搜索

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

复杂背景下目标检测的级联分类器算法研究

高文 汤洋 朱明

引用本文:
Citation:

复杂背景下目标检测的级联分类器算法研究

高文, 汤洋, 朱明

Study on the cascade classifier in target detection under complex background

Gao Wen, Tang Yang, Zhu Ming
PDF
导出引用
  • 目标检测与跟踪一直是图像处理与计算机视觉领域的热门研究方向之一,其对军事上的成像制导、跟踪军事目标等以及民事方面的安防监控、智能人机交互等方面均有着重要的研究价值. 将特征匹配问题看成是一种更普遍的二分类问题,将这种难解的高维计算变成二分类问题,使计算复杂度大大减小,这类方法以大数定律和贝叶斯法则为理论依据,本文提出一种非树形结构的分类器,并从理论上推导出其实现公式,将1bitBP特征应用到分类器中,同时采用计算量由小到大的三个分类器进行级联从而实现鲁棒精确的目标检测. 从实验结果来看,本文算法能够对目标的尺度变化、旋转、部分遮挡、形变、模糊、背景变化等复杂情况有较好鲁棒性,并且检测精度相对较高,而本文算法的计算复杂度低、计算量小,有较高的应用价值.
    Method of target detection and tracking is one of the hot topics in image processing and computer vision field, which is significant not only in military such as imaging guidance and military target tracking, but also for civil use such as security and monitoring and the intelligent man-machine interaction. Treating the feature matching problem as a more general equinoctial classification question, can turn the intractable high-dimensional problem to a classification problem and deplete computer complexity. This method is based on the law of large numbers and Bayes rule. In this paper we propose a non-hierarchy structure classifier, for which the equation for calculation is theoretically derived, and apply 1bitBP feature to the classifier; and for further reducing the amount of calculation, we use integral image and square integral image to variance classifier as preprocessor, and then use non-hierarchy classifier to handle the patches which meet the variance demand and use the nearest neighbor to further improve the accuracy, and finally realize target detection and tracking based on cascade classifier. Our experimental results show that the method proposed is far superior in calculation amount and processing precision, and is robust to scale changing and rotation, so the method proposed in this paper is of high practical value.
    • 基金项目: 中国科学院航空光学成像与测量重点实验室开放基金(批准号:Y2HC1SR121)资助的课题.
    • Funds: Project supported by the Key Laboratory of Airborne Optical Imaging and Measurement, Chinese Academy of Sciences (Grant No. Y2HC1SR121).
    [1]

    Li T W, Shi A G, He S H 2009 Acta Phys. Sin. 58 794 (in Chinese) [李天伟, 石爱国, 何四华 2005 物理学报 58 794]

    [2]

    Guo G R, Wang H Q, Jiang B 2006 Acta Phys. Sin. 55 3985 (in Chinese) [郭桂蓉, 王宏强, 姜斌 2006 55 3985]

    [3]

    Wang M W, Zhai H C, Gao L J 2009 Acta Phys. Sin. 58 1662 (in Chinese) [王明伟, 翟宏琛, 高丽娟 2009 物理学报 58 1662]

    [4]

    Zhang J S, Zhang Z T 2010 Chin. Phys. B 19 104601

    [5]

    Sun J F, Wang Q, Wang L 2010 Chi. Phys. B 19 104203

    [6]

    CHEN G Y, GUO Z X, ZHANG C P 2003 Chinese Phys. Lett. 20 2161

    [7]

    YanJ, Wu M Y 2012 Opt. Precision Eng. 20 439 (in Chinese) [颜佳, 吴敏渊 2012 光学精密工程 20 439]

    [8]

    Zhu Q P, Yan J, Zhang H 2013 Opt. Precision Eng. 21 437 (in Chinese) [朱秋平, 颜佳, 张虎 2013 光学精密工程 21 437]

    [9]

    Gong J L, He X, Wei Z H 2012 Opt. Precision Eng. 20 413 (in Chinese) [龚俊亮, 何昕, 魏仲慧 2012 光学精密工程 20 413]

    [10]

    Shi J, Tomasi C 1994 Conf. on Computer Vision and Pattern Recognition, 1994

    [11]

    Lowe D G 2004 International Journal of Computer Vision 60 91

    [12]

    Ross D, Lim J, Lin R S, Yang M H 2008 Int J Comput Vis. 77 125

    [13]

    Ma Y, Lv Q B, Liu Y Y, Qian L L, Pei L L 2013 Acta Phys. Sin. 62 204202 (in Chinese) [马原, 吕群波, 刘扬阳, 钱路路, 裴琳琳 2005 物理学报 62 204202]

    [14]

    Duarte M F, Baraniuk R G 2012 IEEE Trans. Image Proc. 21 494

    [15]

    Shi W J, Li J 2012 Opt. Precision Eng. 20 2095 (in Chinese) [石文轩, 李婕 2012 光学精密工程 20 2095]

    [16]

    Chen T, Li Z W, Wang J L, Wang B, Guo S 2012 Opt. Precision Eng. 20 2523 (in Chinese) [陈涛, 李正炜, 王建立, 王斌, 郭爽 2012 光学精密工程 20 2523]

    [17]

    Lepetit V, Fua P 2006 IEEE Trans. Pattern Analysis and Machine Intelligence, 28 1465

    [18]

    Grabner H, Leistner C, Bischof H 2008 European Conf. on Computer Vision

    [19]

    Babenko B, Yang M-H, Belongie S 2009 IEEE Conf. on Computer Vision and Pattern Recognition, Washington, DC 2009 p983

    [20]

    Viola P, Platt J, Zhang C 2005 Neural Information Processing Systems, 2005

    [21]

    Saffari A, Leistner C, Santner J 2009 IEEE 12th International Conf. on Computer Vision Workshops Washington DC 2009 p1393

    [22]

    Leistner C, Saffari A, Bischof H 2010 20th International Conf. on Pattern Recognition, Washington DC 2010 p3545

    [23]

    Viola P, Jones M, Snow D 2005 I nternational Journal of Computer Vision 63 153

    [24]

    Tomasz T, Vincent L 2012 European Conf. on Computer Vision (ECCV) 2012

    [25]

    Breiman L 2001 Machine Learning 45 5

    [26]

    Kalal Z, Matas J, Mikolajczyk K 2010 Conf. on Computer Vision and Pattern Recognition, 2010

    [27]

    Viola P, Jones M 2001 Conf. on Computer Vision and Pattern Recognition, 2001 p511

    [28]

    Zheng F, Webb G 2005 the Fourth Australasian DataMining Conference (AusDM05), Sydney, 2005 p141

    [29]

    Hoiem D, Sukthankar R, Schneiderman H, Huston L 2004 Conf. on Computer Vision and Pattern Recognition, 2004, 02 p490

    [30]

    Amit Y, Geman D 1997 Neural Computation, 9 7

    [31]

    Bay H, Ess A, Tuytelaars T 2008 Computer Vision and Image Understanding 10 346

    [32]

    Stalder S, Grabner H, Gool L V 2009 IEEE 12th International Conf. on Computer Vision Workshops (ICCV) 2009

    [33]

    Yu Q, Dinh T B, Medioni G 2008 European Conf. on Computer Vision. 2008

  • [1]

    Li T W, Shi A G, He S H 2009 Acta Phys. Sin. 58 794 (in Chinese) [李天伟, 石爱国, 何四华 2005 物理学报 58 794]

    [2]

    Guo G R, Wang H Q, Jiang B 2006 Acta Phys. Sin. 55 3985 (in Chinese) [郭桂蓉, 王宏强, 姜斌 2006 55 3985]

    [3]

    Wang M W, Zhai H C, Gao L J 2009 Acta Phys. Sin. 58 1662 (in Chinese) [王明伟, 翟宏琛, 高丽娟 2009 物理学报 58 1662]

    [4]

    Zhang J S, Zhang Z T 2010 Chin. Phys. B 19 104601

    [5]

    Sun J F, Wang Q, Wang L 2010 Chi. Phys. B 19 104203

    [6]

    CHEN G Y, GUO Z X, ZHANG C P 2003 Chinese Phys. Lett. 20 2161

    [7]

    YanJ, Wu M Y 2012 Opt. Precision Eng. 20 439 (in Chinese) [颜佳, 吴敏渊 2012 光学精密工程 20 439]

    [8]

    Zhu Q P, Yan J, Zhang H 2013 Opt. Precision Eng. 21 437 (in Chinese) [朱秋平, 颜佳, 张虎 2013 光学精密工程 21 437]

    [9]

    Gong J L, He X, Wei Z H 2012 Opt. Precision Eng. 20 413 (in Chinese) [龚俊亮, 何昕, 魏仲慧 2012 光学精密工程 20 413]

    [10]

    Shi J, Tomasi C 1994 Conf. on Computer Vision and Pattern Recognition, 1994

    [11]

    Lowe D G 2004 International Journal of Computer Vision 60 91

    [12]

    Ross D, Lim J, Lin R S, Yang M H 2008 Int J Comput Vis. 77 125

    [13]

    Ma Y, Lv Q B, Liu Y Y, Qian L L, Pei L L 2013 Acta Phys. Sin. 62 204202 (in Chinese) [马原, 吕群波, 刘扬阳, 钱路路, 裴琳琳 2005 物理学报 62 204202]

    [14]

    Duarte M F, Baraniuk R G 2012 IEEE Trans. Image Proc. 21 494

    [15]

    Shi W J, Li J 2012 Opt. Precision Eng. 20 2095 (in Chinese) [石文轩, 李婕 2012 光学精密工程 20 2095]

    [16]

    Chen T, Li Z W, Wang J L, Wang B, Guo S 2012 Opt. Precision Eng. 20 2523 (in Chinese) [陈涛, 李正炜, 王建立, 王斌, 郭爽 2012 光学精密工程 20 2523]

    [17]

    Lepetit V, Fua P 2006 IEEE Trans. Pattern Analysis and Machine Intelligence, 28 1465

    [18]

    Grabner H, Leistner C, Bischof H 2008 European Conf. on Computer Vision

    [19]

    Babenko B, Yang M-H, Belongie S 2009 IEEE Conf. on Computer Vision and Pattern Recognition, Washington, DC 2009 p983

    [20]

    Viola P, Platt J, Zhang C 2005 Neural Information Processing Systems, 2005

    [21]

    Saffari A, Leistner C, Santner J 2009 IEEE 12th International Conf. on Computer Vision Workshops Washington DC 2009 p1393

    [22]

    Leistner C, Saffari A, Bischof H 2010 20th International Conf. on Pattern Recognition, Washington DC 2010 p3545

    [23]

    Viola P, Jones M, Snow D 2005 I nternational Journal of Computer Vision 63 153

    [24]

    Tomasz T, Vincent L 2012 European Conf. on Computer Vision (ECCV) 2012

    [25]

    Breiman L 2001 Machine Learning 45 5

    [26]

    Kalal Z, Matas J, Mikolajczyk K 2010 Conf. on Computer Vision and Pattern Recognition, 2010

    [27]

    Viola P, Jones M 2001 Conf. on Computer Vision and Pattern Recognition, 2001 p511

    [28]

    Zheng F, Webb G 2005 the Fourth Australasian DataMining Conference (AusDM05), Sydney, 2005 p141

    [29]

    Hoiem D, Sukthankar R, Schneiderman H, Huston L 2004 Conf. on Computer Vision and Pattern Recognition, 2004, 02 p490

    [30]

    Amit Y, Geman D 1997 Neural Computation, 9 7

    [31]

    Bay H, Ess A, Tuytelaars T 2008 Computer Vision and Image Understanding 10 346

    [32]

    Stalder S, Grabner H, Gool L V 2009 IEEE 12th International Conf. on Computer Vision Workshops (ICCV) 2009

    [33]

    Yu Q, Dinh T B, Medioni G 2008 European Conf. on Computer Vision. 2008

  • [1] 张海燕, 徐心语, 马雪芬, 朱琦, 彭丽. 超声图像中复合材料褶皱形态的Mask-RCNN识别方法. 物理学报, 2022, 71(7): 074302. doi: 10.7498/aps.71.20212009
    [2] 徐艳, 王培光, 杨青, 董江涛. 时空相关多通道聚类的运动目标检测. 物理学报, 2019, 68(16): 164203. doi: 10.7498/aps.68.20190161
    [3] 张文杰, 王世元, 冯亚丽, 冯久超. 基于Huber的高阶容积卡尔曼跟踪算法. 物理学报, 2016, 65(8): 088401. doi: 10.7498/aps.65.088401
    [4] 陈典兵, 朱明, 高文, 王慧利, 杨航. 基于残差矩阵估计的稀疏表示目标跟踪算法. 物理学报, 2016, 65(19): 194201. doi: 10.7498/aps.65.194201
    [5] 吴昊, 陈树新, 杨宾峰, 陈坤. 基于广义M估计的鲁棒容积卡尔曼滤波目标跟踪算法. 物理学报, 2015, 64(21): 218401. doi: 10.7498/aps.64.218401
    [6] 逯志宇, 王大鸣, 王建辉, 王跃. 基于时频差的正交容积卡尔曼滤波跟踪算法. 物理学报, 2015, 64(15): 150502. doi: 10.7498/aps.64.150502
    [7] 高文, 汤洋, 朱明. 目标跟踪中目标模型更新问题的半监督学习算法研究. 物理学报, 2015, 64(1): 014205. doi: 10.7498/aps.64.014205
    [8] 杨立学, 陈克安, 张冰瑞, 梁雍. 基于不相似度评价的水下声目标分类与听觉特征辨识. 物理学报, 2014, 63(13): 134304. doi: 10.7498/aps.63.134304
    [9] 金左轮, 韩静, 张毅, 柏连发. 基于纹理显著性的微光图像目标检测. 物理学报, 2014, 63(6): 069501. doi: 10.7498/aps.63.069501
    [10] 吴瑕, 陈建文, 鲍拯, 郭德阳. 火箭尾焰对高频回波的影响及其频域能量凝聚检测方法. 物理学报, 2014, 63(11): 119401. doi: 10.7498/aps.63.119401
    [11] 王保宪, 赵保军, 唐林波, 王水根, 吴京辉. 基于双向稀疏表示的鲁棒目标跟踪算法. 物理学报, 2014, 63(23): 234201. doi: 10.7498/aps.63.234201
    [12] 侯旺, 于起峰, 雷志辉, 刘晓春. 基于分块速度域改进迭代运动目标检测算法的红外弱小目标检测. 物理学报, 2014, 63(7): 074208. doi: 10.7498/aps.63.074208
    [13] 行鸿彦, 龚平, 徐伟. 海杂波背景下小目标检测的分形方法. 物理学报, 2012, 61(16): 160504. doi: 10.7498/aps.61.160504
    [14] 许元男, 赵远, 刘丽萍, 张宇, 孙秀冬. 基于伪Wigner-Ville分布和Rényi熵的显著图目标检测. 物理学报, 2010, 59(2): 980-988. doi: 10.7498/aps.59.980
    [15] 余赟, 惠俊英, 陈阳, 孙国仓, 滕超. 浅海低频声场中目标深度分类方法研究. 物理学报, 2009, 58(9): 6335-6343. doi: 10.7498/aps.58.6335
    [16] 何四华, 杨绍清, 石爱国, 李天伟. 基于图像区域Lyapunov指数的海面舰船目标检测. 物理学报, 2009, 58(2): 794-801. doi: 10.7498/aps.58.794
    [17] 姜 斌, 王宏强, 黎 湘, 郭桂蓉. 海杂波背景下的目标检测新方法. 物理学报, 2006, 55(8): 3985-3991. doi: 10.7498/aps.55.3985
    [18] 郏东耀, 丁天怀. 皮棉杂质透射检测及成像目标增强方法. 物理学报, 2005, 54(9): 4058-4064. doi: 10.7498/aps.54.4058
    [19] 马兆勉, 陶纯堪. 区域分形与人工目标检测. 物理学报, 1999, 48(12): 2202-2207. doi: 10.7498/aps.48.2202
    [20] 申金媛, 刘, 常胜江, 贾 佳, 张文伟, 张延, 母国光. 基于径向基函数的多目标旋转不变分类及其光电实现. 物理学报, 1998, 47(12): 1968-1975. doi: 10.7498/aps.47.1968
计量
  • 文章访问数:  4244
  • PDF下载量:  974
  • 被引次数: 0
出版历程
  • 收稿日期:  2013-12-03
  • 修回日期:  2013-12-23
  • 刊出日期:  2014-05-05

复杂背景下目标检测的级联分类器算法研究

  • 1. 中国科学院长春光学精密机械与物理研究所, 长春 130033;
  • 2. 中国科学院航空光学成像与测量重点实验室, 长春 130033
    基金项目: 中国科学院航空光学成像与测量重点实验室开放基金(批准号:Y2HC1SR121)资助的课题.

摘要: 目标检测与跟踪一直是图像处理与计算机视觉领域的热门研究方向之一,其对军事上的成像制导、跟踪军事目标等以及民事方面的安防监控、智能人机交互等方面均有着重要的研究价值. 将特征匹配问题看成是一种更普遍的二分类问题,将这种难解的高维计算变成二分类问题,使计算复杂度大大减小,这类方法以大数定律和贝叶斯法则为理论依据,本文提出一种非树形结构的分类器,并从理论上推导出其实现公式,将1bitBP特征应用到分类器中,同时采用计算量由小到大的三个分类器进行级联从而实现鲁棒精确的目标检测. 从实验结果来看,本文算法能够对目标的尺度变化、旋转、部分遮挡、形变、模糊、背景变化等复杂情况有较好鲁棒性,并且检测精度相对较高,而本文算法的计算复杂度低、计算量小,有较高的应用价值.

English Abstract

参考文献 (33)

目录

    /

    返回文章
    返回