篮球运动快速位移图像自适应定位方法
|
摘要
针对传统方法在进行篮球位移图像定位时存在计算复杂度高、抗干扰性差、准确性低的问题,提出一种新的篮球运动快速位移图像自适应定位方法.对图像互能量谱进行反转,对得到的背景运动偏移量进行补偿;依据灰度阈值技术,对背景模板和当前帧篮球运动图像进行差分,获取当前帧快速位移区域;将种子像素作为生长起点,将邻域中与种子像素相似的像素聚集起来,得到快速位移目标的最小框,完成定位.结果表明,所提方法得到的定位结果清晰准确,抗干性强扰;在目标出现变形的情况下,本文方法受变形影响小,抗噪能力高,定位精度高.
Aiming at the problem of high computational complexity,poor anti-interference and lowaccuracy in the traditional methods for the basketball displacement image location,a newadaptive location method for the fast displacement images in the basketball sport was proposed. The image mutual energy spectrum was inverted,and the obtained background motion offset was compensated. According to the gray threshold technique,the background template and the current frame basketball sport images were differentiated to obtain the fast displacement region of current frames. The seed pixel was used as a starting growth point,and the pixels similar to the seed pixel in the neighborhood area were collected to obtain the minimum frame of fast displacement target to complete the location. The results showthat the location results obtained by the proposed method are clear and accurate,and are resistant to the strong interference. In the case of target deformation,the proposed method is less affected by the deformation,and has high noise resistance and high location accuracy.
Aiming at the problem of high computational complexity,poor anti-interference and lowaccuracy in the traditional methods for the basketball displacement image location,a newadaptive location method for the fast displacement images in the basketball sport was proposed. The image mutual energy spectrum was inverted,and the obtained background motion offset was compensated. According to the gray threshold technique,the background template and the current frame basketball sport images were differentiated to obtain the fast displacement region of current frames. The seed pixel was used as a starting growth point,and the pixels similar to the seed pixel in the neighborhood area were collected to obtain the minimum frame of fast displacement target to complete the location. The results showthat the location results obtained by the proposed method are clear and accurate,and are resistant to the strong interference. In the case of target deformation,the proposed method is less affected by the deformation,and has high noise resistance and high location accuracy.
引文
[1]李少军,肖利平,高磊.光电成像制导中地面运动目标检测与跟踪[J].系统工程与电子技术,2014,36(1):23-30.(LI Shao-jun,XIAO Li-ping,GAO Lei. Detection andtracking of moving targets in photoelectric imagingguidance[J]. Systems Engineering and Electronics,2014,36(1):23-30.)
[2]蔺俊杰,曹军芳.基于多权值相似均衡的运动图像循环定位算法[J].科技通报,2015,31(8):111-113.(LIN Jun-jie,CAO Jun-fang. Loop tracking algorithmfor moving image based on multi-w eighted similarequilibrium[J]. Bulletin of Science and Technology,2015,31(8):111-113.)
[3]张文娟,王大羽,余梅生.基于两种独立混沌函数的图像加密算法[J].重庆邮电大学学报(自然科学版),2017,29(2):232-239.(ZHANG Wen-juan,WANG Da-yu,YU Mei-sheng.Research of image encryption algorithm based on tw oindependent chaotic functions[J]. Journal of Chong-qing University of Posts and Telecommunications(Natural Science Edition),2017,29(2):232-239.)
[4]冷纪岚.人体运动姿态图像矫正识别仿真研究[J].计算机仿真,2015,32(12):451-454.(LENG Ji-lan. Three dimensional motion pattern re-cognition posture correction[J]. Computer Simula-tion,2015,32(12):451-454.)
[5] Patil N,Biswas P K. Global abnormal events detectionin crow ded scenes using context location and motion-rich spatio-temporal volumes[J]. IET Image Processing,2018,12(4):596-604.
[6]周牧,张巧,田增山,等.基于图像边缘检测的WLAN室内用户运动地图构建[J].电子科技大学学报,2016,45(2):240-245.(ZHOU Mu,ZHANG Qiao,TIAN Zeng-shan,et al.Edge detection based individual mobility graph con-struction in WLAN indoor environment[J]. Journalof University of Electronic Science and Technology ofChina,2016,45(2):240-245.)
[7]卢军,王子梁,朱佳磊.一种基于时-空多尺度的运动目标检测方法[J].西北工业大学学报,2017,35(1):98-102.(LU Jun,WANG Zi-liang,ZHU Jia-lei. Space-timemultiscale based moving object detection method[J].Journal of Northw estern Polytechnical University,2017,35(1):98-102.)
[8]李娜,王维哲,赵伟.高速移动图像丢帧误差的矫正方法研究[J].控制工程,2015,22(6):1196-1200.(LI Na,WANG Wei-zhe,ZHAO Wei. High-speedmobile image frame error correction method[J].Control Engineering of China,2015,22(6):1196-1200.)
[9] Schellekens W,Ramsey N F,Raemaekers M. Predic-tions to motion stimuli in human early visual cortex:effects of motion displacement on motion predictabili-ty[J]. Neuroimage,2015,18(5):118-125.
[10]朱闻亚.结合稳健估计和Meanshift的视频目标跟踪算法[J].沈阳工业大学学报,2017,39(2):177-182.(ZHU Wen-ya. Video target tracking algorithm withcombining robust estimation and M eanshift[J]. Jour-nal of Shenyang University of Technology,2017,39(2):177-182.)
[11]史洪印,张诺.基于稀疏表示和道路辅助的单幅SAR图像运动目标检测方法[J].电子学报,2015,43(3):431-439.(SHI Hong-yin,ZHANG Nuo. Moving targets indica-tion method in single SAR imagery based on sparserepresentation and road information[J]. Acta Elec-tronica Sinica,2015,43(3):431-439.)
[12]金兆飞,雷仲魁,李倩.基于运动目标跟踪与定位的无人机测速技术[J].光电工程,2015,42(2):59-65.(JIN Zhao-fei,LEI Zhong-kui,LI Qian. UAV measuringobject velocity based on moving target tracking andorientation[J]. Opto-Electronic Engineering,2015,42(2):59-65.)
[13]Guignard J,Cheng J T,Ravicz M E,et al. Aligningdigital holography images of tympanic membrane mo-tion[J]. Journal of the Acoustical Society of Ameri-ca,2014,135(4):2416-2416.
[14]刘巍,陈玲,马鑫,等.基于彩色图像的高速目标单目位姿测量方法[J].仪器仪表学报,2016,37(3):675-682.(LIU Wei,CHEN Ling,MA Xin,et al. Monocular po-sition and pose measurement method for high-speedtargets based on colored images[J]. Chinese Journalof Scientific Instrument,2016,37(3):675-682.)
[2]蔺俊杰,曹军芳.基于多权值相似均衡的运动图像循环定位算法[J].科技通报,2015,31(8):111-113.(LIN Jun-jie,CAO Jun-fang. Loop tracking algorithmfor moving image based on multi-w eighted similarequilibrium[J]. Bulletin of Science and Technology,2015,31(8):111-113.)
[3]张文娟,王大羽,余梅生.基于两种独立混沌函数的图像加密算法[J].重庆邮电大学学报(自然科学版),2017,29(2):232-239.(ZHANG Wen-juan,WANG Da-yu,YU Mei-sheng.Research of image encryption algorithm based on tw oindependent chaotic functions[J]. Journal of Chong-qing University of Posts and Telecommunications(Natural Science Edition),2017,29(2):232-239.)
[4]冷纪岚.人体运动姿态图像矫正识别仿真研究[J].计算机仿真,2015,32(12):451-454.(LENG Ji-lan. Three dimensional motion pattern re-cognition posture correction[J]. Computer Simula-tion,2015,32(12):451-454.)
[5] Patil N,Biswas P K. Global abnormal events detectionin crow ded scenes using context location and motion-rich spatio-temporal volumes[J]. IET Image Processing,2018,12(4):596-604.
[6]周牧,张巧,田增山,等.基于图像边缘检测的WLAN室内用户运动地图构建[J].电子科技大学学报,2016,45(2):240-245.(ZHOU Mu,ZHANG Qiao,TIAN Zeng-shan,et al.Edge detection based individual mobility graph con-struction in WLAN indoor environment[J]. Journalof University of Electronic Science and Technology ofChina,2016,45(2):240-245.)
[7]卢军,王子梁,朱佳磊.一种基于时-空多尺度的运动目标检测方法[J].西北工业大学学报,2017,35(1):98-102.(LU Jun,WANG Zi-liang,ZHU Jia-lei. Space-timemultiscale based moving object detection method[J].Journal of Northw estern Polytechnical University,2017,35(1):98-102.)
[8]李娜,王维哲,赵伟.高速移动图像丢帧误差的矫正方法研究[J].控制工程,2015,22(6):1196-1200.(LI Na,WANG Wei-zhe,ZHAO Wei. High-speedmobile image frame error correction method[J].Control Engineering of China,2015,22(6):1196-1200.)
[9] Schellekens W,Ramsey N F,Raemaekers M. Predic-tions to motion stimuli in human early visual cortex:effects of motion displacement on motion predictabili-ty[J]. Neuroimage,2015,18(5):118-125.
[10]朱闻亚.结合稳健估计和Meanshift的视频目标跟踪算法[J].沈阳工业大学学报,2017,39(2):177-182.(ZHU Wen-ya. Video target tracking algorithm withcombining robust estimation and M eanshift[J]. Jour-nal of Shenyang University of Technology,2017,39(2):177-182.)
[11]史洪印,张诺.基于稀疏表示和道路辅助的单幅SAR图像运动目标检测方法[J].电子学报,2015,43(3):431-439.(SHI Hong-yin,ZHANG Nuo. Moving targets indica-tion method in single SAR imagery based on sparserepresentation and road information[J]. Acta Elec-tronica Sinica,2015,43(3):431-439.)
[12]金兆飞,雷仲魁,李倩.基于运动目标跟踪与定位的无人机测速技术[J].光电工程,2015,42(2):59-65.(JIN Zhao-fei,LEI Zhong-kui,LI Qian. UAV measuringobject velocity based on moving target tracking andorientation[J]. Opto-Electronic Engineering,2015,42(2):59-65.)
[13]Guignard J,Cheng J T,Ravicz M E,et al. Aligningdigital holography images of tympanic membrane mo-tion[J]. Journal of the Acoustical Society of Ameri-ca,2014,135(4):2416-2416.
[14]刘巍,陈玲,马鑫,等.基于彩色图像的高速目标单目位姿测量方法[J].仪器仪表学报,2016,37(3):675-682.(LIU Wei,CHEN Ling,MA Xin,et al. Monocular po-sition and pose measurement method for high-speedtargets based on colored images[J]. Chinese Journalof Scientific Instrument,2016,37(3):675-682.)