浏览全部资源
扫码关注微信
Received：07 June 2019，
Revised：2019-12-25，
Accepted：01 January 2020，
Published：16 August 2020
移动端阅览
目的
2
针对基于Haar-like特征的Adaboost人脸检测算法，在应用于视频流时训练的时间较长，以及检测效率较低的问题，提出了一种基于区间阈值的Adaboost人脸检测算法。
方法
2
通过运行传统的Adaboost算法对人脸图像Haar-like特征值进行提取分析后，对人脸样本与非人脸样本特征值进行比较，发现在某一特定的特征值区间内，人脸和非人脸区域能够得到准确区分，根据此特性，进行分类器的选择，在简化弱分类器计算步骤的同时，降低训练时间，提高对人脸的识别能力。除此之外，弱分类器的增强通过Adaboost算法的放大使得强分类器分类精度提高，与级联结构的配合使用也提升了最终模型检测人脸的准确率。
结果
2
利用MIT（Massachusetts Institute of Technology）标准人脸库对改进Adaboost算法的性能进行验证，通过实验验证结果可知，改进后的Adaboost人脸检测算法训练速度提升为原来的1.44倍，检测率上升到94.93%，虚警率下降到6.03%。并且将改进算法在ORL（Olivetti Research Laboratory）、FERET（face recognition technology）以及CMU Multi-PIE（the CMU Multi-PIE face database）这3种标准人脸库中，分别与SVM（support vector machine）、DL（deep learning）、CNN（convolutional neural networks）以及肤色模型等4种算法进行了人脸检测对比实验，实验结果显示，改进后的Adaboost算法在进行人脸检测时，检测率提升了2.66%，训练所需时间减少至624.45 s，检测效果明显提升。
结论
2
提出的基于区间阈值的Adaboost人脸检测算法，在分类器的训练和人脸检测方面都比传统的Adaboost算法性能更高，能够更好地满足人员较密集处（如球场等地）对多人脸同时检测的实际需求。
Objective
2
With the continuous improvement of artificial intelligence technology
Adaboost algorithm based on Haar-like features has gained an important position in the endless stream of machine learning algorithms. Adaboost is widely used in the fields of medicine
transportation
and security. However
the Adaboost face detection algorithm based on Haar-like features
which is applied to video streams
has a long training time and low detection efficiency. This paper then proposes a novel Adaboost face detection algorithm based on the interval threshold.
Method
2
The integral value is used to quickly calculate the feature value of a face image
whereas the extracted feature values are used to analyze the Haar-like features of the final face recognition result. Comparing the feature values of face smaples and non-human face samples
and realized that the human and non-human faces are then determined by calculating the frequencies of different feature value intervals
and the final image is drawn. After a statistical analysis of the Haar-like eigenvalues of images
these two types of faces have been well distinguished within a certain eigenvalue interval
and a weak classifier based on double threshold is proposed accordingly. The interval threshold is used to select the weak classifier to simplify the calculation steps
shorten the training time
improve the face recognition and classification ability
and reduce the false alarm rate. Enhancing the weak classifier also improves the classification accuracy of the strong classifier by amplifying the Adaboost algorithm
whereas using a cascade structure increases the final face detection accuracy. This approach greatly accelerates the threshold search
and using the interval threshold instead of a single threshold guarantees an accurate threshold search. An interval threshold weak classifier corresponds to two single threshold weak classifiers
and using a strong classifier can effectively increased detection effect by 2.66%
Besides
training time reduced to 624.45 s.
Result
2
The performance of improved algorithm was compared with that of the traditional Adaboost algorithm and is verified by using the MIT(massachusetts Institute of Technology) standard face database. In the experiment
1 500 and 3 000 face and non-face samples were randomly selected for testing. The experimental simulation results of the traditional and proposed Adaboost algorithms were then compared in terms of training time
detection time
detection rate
and false alarm rate. The training time of the improved algrithm was 1.44 times faster than traditional ones. And the detection rate was improved to 94.93%. Both algorithms showed a low detection rate
low false alarm rate
and poor recognition ability with a small number of weak classifiers. However
increasing the number of weak classifiers improved both detection and false alarm rates. In the case of the same detection rate
the improved Adaboost algorithm demonstrated less requirements for weak classifiers
shorter detection time
and 6.03% lower false alarm rate compared with its traditional counterpart. To verify its high practicability
the improved Adaboost algorithm was applied to real face detection
and the experimental results show that this algorithm outperforms the traditional algorithm in terms of detection efficiency
error rate
and detection accuracy. To verify whether the improved Adaboost algorithm is highly advanced
400
1 400
and 2 592 face images were selected from the ORL(Olivetti Research Laboratory)
FERET(face recognition technology)
and CMU Multi-PIE databases
respectively. The performance of the improved Adaboost algorithm was separately tested on these three groups of faces with the popular SVM(support vector machine)
DL(deep learning)
CNN(convolutional neural networks)
and skin color models. The improved algorithm shows a higher detection efficiency compared with the traditional algorithm and has a 95.3% correct detection rate.
Conclusion
2
This paper proposes a dual-threshold Adaboost fast training algorithm that demonstrates efficient face detection
fast training speed
and excellent test results. Experimental results show that under the same cascade structure and number of weak classifiers
the improved Adaboost algorithm is generally superior to the single threshold. The improved algorithm also has shorter face training time requirements
improved face detection accuracy
and better performance and practicability compared with its traditional counterpart. The ROC(receiver operating characteristic) curves show that the improved algorithm achieves better results with fewer weak classifiers. In actual picture detection applications
the improved Adaboost algorithm is superior to the traditional algorithm in terms of training and detection and can meet actual needs. Therefore
it can better meet the actual demand for simultaneous detection of multiple faces in densely populated areas
such as stadiums.
Ahmad J, Muhammad K, Ahmad I, Ahmad W, Smith M L, Smith L N, Jain D K, Wang H X and Mehmood I. 2018. Visual features based boosted classification of weeds for real-time selective herbicide sprayer systems. Computers in Industry, 98:23-33[DOI:10.1016/j.compind.2018.02.005]
Chen G L, Teng H, Zhang F and Yuan S L. 2019. AdaBoost-SVM Based Undergraduates Evaluations//Proceedings of the 2nd International Conference on Informatics, Control and Automation (ICA 2019).[s.l.]: Science and Engineering Research Center: #5[ DOI: 10.12783/dtcse/ica2019/30732 http://dx.doi.org/10.12783/dtcse/ica2019/30732 ]
Chen H L and Chen H Y. 2019. Face recognition algorithm based on VGG network model and SVM. Journal of Physics:Conference Series, 1229(1):#012015[DOI:10.1088/1742-6596/1229/1/012015]
Chen Y M, Bai Y and Li C C. 2018. Fast face detection algorithm based on convolutional neural network. Natural Science Journal of Hainan University, 36(4):324-329
陈益民, 白勇, 黎传琛. 2018.基于卷积神经网络的快速人脸检测算法.海南大学学报(自然科学版), 36(4):324-329)[DOI:10.15886/j.cnki.hdxbzkb.2018.0045]
Cheng C, Kaldy Z and Blaser E. 2019. Two-year-olds succeed at MIT:multiple identity tracking in 20-and 25-month-old infants. Journal of Experimental Child Psychology, 187:#104649[DOI:10.1016/j.jecp.2019.06.002]
Coskun K. 2019. Development of facial emotion recognition and empathy test (FERET) for primary school children. Children Australia, 44(1):23-31[DOI:10.1017/cha.2018.51]
Cui P and Yan T T. 2018. Face detection combining the YCbCr skin color model with improved Adaboost algorithm. Journal of Harbin University of Science and Technology, 23(2):91-96
崔鹏, 燕天天. 2018.融合YCbCr肤色模型与改进的Adaboost算法的人脸检测.哈尔滨理工大学学报, 23(2):91-96)[DOI:10.15938/j.jhust.2018.02.016]
Fhonna R P, Nasution M K M and Tulus. 2018. Real-time detection with AdaBoost-svm combination in various face orientation. Journal of Physics:Conference Series, 978(1):#012015[DOI:10.1088/1742-6596/978/1/012015]
Freund Y, Iyer R D, Schapire R E and Singer Y. 1998. An efficient boosting algorithm for combining preferences//Proceedings of the 15th International Conference on Machine Learning. San Francisco: ACM: 170-178
Gallos L K, Potiguar F Q, Andrade Jr J S and Makse H A. 2013. IMDB network revisited:unveiling fractal and modular properties from a typical small-world network. PLoS One, 8(6):e66443[DOI:10.1371/journal.pone.0066443]
Guo C Z and Zhan Y W. 2018. Fast deformation part model with CNN for face detection//Proceedings of 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018).[s.l.]: Atlantis Press: #65[ DOI: 10.2991/ncce-18.2018.65 http://dx.doi.org/10.2991/ncce-18.2018.65 ]
Guo D F. 2013. The construction of strong classifier algorithm for face detection based on the principal component analysis. Bulletin of Science and Technology, 29(10):220-221+224
郭东峰. 2013.基于主元分析构造强分类器的人脸检测算法.科技通报, 29(10):220-221+224.
Huang H X and Zhang D. 2019. Face liveness detection algorithmbased on deep learning. Application of Electronic Technique, 45(8):44-47
黄海新, 张东. 2019.基于深度学习的人脸活体检测算法.电子技术应用, 45(8):44-47
Hong Z Y, Wang M F, Hou D Q and Yang G L. 2018. A method of driver fatigue detection based on machine vision. Internet of Things Technologies, 8(7):78-79
洪志阳, 王猛飞, 侯东强, 杨国亮. 2018.基于机器视觉的驾驶员疲劳检测方法.物联网技术, 8(7):78-79)[DOI:10.16667/j.issn.2095-1302.2018.07.021]
Junaidy D, Wulandari M and Tanudjaja H. 2019. Real time face detection using Haar-like feature method and local binary pattern method. IOP Conference Series:Materials Science and Engineering, 508(1):#012076[DOI:10.1088/1757-899X/508/1/012076]
Li P, Wang H J, Li Y L and Liu M Y. 2020. Analysis and research based on AdaBoost face detection algorithm. Journal of Beijing Institute of Graphic Communication, 28(01):128-132
李佩, 汪红娟, 李业丽, 刘梦阳. 2020.基于Adaboost算法的人脸检测的研究.北京印刷学院学报, 28(01):128-132
Li X. 2018. Pulmonary Nodule Detection Algorithm Based on Adaboost and Faster-RCNN. Jinan:Shandong University
李夏.2018.基于Adaboost和Faster-RCNN的肺结节检测算法.济南:山东大学
Rajiv S and Soma B. 2019. Machine vision quality assessment for robust face detection. Signal Processing:Image Communication, 72:92-104[DOI:10.1016/j.image.2018.12.012]
Stoyanov S, Asenova K, Kolev P, Vasileva S, Adamova S and Skerleva D. 2018. DISH syndrome in ORL practice. Laryngo-Rhino-Otologie, 97(S2):S18[DOI:10.1055/s-0038-1639778]
Song H L. 2019. Infrared and visible light target tracking based on AdaBoost confidence map. Research on dancing people tracking technology based on contout model and AdaBoost algorithm, Mordern Electronics Technique, 42(15):51-53
宋红莲. 2019.基于轮廓模型和AdaBoost算法的舞蹈人员跟踪技术研究.现代电子技术, 42(15):51-53
Thorne C and Newell M L. 2018. Epidemiology of HIV infection in the newborn. Early Human Development, 58(1):1-16[DOI:10.1016/S0378-3782(00)00049-9]
Xia L L, Zhao Y, Ma W J, Cong J Y and Xiao J L. 2018. An Adaboost optimization based information fusion algorithm for MIMUs/GPS. Electronics Optics and Control, 25(12):21-25
夏琳琳, 赵耀, 马文杰, 丛靖宇, 肖建磊. 2018.一种Adaboost优化的MIMUs/GPS信息融合算法.电光与控制, 25(12):21-25)[DOI:10.3969/j.issn.1671-637X.2018.12.005]
Zhang S, Gao M W and Hu Q H. 2018. Research and application of adaboost based prediction of student's academic achievement//Proceedings of the 3rd Annual International Conference on Education and Development (ICED 2018).[s.l.]: Atlantis Press: #35[ DOI: 10.2991/iced-18.2018.35 http://dx.doi.org/10.2991/iced-18.2018.35 ]
Zhao Y B. 2018. Adaboost based gait analysis and its application in the diagnosis of knee osteo arthritis. Guangzhou:Guangdong University of Technology
赵宇斌. 2018.基于Adaboost的步态分析及在膝骨性关节炎诊断的应用.广州:广东工业大学
Zhu Z M and Qiao J. 2017. Research of preceding vehicle identification based on Haar-like features and Adaboost algorithm. Electronic Measurement Technology, 40(05):180-184
朱志明, 乔洁. 2017.基于Haar-like特征与Adaboost算法的前方车辆辨识技术研究.电子测量技术, 40(05):180-184
相关文章
相关作者
相关机构
京公网安备11010802024621
