(xùn)練過程中訓(xùn)練特征數(shù)目過多導(dǎo)致訓(xùn)練過程過于耗時(shí)的問題,提出了一種基于“大T”型區(qū)域的AdaBoost人臉檢測算法。通過隨機(jī)抽取500張人臉樣本并提取出面部的主要特征區(qū)域,投影到一個(gè)大小為20?20的模板中,然后對投影區(qū)域的重疊部分取其并集求得“大T”型Haar特征篩選模板,最后采用該模板對所有的Haar特征進(jìn)行篩選,使得用于AdaBoost算法的訓(xùn)練特征全部集中于人臉面部的關(guān)鍵區(qū)域,而且“大T”型特征篩選模板只限于AdaBoost人臉檢測算法訓(xùn)練特征的優(yōu)化與不同的訓(xùn)練樣本庫無關(guān)。實(shí)驗(yàn)結(jié)果表明,采用“大T”型特征篩選模板對Haar特征進(jìn)行篩選不僅降低訓(xùn)練特征的數(shù)目起到了優(yōu)化訓(xùn)練時(shí)間的目的,而且在LFW、PKU-SVD-B數(shù)據(jù)庫中的檢測結(jié)果也表明本文的改進(jìn)方法在不降低AdaBoost算法檢測率的同時(shí),對PKU數(shù)據(jù)庫中的多人臉檢測時(shí)算法的漏檢率有所改善,因此對于AdaBoost算法訓(xùn)練性能的提高有一定的作用。其次,針對于監(jiān)控設(shè)備采集到的視頻圖像中不可避免的存在多人臉、復(fù)雜背景、光照干擾、多姿態(tài)人臉等干擾因素,使得單獨(dú)采用AdaBoost算法進(jìn)行視頻人臉檢測時(shí),算法的誤檢率和漏檢率都比較高的問題,提出了采用AdaBoost算法與YCgCr混合高斯膚色模型相結(jié)合的方式進(jìn)行視頻人臉檢測。通過對膚色樣本在YCgCr色彩空間進(jìn)行建模,發(fā)現(xiàn)亮度分量同樣具有近似高斯分布的特性,因此采用線性加權(quán)的方式構(gòu)建Y分量與CgCr分量的混合高斯膚色模型,并采用該混合高斯模型對視頻圖像進(jìn)行膚色分割。實(shí)驗(yàn)結(jié)果表明,采用AdaBoost算法與新的膚色模型進(jìn)行視頻人臉檢測時(shí),可以較好的避免視頻圖像中復(fù)雜背景對人臉檢測的影響,進(jìn)而起到了降低算法的誤檢率和漏檢率的目的。
[Abstract]:Face detection is the premise and foundation of face recognition, facial expression analysis, face tracking and so on. It has wide application value in the field of video surveillance.Face detection has the obvious advantage that it is not easy to be detected by the target detection, and can be monitored without matching the target, which provides a strong guarantee for intelligent video surveillance.The content of this paper is video face detection based on AdaBoost algorithm, because many non-ideal factors (such as complex environment, multi-face) need to be considered in video face detection.However, the existing detection algorithms usually have better detection effect only for some cases.Therefore, it is still a challenging problem to establish a face detection algorithm with high detection rate, low false detection rate and strong robustness.Because the AdaBoost face detection algorithm has good detection effect, and can basically achieve the purpose of real-time face detection, this paper uses the algorithm to carry out video detection. The following are the main contents of this paper: first,Based on the AdaBoost algorithm, and aiming at the problem that the excessive number of training features in the training process of face detection classifier leads to the time-consuming training process, a AdaBoost face detection algorithm based on "large T" region is proposed.By randomly extracting 500 face samples and extracting the main feature region of the face, the main feature region of the face is projected into a template of 20 ~ 20, and then the "large T" type Haar feature screening template is obtained by the union of the overlapping parts of the projection region.Finally, the template is used to filter all the Haar features, so that the training features used in the AdaBoost algorithm concentrate on the key areas of the face.Moreover, the "large T" feature selection template is limited to the optimization of training features of AdaBoost face detection algorithm and is independent of different training sample bases.The experimental results show that "large T" feature screening template can not only reduce the number of training features, but also optimize the training time.The detection results in LFWU PKU-SVD-B database also show that the improved method does not reduce the detection rate of AdaBoost algorithm, but also improves the detection rate of multi-face detection algorithm in PKU database.Therefore, it can improve the training performance of AdaBoost algorithm.Secondly, in view of the inevitable interference factors such as multi-face, complex background, illumination interference, multi-pose face and so on in the video image collected by the monitoring equipment, the AdaBoost algorithm is used to detect the video face alone.The false detection rate and missed detection rate of the algorithm are both high. This paper proposes a method of video face detection based on the combination of AdaBoost algorithm and YCgCr mixed Gao Si skin color model.By modeling the skin color sample in the YCgCr color space, it is found that the luminance component also has the property of similar Gao Si distribution, so we use linear weighting method to construct the mixed skin color model of Y component and CgCr component.And the mixed Gao Si model is used to segment the color of the video image.The experimental results show that the AdaBoost algorithm and the new skin color model can avoid the influence of the complex background on the face detection and thus reduce the false detection rate and the missed detection rate of the algorithm.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 鐘銳;吳懷宇;吳若鴻;;基于強(qiáng)跟蹤Kalman濾波的魯棒人臉跟蹤算法[J];計(jì)算機(jī)工程與設(shè)計(jì);2016年02期

2 劉王勝;馮瑞;;一種基于AdaBoost的人臉檢測算法[J];計(jì)算機(jī)工程與應(yīng)用;2016年11期

3 王慶偉;應(yīng)自爐;;一種基于Haar-Like T特征的人臉檢測算法[J];模式識別與人工智能;2015年01期

4 許劍;張洪偉;;Adaboost算法分類器設(shè)計(jì)及其應(yīng)用[J];四川理工學(xué)院學(xué)報(bào)(自然科學(xué)版);2014年01期

5 趙紅雨;吳樂華;史燕軍;王志中;;基于HSV顏色空間的運(yùn)動目標(biāo)檢測方法[J];現(xiàn)代電子技術(shù);2013年12期

6 曹瑩;苗啟廣;劉家辰;高琳;;AdaBoost算法研究進(jìn)展與展望[J];自動化學(xué)報(bào);2013年06期

7 張君昌;張譯;;基于改進(jìn)AdaBoost算法的人臉檢測[J];計(jì)算機(jī)仿真;2011年07期

8 萬麗;陳普春;尹志勇;陳卓;夏巨武;;基于YCgCr色彩空間的人臉檢測技術(shù)研究[J];現(xiàn)代電子技術(shù);2011年04期

9 趙敏岑;王成儒;;基于YCgCb顏色空間的高斯膚色模型的人臉檢測[J];計(jì)算機(jī)安全;2010年12期

10 張爭珍;石躍祥;;YCgCr與YCgCb顏色空間的膚色檢測[J];計(jì)算機(jī)工程與應(yīng)用;2010年34期

相關(guān)碩士學(xué)位論文 前10條

1 陳雪婷;面向視頻人臉檢測的深度學(xué)習(xí)算法研究[D];杭州電子科技大學(xué);2016年

2 徐信;基于Adaboost人臉檢測算法的研究及實(shí)現(xiàn)[D];太原理工大學(xué);2015年

3 張靜;基于AdaBoost的視頻人臉檢測[D];杭州電子科技大學(xué);2015年

4 任璐;基于膚色分割和改進(jìn)Adaboost算法的人臉檢測[D];東北石油大學(xué);2014年

5 葛開標(biāo);基于改進(jìn)的LBP特征的AdaBoost算法與膚色檢測相結(jié)合的人臉檢測[D];重慶大學(xué);2012年

6 尹雪聰;基于可變形部件模型的人臉檢測方法研究[D];西安電子科技大學(xué);2012年

7 王健;基于Gentle Adaboost算法的人臉檢測研究[D];電子科技大學(xué);2011年

8 隋靜;基于視頻圖像的人臉檢測方法研究[D];西安電子科技大學(xué);2011年

9 陳抒;高速監(jiān)控視頻中的人臉檢測研究與實(shí)現(xiàn)[D];華東師范大學(xué);2009年

10 左登宇;基于Adaboost算法的人臉檢測研究[D];中國科學(xué)技術(shù)大學(xué);2009年

，

本文編號：1749451