交通视频中视点无关目标分类与检索方法研究
|
摘要
随着网络、通信和微电子技术的飞速发展,一些特定功能的视觉分析系统以其直观、方便和内容丰富等特点,日益受到人们的亲睐,其中交通监控领域应用最为广泛。然而,全天候监控捕获的大量视频信息,若采用人工搜索方法来寻找目标,不仅低效,由人为因素造成的失误亦是难以避免。因此,人们希望计算机能具有类似人类视觉系统的能力,可以分析、理解图像或视频的内容,以实现视频分析系统的智能化、实用化,视频分析技术应运而生。
视频分析技术主要是处理包含各种运动目标的视频序列,从场景中检测、跟踪、分类识别目标,并对其行为进行理解和描述。其中,目标分类是基于视频的运动分析课题中的一个重要方面,其研究内容是在运动检测和跟踪的基础上,依据提取的运动目标区域形状特征和运动属性,对运动目标区域进行语义上的分类。目标分类技术研究对更高层次的视频理解技术的发展有重要意义。
目前国外研究机构和国内高校在目标分类技术上取得了一定进展,但仍存在一些应用上的限制和不足,其中目标分类的视点依赖性问题是影响分类稳定性的主要因素。所谓视点依赖性是指目标的2D特征在投影到图像平面时发生了透视形变,从而导致其无法准确的用于分类。本文围绕运动目标分类及其在交通视频目标检索中的应用这一课题,重点阐述分类过程中视点依赖性问题的解决方法,并对各类相关技术进行了研究,具有重要的理论意义和实际价值。本文的工作主要分以下几个部分:
(1)介绍目标分类领域基础理论及相关研究,包括目标的特征表达、目标分类方法、运动目标检测与跟踪以及场景知识在目标分类中的作用。
(2)提出了一种基于kalman预估模型和最大化后验概率匹配的粘连目标跟踪方法,实现了目标相互遮挡时连续稳定跟踪。
(3)针对目标分类中遇到的视点依赖性问题,介绍当前具有代表性的三种目标2D特征透视变形的矫正方法,并提出了基于地平面矫正的目标2D特征恢复算法。在标准化后特征的基础上,采用多类支持向量机实现视点无关运动目标分类。
(4)研究基于目标特征和语义类别的运动目标检索方法,对目标特征数据组织形式、目标检索方式、查询结果显示方式等进行了探讨;
(5)本文搭建了基于Visual C++平台和OpenCV图像处理库的实验环境,利用程序验证了提出了的各类算法,实验证明本文算法实现了不同视点下运动目标的准确分类。
As the rapid development of the network, communications,microelectronic technology,a number of visual analysis system with specific function get the growing popularity of the pro-gaze for its intuitive, convenient and rich in content. Traffic monitoring is the most widely used area. However, the all-weather surveillance system captured a large number of video information, in which it is inefficient and mistakable to find the target with the use of pure-manual search methods. Therefore, it was hoped that the computer can analyze and understand video content, in order to achieve video analysis intelligent and practical system, video analysis techniques have emerged.
Video analysis techniques deal primarily with video sequences which contain a variety of moving target, and detect, track, and classify and recognition goals, then understand and describe their behavior. Among them, target classification is an important aspect of the Video analysis with the content of classifying the object area based on motion analysis by their features of shape and motion, and is important to the development of high level video understanding techniques.
Currently overseas research institutions and national universities have made some progress in the target classification technology, but there are still some constraints and lack of applications, in which the view-dependent issue of classification is the main factor affecting the stability of classification. The so-called view-dependent issue means 2D feature of object has some distortion because of projection, causing it is not correct for classification. The major work is described as follows:
(1) Introduce basic theory of target classification and related research, including the expression of object features, common methods of object classification, moving target detection and tracking, and the use of scene knowledge.
(2) The predict model based on kalman filter combine with maximum posterior probability for object matching is proposed to track moving objects with occlusions.
(3) For the view-dependent problems encountered in target classification, Describes three representative kinds of 2D features rectification methods,and then promote 2D features recovery algorithm based on ground plane rectification. Based on the normalized characteristics, using Multi-Class Support Vector Machines (SVM) to achieve view-independent classification of moving object.
(4) Research on moving object retrieval method based on video features,discuss the organization of feature data,the way of object retrieval and display form of retrival results.
(5) By building the experiment platform based on Visual C++ and OpenCV graphics library. We verify the various algorithms with coding. Results shows the algorithm solve the view-dependent problems in object classification and achieve view-independent object classification.
视频分析技术主要是处理包含各种运动目标的视频序列,从场景中检测、跟踪、分类识别目标,并对其行为进行理解和描述。其中,目标分类是基于视频的运动分析课题中的一个重要方面,其研究内容是在运动检测和跟踪的基础上,依据提取的运动目标区域形状特征和运动属性,对运动目标区域进行语义上的分类。目标分类技术研究对更高层次的视频理解技术的发展有重要意义。
目前国外研究机构和国内高校在目标分类技术上取得了一定进展,但仍存在一些应用上的限制和不足,其中目标分类的视点依赖性问题是影响分类稳定性的主要因素。所谓视点依赖性是指目标的2D特征在投影到图像平面时发生了透视形变,从而导致其无法准确的用于分类。本文围绕运动目标分类及其在交通视频目标检索中的应用这一课题,重点阐述分类过程中视点依赖性问题的解决方法,并对各类相关技术进行了研究,具有重要的理论意义和实际价值。本文的工作主要分以下几个部分:
(1)介绍目标分类领域基础理论及相关研究,包括目标的特征表达、目标分类方法、运动目标检测与跟踪以及场景知识在目标分类中的作用。
(2)提出了一种基于kalman预估模型和最大化后验概率匹配的粘连目标跟踪方法,实现了目标相互遮挡时连续稳定跟踪。
(3)针对目标分类中遇到的视点依赖性问题,介绍当前具有代表性的三种目标2D特征透视变形的矫正方法,并提出了基于地平面矫正的目标2D特征恢复算法。在标准化后特征的基础上,采用多类支持向量机实现视点无关运动目标分类。
(4)研究基于目标特征和语义类别的运动目标检索方法,对目标特征数据组织形式、目标检索方式、查询结果显示方式等进行了探讨;
(5)本文搭建了基于Visual C++平台和OpenCV图像处理库的实验环境,利用程序验证了提出了的各类算法,实验证明本文算法实现了不同视点下运动目标的准确分类。
As the rapid development of the network, communications,microelectronic technology,a number of visual analysis system with specific function get the growing popularity of the pro-gaze for its intuitive, convenient and rich in content. Traffic monitoring is the most widely used area. However, the all-weather surveillance system captured a large number of video information, in which it is inefficient and mistakable to find the target with the use of pure-manual search methods. Therefore, it was hoped that the computer can analyze and understand video content, in order to achieve video analysis intelligent and practical system, video analysis techniques have emerged.
Video analysis techniques deal primarily with video sequences which contain a variety of moving target, and detect, track, and classify and recognition goals, then understand and describe their behavior. Among them, target classification is an important aspect of the Video analysis with the content of classifying the object area based on motion analysis by their features of shape and motion, and is important to the development of high level video understanding techniques.
Currently overseas research institutions and national universities have made some progress in the target classification technology, but there are still some constraints and lack of applications, in which the view-dependent issue of classification is the main factor affecting the stability of classification. The so-called view-dependent issue means 2D feature of object has some distortion because of projection, causing it is not correct for classification. The major work is described as follows:
(1) Introduce basic theory of target classification and related research, including the expression of object features, common methods of object classification, moving target detection and tracking, and the use of scene knowledge.
(2) The predict model based on kalman filter combine with maximum posterior probability for object matching is proposed to track moving objects with occlusions.
(3) For the view-dependent problems encountered in target classification, Describes three representative kinds of 2D features rectification methods,and then promote 2D features recovery algorithm based on ground plane rectification. Based on the normalized characteristics, using Multi-Class Support Vector Machines (SVM) to achieve view-independent classification of moving object.
(4) Research on moving object retrieval method based on video features,discuss the organization of feature data,the way of object retrieval and display form of retrival results.
(5) By building the experiment platform based on Visual C++ and OpenCV graphics library. We verify the various algorithms with coding. Results shows the algorithm solve the view-dependent problems in object classification and achieve view-independent object classification.
引文
[1] R. C. Gonzalez and R. E. Woods. Digital Image Processing. Englewood Cliffs, NJ: Prentice -Hall, 2002.
[2] Ming-Kuei Hu. Visual Pattern Recognition by Moment Invariants. Information Theory, IRE Transactions.1962:179-187.
[3] Michael J. Swain, Dana H. Ballard. Color indexing. International Journal of Computer Vision 1991: 11-32.
[4]冯桂,卢健,林宗坚.图像直方图不变特征在影像匹配定位中的应用[J].计算机辅助设计与图形学学报.2000第二期:146-148
[5] Smith, J.R., Shih-Fu Chang. Single color extraction and image query. International Conference of Image Processing. 1995:528-531.
[6] Grag Pass, Ramin Zabih. Comparing Images Using Color Coherence Vectors. Proceedings of the fourth ACM international conference on Multimedia. 1997:65-73.
[7] Jing Huang, S Ravi Kumar. Image Indexing Using Color Correlograms. Computer Vision and Pattern Recognition, 1997:762-768.
[8] M. Stricker, M. Orengo. Similarity of Color Images. Proc. SPIE Vol. 2420, 1995:381-392.
[9] Tuceryan M, Jain A K. Texture Analysis, Handbook Pattern Recognition and Computer Vision [M]. Singapore: World Scientific, 1993:235-276.
[10] Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification [J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973, 3 (6): 610-621.
[11] Bruce H. McCormick and Sadali N. Jayaramamurthy. Time series model for texture synthesis. International Journal of Parallel Programming. 1974, 3 (4): 329-343.
[12]毛建昌,王成道,万嘉若.多分辨率自回归纹理模型[J].电子学报, 1988, 16 (6) : 63-68.
[13] Mao J, Jain A K. Texture classification and segmentation using Multiresolution Simultaneous Autoregressive Models [ J ]. Pattern Recognition, 1992, 25 (2): 173-188.
[14] Cross G R, Jain A K. Markov random field texture models [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1983,5 (1) : 25-39.
[15] German S, German D. Stochastic relaxation, Gibbs distribution and bayesian restoration of images [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1984, 6 (6) : 721-741.
[16] Chitre Y, Dhawan A P. M2band wavelet discrimination of natural textures [J]. Pattern Recognition, 1999, 32 (5): 773-789.
[17]章毓晋.图像工程(上册)一图像处理与分析[M].北京:清华大学出版社,1999年3月.
[18] B.Horn, B.Schunck. Determining optical flow. Artificial Intelligence, 1981:185-203.
[19] Robert Collins, Alan Lipton, Takeo Kanade et. A system for video surveillance and moni- toring: VSAM final report CMU-RI-TR-00-12. Carnegie Mellon University, May, 2000.
[20] Alan J. Lipton Hironobu Fujiyoshi. Moving object classification and tracking from real-time video. Applications of Computer Vision, 1998:8-14
[21] Kuno Y., Watanabe T. Automated detection of human for visual surveillance system. International Conference of Pattern Recognition. 1996: 865-869.
[22] Daniel Toth and Til Aach. Detection and recognition of moving objects using statistical mo- tion detection and Fourier descriptors [J]. Image Analysis and Processing, 2003: 430– 435.
[23] Razali M.T., Adznan B.J. Detection and Classification of Moving Object for Smart Vision Sensor. Information and Communication Technologies, 2006: 733 -737.
[24] Jin Soo Noh and Kang Hyeon Rhee. Palmprint Identification Algorithm Using Hu Invariant Moments [J]. Fuzzy Systems and Knowledge Discovery. 2005.7:91-95.
[25] Cutler R, Davis L.S. Robust real-time periodic motion detection, analysis, and applications. Pattern Analysis and Machine Intelligence. 2000:781-796.
[26] Alan J Lipton. Local Application of Optic Flow to Analyse Rigid versus Non-Rigid Motion. International Journal of Computer Vision. 2004:255-279.
[27] Zhaoxiang Zhang, Yinghao Cai, Kaiqi Huang, Tieniu Tan. Real-Time Moving Object Classification with Automatic Scene Division. ICIP, 2007: 149-152.
[28] Y Bogomolov, G Dror, S Lapchev, E Rivlin, and M. Rudzsky. Classification of Moving Objects Based on Motion and Appearance. Proceedings of the British Machine Vision Conference, 2003: 429-438.
[29] Anderso, Peter B. Change detection and tracking using pyramid transform techniques. In proceedings of SPIE-Intelligent Robots and Computer Vision, 1985:72-78.
[30] J. L. Barron. Performance of optical flow techniques. International Journal of Computer Vision. 2005:43-77
[31] Horn and Schunck. Determining optical flow. Artificial Intelligence. 1981:185-203.
[32] Zhaoxiang Zhang, Min Li. View Independent Object Classification Based on Automated Ground Plane Rectification for Traffic Scene Surveillance. In Proc. of the 2008 IEEE International Workshop on Visual Surveillance. Marseille, France, 2008.10: 1364-1367.
[33] Zhaoxiang Zhang, Yinghao Cai, Kaiqi Huang, Tieniu Tan. Real-Time Moving Object Classification with Automatic Scene Division. ICIP, 2007: 149-152.
[34] L. M. Brown. View independent vehicle/person classification. In Proc. of the ACM 2nd international workshop on Video Surveillance and Sensor Networks, 2004: 114-123.
[35] Torralba, A. Sinha P. Statistical context priming for object detection. International Conference of Computer Vision, 2001:763-770.
[36] Y Bogomolov, G Dror, S Lapchev, E Rivlin, and M. Rudzsky. Classification of Moving Objects Based on Motion and Appearance. Proceedings of the British Machine Vision Conference, 2003: 429-438.
[37] Hampapur, A., et al., IBM Smart Surveillance System, IEEE Computer Vision and Pattern Recognition, Demonstration, Washington, DC, June 28-30, 2004.
[38] R.I.Hartley and Zizzerman. Multiple view geometry in computer vision. Cambridge University Press. 2000.
[39] Liebowitz D, Zisserman A. Metric rectification for perspective images of planes. Computer Vision and Pattern Recognition. 1998:482-488.
[40] Zhou Liu, Kaiqi Huang, and Tieniu Tan. Cast shadow removal with gmm for surface reflectance component. in Proceedings of 18th IEEE International Conference of Pattern Recognition, 2006.
[41] KaewTraKulPong P, Bowden R. An improved adaptive backgroundmixture model for real-time tracking with shadow detection [ C ]. Proceedings of the 2nd European Workshop on Advanced VideoBased Surveillance Systems. USA: Kluwer Academic Publishers, 2001: 135-144.
[42] Razali M.T., Adznan B.J. Detection and Classification of Moving Object for Smart Vision Sensor. Information and Communication Technologies, 2006. On page(s): 733 -737.
[43]冯祖仁,吕娜,李良福等.基于最大后验概率的图像匹配相似性指标研究[J].自动化学报. 2007.33卷01期: 1-8.
[44] I. W. Faig. Calibration of Close Range Photogrammetric System: Mathematical Formulation [J]. Photogrammeric Eng. Remote Sensing, VOL. 41, PP. 1479-1486, 1975.
[45] R. Y. Tsai. An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision[C],Proc. of IEEE Conference of Computer Vision and Pattern Recognition, 1986:364-374,
[46] Arun Hampapur, Lisa Brown.et.The IBM Smart Surveillance System.
[47] Ying-li Tian, Lisa Brown. IBM smart surveillance system (S3): event based video surveillance system with an open and extensible framework.. Machine Vision and Applications. 2008:315-327.
[48] Yingli Tian. the IBM smart surveillance system release 1. Wireless and Optical Communications. 2005:98-.
[49]边肇棋,张学工.模式识别第二版.北京:清华大学出版社,2000, 284-299.
[50] John C Platt, Nello Cristianini, John Shawe-taylor. Large Margin DAGs for Multiclass Classification. Advances in Neural Information Processing Systems. 2005:547-553.
[51]吴成东等.基于误差修正码的支持向量机大类别分类方法[J] .沈阳建筑工程学院学报(自然科学版) ,2004 ,20 (1) :66-70.
[52] Thomas G. Dietterich, Ghulum Bakiri. Solving Multiclass Learning Problems via Error- Correcting Output Codes. 1995:263-286.
[2] Ming-Kuei Hu. Visual Pattern Recognition by Moment Invariants. Information Theory, IRE Transactions.1962:179-187.
[3] Michael J. Swain, Dana H. Ballard. Color indexing. International Journal of Computer Vision 1991: 11-32.
[4]冯桂,卢健,林宗坚.图像直方图不变特征在影像匹配定位中的应用[J].计算机辅助设计与图形学学报.2000第二期:146-148
[5] Smith, J.R., Shih-Fu Chang. Single color extraction and image query. International Conference of Image Processing. 1995:528-531.
[6] Grag Pass, Ramin Zabih. Comparing Images Using Color Coherence Vectors. Proceedings of the fourth ACM international conference on Multimedia. 1997:65-73.
[7] Jing Huang, S Ravi Kumar. Image Indexing Using Color Correlograms. Computer Vision and Pattern Recognition, 1997:762-768.
[8] M. Stricker, M. Orengo. Similarity of Color Images. Proc. SPIE Vol. 2420, 1995:381-392.
[9] Tuceryan M, Jain A K. Texture Analysis, Handbook Pattern Recognition and Computer Vision [M]. Singapore: World Scientific, 1993:235-276.
[10] Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification [J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973, 3 (6): 610-621.
[11] Bruce H. McCormick and Sadali N. Jayaramamurthy. Time series model for texture synthesis. International Journal of Parallel Programming. 1974, 3 (4): 329-343.
[12]毛建昌,王成道,万嘉若.多分辨率自回归纹理模型[J].电子学报, 1988, 16 (6) : 63-68.
[13] Mao J, Jain A K. Texture classification and segmentation using Multiresolution Simultaneous Autoregressive Models [ J ]. Pattern Recognition, 1992, 25 (2): 173-188.
[14] Cross G R, Jain A K. Markov random field texture models [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1983,5 (1) : 25-39.
[15] German S, German D. Stochastic relaxation, Gibbs distribution and bayesian restoration of images [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1984, 6 (6) : 721-741.
[16] Chitre Y, Dhawan A P. M2band wavelet discrimination of natural textures [J]. Pattern Recognition, 1999, 32 (5): 773-789.
[17]章毓晋.图像工程(上册)一图像处理与分析[M].北京:清华大学出版社,1999年3月.
[18] B.Horn, B.Schunck. Determining optical flow. Artificial Intelligence, 1981:185-203.
[19] Robert Collins, Alan Lipton, Takeo Kanade et. A system for video surveillance and moni- toring: VSAM final report CMU-RI-TR-00-12. Carnegie Mellon University, May, 2000.
[20] Alan J. Lipton Hironobu Fujiyoshi. Moving object classification and tracking from real-time video. Applications of Computer Vision, 1998:8-14
[21] Kuno Y., Watanabe T. Automated detection of human for visual surveillance system. International Conference of Pattern Recognition. 1996: 865-869.
[22] Daniel Toth and Til Aach. Detection and recognition of moving objects using statistical mo- tion detection and Fourier descriptors [J]. Image Analysis and Processing, 2003: 430– 435.
[23] Razali M.T., Adznan B.J. Detection and Classification of Moving Object for Smart Vision Sensor. Information and Communication Technologies, 2006: 733 -737.
[24] Jin Soo Noh and Kang Hyeon Rhee. Palmprint Identification Algorithm Using Hu Invariant Moments [J]. Fuzzy Systems and Knowledge Discovery. 2005.7:91-95.
[25] Cutler R, Davis L.S. Robust real-time periodic motion detection, analysis, and applications. Pattern Analysis and Machine Intelligence. 2000:781-796.
[26] Alan J Lipton. Local Application of Optic Flow to Analyse Rigid versus Non-Rigid Motion. International Journal of Computer Vision. 2004:255-279.
[27] Zhaoxiang Zhang, Yinghao Cai, Kaiqi Huang, Tieniu Tan. Real-Time Moving Object Classification with Automatic Scene Division. ICIP, 2007: 149-152.
[28] Y Bogomolov, G Dror, S Lapchev, E Rivlin, and M. Rudzsky. Classification of Moving Objects Based on Motion and Appearance. Proceedings of the British Machine Vision Conference, 2003: 429-438.
[29] Anderso, Peter B. Change detection and tracking using pyramid transform techniques. In proceedings of SPIE-Intelligent Robots and Computer Vision, 1985:72-78.
[30] J. L. Barron. Performance of optical flow techniques. International Journal of Computer Vision. 2005:43-77
[31] Horn and Schunck. Determining optical flow. Artificial Intelligence. 1981:185-203.
[32] Zhaoxiang Zhang, Min Li. View Independent Object Classification Based on Automated Ground Plane Rectification for Traffic Scene Surveillance. In Proc. of the 2008 IEEE International Workshop on Visual Surveillance. Marseille, France, 2008.10: 1364-1367.
[33] Zhaoxiang Zhang, Yinghao Cai, Kaiqi Huang, Tieniu Tan. Real-Time Moving Object Classification with Automatic Scene Division. ICIP, 2007: 149-152.
[34] L. M. Brown. View independent vehicle/person classification. In Proc. of the ACM 2nd international workshop on Video Surveillance and Sensor Networks, 2004: 114-123.
[35] Torralba, A. Sinha P. Statistical context priming for object detection. International Conference of Computer Vision, 2001:763-770.
[36] Y Bogomolov, G Dror, S Lapchev, E Rivlin, and M. Rudzsky. Classification of Moving Objects Based on Motion and Appearance. Proceedings of the British Machine Vision Conference, 2003: 429-438.
[37] Hampapur, A., et al., IBM Smart Surveillance System, IEEE Computer Vision and Pattern Recognition, Demonstration, Washington, DC, June 28-30, 2004.
[38] R.I.Hartley and Zizzerman. Multiple view geometry in computer vision. Cambridge University Press. 2000.
[39] Liebowitz D, Zisserman A. Metric rectification for perspective images of planes. Computer Vision and Pattern Recognition. 1998:482-488.
[40] Zhou Liu, Kaiqi Huang, and Tieniu Tan. Cast shadow removal with gmm for surface reflectance component. in Proceedings of 18th IEEE International Conference of Pattern Recognition, 2006.
[41] KaewTraKulPong P, Bowden R. An improved adaptive backgroundmixture model for real-time tracking with shadow detection [ C ]. Proceedings of the 2nd European Workshop on Advanced VideoBased Surveillance Systems. USA: Kluwer Academic Publishers, 2001: 135-144.
[42] Razali M.T., Adznan B.J. Detection and Classification of Moving Object for Smart Vision Sensor. Information and Communication Technologies, 2006. On page(s): 733 -737.
[43]冯祖仁,吕娜,李良福等.基于最大后验概率的图像匹配相似性指标研究[J].自动化学报. 2007.33卷01期: 1-8.
[44] I. W. Faig. Calibration of Close Range Photogrammetric System: Mathematical Formulation [J]. Photogrammeric Eng. Remote Sensing, VOL. 41, PP. 1479-1486, 1975.
[45] R. Y. Tsai. An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision[C],Proc. of IEEE Conference of Computer Vision and Pattern Recognition, 1986:364-374,
[46] Arun Hampapur, Lisa Brown.et.The IBM Smart Surveillance System.
[47] Ying-li Tian, Lisa Brown. IBM smart surveillance system (S3): event based video surveillance system with an open and extensible framework.. Machine Vision and Applications. 2008:315-327.
[48] Yingli Tian. the IBM smart surveillance system release 1. Wireless and Optical Communications. 2005:98-.
[49]边肇棋,张学工.模式识别第二版.北京:清华大学出版社,2000, 284-299.
[50] John C Platt, Nello Cristianini, John Shawe-taylor. Large Margin DAGs for Multiclass Classification. Advances in Neural Information Processing Systems. 2005:547-553.
[51]吴成东等.基于误差修正码的支持向量机大类别分类方法[J] .沈阳建筑工程学院学报(自然科学版) ,2004 ,20 (1) :66-70.
[52] Thomas G. Dietterich, Ghulum Bakiri. Solving Multiclass Learning Problems via Error- Correcting Output Codes. 1995:263-286.