基于移动端的“非受控”物体识别算法的实现
|
摘要
针对现有的物体识别方法在复杂环境下易受光照、角度、尺寸、复杂背景等"非受控"因素的影响,且识别率低、实时性差、占用内存大等问题,提出一种新的物体识别算法,并在此基础上实现了基于移动端的物体识别系统。该方法首先利用粒子滤波算法对检测范围进行加窗跟踪,接着用分水岭分割算法对物体进行分割,然后用HOG(Histogram of Oriented Gradient)算法提取物体特征,最后运用随机森林算法进行物体匹配。实验结果表明该方法能基于移动端在"非受控"的环境下进行较快速且准确的识别,从而证明了该方法的有效性。
Aiming at the problems that the existing object recognition methods are easy to be influenced by "uncontrolled" factors such as illumination,angle,size and complex environment,and have the problems such as low recognition rate,poor real-time performance and large memory consumption,this paper proposed a new object recognition algorithm,on which the object recognition system based on mobile terminal was realized.This method first employs particle filter algorithm to track the detection range by adding windows,and then applies the watershed segmentation algorithm to segment objects,then uses the HOG(Histogram of Oriented Gradient) algorithm to extract object features.Finally,the random forest algorithm is utilized to recognize objects.The experimental results show that this method can be used to identify the mobile terminal quickly and accurately in an "uncontrolled" environment.
Aiming at the problems that the existing object recognition methods are easy to be influenced by "uncontrolled" factors such as illumination,angle,size and complex environment,and have the problems such as low recognition rate,poor real-time performance and large memory consumption,this paper proposed a new object recognition algorithm,on which the object recognition system based on mobile terminal was realized.This method first employs particle filter algorithm to track the detection range by adding windows,and then applies the watershed segmentation algorithm to segment objects,then uses the HOG(Histogram of Oriented Gradient) algorithm to extract object features.Finally,the random forest algorithm is utilized to recognize objects.The experimental results show that this method can be used to identify the mobile terminal quickly and accurately in an "uncontrolled" environment.
引文
[1] SHAH S A A,BENNAMOUN M,BOUSSAID F.Automatic object detection using objectness measure[C]//International Conference on Communications,Signal Processing,and Their Applications.IEEE,2013:1-6.
[2] 徐晓.计算机视觉中物体识别综述[J].电脑与信息技术,2013,21(5):4-6.
[3] 黄凯奇,任伟强,谭铁牛.图像物体分类与检测算法综述[J].计算机学报,2014,36(6):1225-1240.
[4] 林志强,陈小平.一种结合多特征的实时物体识别系统[J].小型微型计算机系统,2015,36(6):1310-1315.
[5] 刘曦,史忠植,石志伟,等.一种基于特征捆绑计算模型的物体识别方法[J].软件学报,2010,21(3):452-460.
[6] 卢良锋,谢志军,叶宏武.基于RGB特征与深度特征融合的物体识别算法[J].计算机工程,2016,42(5):186-193.
[7] 孙利娟,张继栋,杨新锋.基于多稀疏分布特征和最近邻分类的物体识别方法[J].计算机应用研究,2016,33(10):3156-3159.
[8] 尚俊.基于HOG特征的目标识别算法研究[D].武汉:华中科技大学,2012.
[9] 唐发明,王仲东,陈绵云.一种新的二叉树多类支持向量机算法[J].计算机工程与应用,2005,41(7):24-26.
[10] 苏亚麟,吕开云.基于随机森林算法的特征选择的水稻分类——以南昌市为例[J]江西科学,2018(1):161-167.
[11] 周雪晴,张占松,张超谟,等.基于粗糙集—随机森林算法的复杂岩性识别[J].大庆石油地质与开发,2017,36(6):127-133.
[12] 闫月影.非受控场景下的二维人脸识别研究[J].数码世界,2017(11):394-395.
[13] 李安平.复杂环境下的视频目标跟踪算法研究[D].上海:上海交通大学,2007.
[14] 陈代武.基于移动终端的多角度实物识别方法[D].北京:北京邮电大学,2015.
[15] FRIEDMAN N,RUSSELL S.Image Segmentation in Video Sequences:A Probabilistic Approach,Uncertainty in Artificial Intelligence[J].arXiv:1302.1539,1997.
[16] HALEVY G,WEINSHALL D.Motion of Disturbances:Detection and Tracking of Multi-Body Non-Rigid Motion[J].Machine Vision and Applications,1999,11(3):122-137.
[17] KUMAR S,DAI Y,LI H.Spatio-Temporal Union of subspaces for Multi-body Non-rigid Structure-from-Motion[J].Pattern Recognition,2017,71:428-443.
[18] 曹正凤.随机森林算法优化研究[D].北京:首都经济贸易大学,2014.
[19] 许保勋.面向高维数据的随机森林算法优化研究[D].哈尔滨:哈尔滨工业大学,2013.
[20] 程广涛,陈雪,郭照庄.基于HOG特征的行为人视觉检测方法[J].传感器与微系统,2011,30(7):68-70.
[21] 赵桂儒.较大规模数据应用PCA降维的一种方法[J].电脑知识与技术,2014(8):1835-1837.
[22] 杨彪,倪蓉蓉,江大鹏.一种对光照变化鲁棒的移动目标前景提取方法[J].计算机科学,2016,43(s2):186-189.
[23] 郭宇,郝晓燕,张兴忠.基于预测的多特征融合Mean-Shift跟踪算法[J].计算机科学,2018,45(s1):171-173.
[2] 徐晓.计算机视觉中物体识别综述[J].电脑与信息技术,2013,21(5):4-6.
[3] 黄凯奇,任伟强,谭铁牛.图像物体分类与检测算法综述[J].计算机学报,2014,36(6):1225-1240.
[4] 林志强,陈小平.一种结合多特征的实时物体识别系统[J].小型微型计算机系统,2015,36(6):1310-1315.
[5] 刘曦,史忠植,石志伟,等.一种基于特征捆绑计算模型的物体识别方法[J].软件学报,2010,21(3):452-460.
[6] 卢良锋,谢志军,叶宏武.基于RGB特征与深度特征融合的物体识别算法[J].计算机工程,2016,42(5):186-193.
[7] 孙利娟,张继栋,杨新锋.基于多稀疏分布特征和最近邻分类的物体识别方法[J].计算机应用研究,2016,33(10):3156-3159.
[8] 尚俊.基于HOG特征的目标识别算法研究[D].武汉:华中科技大学,2012.
[9] 唐发明,王仲东,陈绵云.一种新的二叉树多类支持向量机算法[J].计算机工程与应用,2005,41(7):24-26.
[10] 苏亚麟,吕开云.基于随机森林算法的特征选择的水稻分类——以南昌市为例[J]江西科学,2018(1):161-167.
[11] 周雪晴,张占松,张超谟,等.基于粗糙集—随机森林算法的复杂岩性识别[J].大庆石油地质与开发,2017,36(6):127-133.
[12] 闫月影.非受控场景下的二维人脸识别研究[J].数码世界,2017(11):394-395.
[13] 李安平.复杂环境下的视频目标跟踪算法研究[D].上海:上海交通大学,2007.
[14] 陈代武.基于移动终端的多角度实物识别方法[D].北京:北京邮电大学,2015.
[15] FRIEDMAN N,RUSSELL S.Image Segmentation in Video Sequences:A Probabilistic Approach,Uncertainty in Artificial Intelligence[J].arXiv:1302.1539,1997.
[16] HALEVY G,WEINSHALL D.Motion of Disturbances:Detection and Tracking of Multi-Body Non-Rigid Motion[J].Machine Vision and Applications,1999,11(3):122-137.
[17] KUMAR S,DAI Y,LI H.Spatio-Temporal Union of subspaces for Multi-body Non-rigid Structure-from-Motion[J].Pattern Recognition,2017,71:428-443.
[18] 曹正凤.随机森林算法优化研究[D].北京:首都经济贸易大学,2014.
[19] 许保勋.面向高维数据的随机森林算法优化研究[D].哈尔滨:哈尔滨工业大学,2013.
[20] 程广涛,陈雪,郭照庄.基于HOG特征的行为人视觉检测方法[J].传感器与微系统,2011,30(7):68-70.
[21] 赵桂儒.较大规模数据应用PCA降维的一种方法[J].电脑知识与技术,2014(8):1835-1837.
[22] 杨彪,倪蓉蓉,江大鹏.一种对光照变化鲁棒的移动目标前景提取方法[J].计算机科学,2016,43(s2):186-189.
[23] 郭宇,郝晓燕,张兴忠.基于预测的多特征融合Mean-Shift跟踪算法[J].计算机科学,2018,45(s1):171-173.