人脸检测及跟踪算法的研究
|
摘要
复杂背景下实时人脸的检测与跟踪属于生物特征识别技术的研究领域，包括图像获取、图像预处理、人脸的检测定位、人脸的跟踪等几个方面。
文中首先介绍了图像预处理的各种方法，重点介绍了图像分割和人脸检测定位，针对实际应用的要求，着重处理了两种情况的人像检测定位：(1)在灰度图像下对连续输入的多帧进行差分，对于得到的边缘进行处理初步确定人像的大概位置，然后采用模板匹配进行精确定位。(2)应用双模板匹配的方法，由多人脸平均得到人脸模板和眼睛模板，在检测中采取了首先使用尺度相对较小的双眼模板搜索候选人脸，再用人脸模板匹配进一步筛选候选人脸的方法。在人脸跟踪的算法中，采用了在差分图像粗定位的基础上，在一个确定的小范围内利用双模板匹配进行跟踪。在相对复杂的背景下，对图像尺度为320×240，检测的速度为每秒6到10帧，而定位的准确率可以达到75％左右。最后在已有人脸检测算法的基础上，提高检测的速度。本文实验所用的图像均是由CCD摄像头在自然光下实时采集的，文中的算法由VC语言在Windows 98平台下编程实现。
Real-time face detection and face tracking in complex background is a study field of Biometric features identification technique, It includes image obtain, image pretreatment, face detection and face tracking etc.
In this paper we first introduce some methods of image pretreatment, then introduce some face detection methods. Aims at application, we mainly discuss face detection in two circumstance.(1)In the gray image, we use difference image in multi-frames, then get the edge of the face ,we make the raw-location to ensure the approximate position of the face, next ,we use the template of the face to occlude the pseudo face region and get the region which near the exact face region.(2)Use method of two template matching, first, we get the face template from many face average, second, we copy the eye section of the face template then get the eye template. In the course of face detection, we first use eye template to search face in image, if the result meet our request, then we can use the face template to search face further. Among the face tracking algorithm, we first use difference image to gain the raw-location, then in the limited position, we use two template matching to tracking face. In a relative complex background, with image size of 320 X 240,our system can get a speed of six to ten image per second, and the accurate rate is around 75%.
In this paper, CCD vidicon grabbed the tracked images and all algorithms were accomplished by VC++ in Windows 98 environment.
文中首先介绍了图像预处理的各种方法，重点介绍了图像分割和人脸检测定位，针对实际应用的要求，着重处理了两种情况的人像检测定位：(1)在灰度图像下对连续输入的多帧进行差分，对于得到的边缘进行处理初步确定人像的大概位置，然后采用模板匹配进行精确定位。(2)应用双模板匹配的方法，由多人脸平均得到人脸模板和眼睛模板，在检测中采取了首先使用尺度相对较小的双眼模板搜索候选人脸，再用人脸模板匹配进一步筛选候选人脸的方法。在人脸跟踪的算法中，采用了在差分图像粗定位的基础上，在一个确定的小范围内利用双模板匹配进行跟踪。在相对复杂的背景下，对图像尺度为320×240，检测的速度为每秒6到10帧，而定位的准确率可以达到75％左右。最后在已有人脸检测算法的基础上，提高检测的速度。本文实验所用的图像均是由CCD摄像头在自然光下实时采集的，文中的算法由VC语言在Windows 98平台下编程实现。
Real-time face detection and face tracking in complex background is a study field of Biometric features identification technique, It includes image obtain, image pretreatment, face detection and face tracking etc.
In this paper we first introduce some methods of image pretreatment, then introduce some face detection methods. Aims at application, we mainly discuss face detection in two circumstance.(1)In the gray image, we use difference image in multi-frames, then get the edge of the face ,we make the raw-location to ensure the approximate position of the face, next ,we use the template of the face to occlude the pseudo face region and get the region which near the exact face region.(2)Use method of two template matching, first, we get the face template from many face average, second, we copy the eye section of the face template then get the eye template. In the course of face detection, we first use eye template to search face in image, if the result meet our request, then we can use the face template to search face further. Among the face tracking algorithm, we first use difference image to gain the raw-location, then in the limited position, we use two template matching to tracking face. In a relative complex background, with image size of 320 X 240,our system can get a speed of six to ten image per second, and the accurate rate is around 75%.
In this paper, CCD vidicon grabbed the tracked images and all algorithms were accomplished by VC++ in Windows 98 environment.
引文
1. Ching W.S. A novel change detection algorithm using adaptive threshold. Image and Vision Computing. 1994, 12(7): 459-463
2. Fathy M.Siyal M Y. An image detection technique based on morphological edge detection and background differencing for real-time traffic analysis. Pattern Recognition Letters 16, 1995:1321-1350
3. Yang G, Huang T.Human face detection in a complex background[j]. Pattern Recognition, 1994,27(1):53-63.
4. Brunelli R, Poggio T.Face recognition: features versus templates[j]. IEEE Trans. PAMI, 1993,15(10):1042-1052
5. Julle P, Marsh R. A hierarchical neural network for human detection. Pattern Recognition, 1996,29(5):781-787
6. Rowley H A, Baluja S, Kanade T.Neural network-based face detection. IEEE Trans. PAMI, 1998,20(1):23-28
7. Sobottka K, Pitas I. A novel method for automatic face segmentation, facial feature extraction and tracking[j]. Signal Processing: Image Communication, 1998,12(3);263-281.
8. Herodotou N, Plataniotis K N, Venetsanopoulos A N. Automatic location and tracking of the facial region in color video sequences[J], Signal Processing: Image Communicaion, 1999,14(5):359-388.
9. Alex(Sandy) Pentland, Tanzeem Choudhury. "Face Recognition for Smart Environments". IEEE Computer, February 2000, p50-p55
10. Robert W.Frischholz, Ulrich Dieckmann, "BioID: A Multimodal Biometric Identification System", IEEE Computer, February 2000, p1-p6
11. Kah-Kay Sung, Tomaso Poggio, "Example-based Learning for View-based Human Face Detection", Massachusetts Institute of Technology, A.I.Memo No.1521, C.B.C.L Paper No.112,1994.
12. Surendra Ranganath, Krishnamurthy Arun, "Face Recognition Using Transform Features And Neural Networks", Pattern Recognition, Vol.30. No.10, pp.1615-1622, 1997
13. Qi Bin Sun, Chian Prong Lam and Jian Kang Wu, A Practical Automatic Face Recognition System. RWC-ISS Lab
14. Lee, C.H. Kim, J.S., Park, K.H. Automatic human face location in a complex background using motion and color information. Pattern recognition, 1996,29(11):1877-1889.
15. Kapfer, M., Benois-Pineau J.Detection of human faces in color image sequences with arbitrary motions for very low bitrate videophone coding Pattern Recognition Letters 1997,18(14):1503-1518
16. Dai, Y., Nakano, Y. Face-Texture model based on SGLD and its application in face detection in a color scene. Pattern recognition, 1996,29(6):1007-1017
17. Yang, M.H., Ahuja, N.Detecting human faces in color images. Proceedings of the international Conference on Imaging processing, vol 1,1998,127-130
18. Schwerd, K., Crowlet, J.L. robust face tracking using color. In:Proceedings of the 4th IEEE International Conference on Automatic Face and Gesture Recognition. 2000. 90-95
19. Garcia, C., Tzirilas, G. Face detection using quantized skin color regions merging and wavelet packet analysis. IEEE Trans. 1999,1(3):264-277.
20. S.Y.LEE, Y.K.HAM and R. -H. PARK, Recognition of Human Front Faces Using Knowledge-Based Feature Extraction and Neuro-Fuzzy Algorithm, Pattern Recognition, Vol.29, No.11, pp.1863-1876,1996.
21. Chellapps R.Wilson C L, Sirohey S.Human and machine recognition of faces: A survey, Proc. of The TEEE, 1995,83(5):705-740
22. Agui T.Kokubo Y. et al. Extraction of face regions from monochromatic photographs using neural networks. ICARCVz' 92,1992
23. Sugiyama Y, ArikiY. Facial region tracking and recognition by subspace projection, Proc. of Virtual Sys. And MultiMedia, Gifa, Japan. 1996:18-20
24．徐士良 《计算机常用算法》，北京：清华大学出版社，第二版，137～146
25．王国印译，《Visual C++技术内幕》北京：清华大学出版社，1994
26．Kenneth. R.Castleman 著，朱志刚等译，《数字图像处理》北京：电子工业出版社，1998年9月第一版
27．何斌 马天予 王云坚 等，《Visual C++ 数字图象处理》北京：人民邮电出版社，2001
28．梁路宏 艾海舟，《基于人脸检测的人脸跟踪算法》，计算机工程与应用，2001，17：41～45
29．艾海舟 王栓 何克忠，《基于差分图像的人脸检测》，中国图象图形学报，1998，3：987～992
30．马丽红 余英林 张宇，《复杂背景中的人脸特征定位与提取》，数据采集与处理，2000，15(1)：33～38
31．王栓 艾海舟 何克忠，《基于差分图象的多运动目标的检测与跟踪》，中国图象图形学报，1999，4(6)：470～475
32．梁路宏 艾海舟 何克忠，《基于多关联模板匹配的人脸检测》，软件学报，2001，12：94～102
33．邢昕 汪孔桥 沈兰荪，《基于器官跟踪的人脸实时跟踪方法》，电子学报，2000，6：29～31
34．彭振云 陶霖密 张宏江，《基于颜色分割和KL变换的人脸特征检测》，清华大学学报，2001，41(4／5)：218～221
35．马丽红 余英林 张宇，《人脸特征提取及自动定位的研究》，华南理工大学学报，2000，28(4)：96～102
2. Fathy M.Siyal M Y. An image detection technique based on morphological edge detection and background differencing for real-time traffic analysis. Pattern Recognition Letters 16, 1995:1321-1350
3. Yang G, Huang T.Human face detection in a complex background[j]. Pattern Recognition, 1994,27(1):53-63.
4. Brunelli R, Poggio T.Face recognition: features versus templates[j]. IEEE Trans. PAMI, 1993,15(10):1042-1052
5. Julle P, Marsh R. A hierarchical neural network for human detection. Pattern Recognition, 1996,29(5):781-787
6. Rowley H A, Baluja S, Kanade T.Neural network-based face detection. IEEE Trans. PAMI, 1998,20(1):23-28
7. Sobottka K, Pitas I. A novel method for automatic face segmentation, facial feature extraction and tracking[j]. Signal Processing: Image Communication, 1998,12(3);263-281.
8. Herodotou N, Plataniotis K N, Venetsanopoulos A N. Automatic location and tracking of the facial region in color video sequences[J], Signal Processing: Image Communicaion, 1999,14(5):359-388.
9. Alex(Sandy) Pentland, Tanzeem Choudhury. "Face Recognition for Smart Environments". IEEE Computer, February 2000, p50-p55
10. Robert W.Frischholz, Ulrich Dieckmann, "BioID: A Multimodal Biometric Identification System", IEEE Computer, February 2000, p1-p6
11. Kah-Kay Sung, Tomaso Poggio, "Example-based Learning for View-based Human Face Detection", Massachusetts Institute of Technology, A.I.Memo No.1521, C.B.C.L Paper No.112,1994.
12. Surendra Ranganath, Krishnamurthy Arun, "Face Recognition Using Transform Features And Neural Networks", Pattern Recognition, Vol.30. No.10, pp.1615-1622, 1997
13. Qi Bin Sun, Chian Prong Lam and Jian Kang Wu, A Practical Automatic Face Recognition System. RWC-ISS Lab
14. Lee, C.H. Kim, J.S., Park, K.H. Automatic human face location in a complex background using motion and color information. Pattern recognition, 1996,29(11):1877-1889.
15. Kapfer, M., Benois-Pineau J.Detection of human faces in color image sequences with arbitrary motions for very low bitrate videophone coding Pattern Recognition Letters 1997,18(14):1503-1518
16. Dai, Y., Nakano, Y. Face-Texture model based on SGLD and its application in face detection in a color scene. Pattern recognition, 1996,29(6):1007-1017
17. Yang, M.H., Ahuja, N.Detecting human faces in color images. Proceedings of the international Conference on Imaging processing, vol 1,1998,127-130
18. Schwerd, K., Crowlet, J.L. robust face tracking using color. In:Proceedings of the 4th IEEE International Conference on Automatic Face and Gesture Recognition. 2000. 90-95
19. Garcia, C., Tzirilas, G. Face detection using quantized skin color regions merging and wavelet packet analysis. IEEE Trans. 1999,1(3):264-277.
20. S.Y.LEE, Y.K.HAM and R. -H. PARK, Recognition of Human Front Faces Using Knowledge-Based Feature Extraction and Neuro-Fuzzy Algorithm, Pattern Recognition, Vol.29, No.11, pp.1863-1876,1996.
21. Chellapps R.Wilson C L, Sirohey S.Human and machine recognition of faces: A survey, Proc. of The TEEE, 1995,83(5):705-740
22. Agui T.Kokubo Y. et al. Extraction of face regions from monochromatic photographs using neural networks. ICARCVz' 92,1992
23. Sugiyama Y, ArikiY. Facial region tracking and recognition by subspace projection, Proc. of Virtual Sys. And MultiMedia, Gifa, Japan. 1996:18-20
24．徐士良 《计算机常用算法》，北京：清华大学出版社，第二版，137～146
25．王国印译，《Visual C++技术内幕》北京：清华大学出版社，1994
26．Kenneth. R.Castleman 著，朱志刚等译，《数字图像处理》北京：电子工业出版社，1998年9月第一版
27．何斌 马天予 王云坚 等，《Visual C++ 数字图象处理》北京：人民邮电出版社，2001
28．梁路宏 艾海舟，《基于人脸检测的人脸跟踪算法》，计算机工程与应用，2001，17：41～45
29．艾海舟 王栓 何克忠，《基于差分图像的人脸检测》，中国图象图形学报，1998，3：987～992
30．马丽红 余英林 张宇，《复杂背景中的人脸特征定位与提取》，数据采集与处理，2000，15(1)：33～38
31．王栓 艾海舟 何克忠，《基于差分图象的多运动目标的检测与跟踪》，中国图象图形学报，1999，4(6)：470～475
32．梁路宏 艾海舟 何克忠，《基于多关联模板匹配的人脸检测》，软件学报，2001，12：94～102
33．邢昕 汪孔桥 沈兰荪，《基于器官跟踪的人脸实时跟踪方法》，电子学报，2000，6：29～31
34．彭振云 陶霖密 张宏江，《基于颜色分割和KL变换的人脸特征检测》，清华大学学报，2001，41(4／5)：218～221
35．马丽红 余英林 张宇，《人脸特征提取及自动定位的研究》，华南理工大学学报，2000，28(4)：96～102