摘要
当前利用人工的方式对篮球视频图像进行监测时,存在图像采集效果不理想、所得图像噪声大及图像分类准确性低等问题,无法得到高效的监测结果。采用线阵摄像机、图像采集卡、摄像头、照明灯光结合的结构实现篮球视频图像的采集,其中的镜头型号为尼康Nikkor 28 m mf/2. 8D,固定焦距是28 mm,支持光圈自动调整,可以高效采集篮球视频图像;对所得图像进行线性灰度变换操作;并分别采用均值、中值与同态滤波,实现篮球视频图像的去噪;利用篮球图像运动轨迹骨架裁剪的方式,正确将图像进行连接,接着对篮球图像参数进行计算;利用三角网的构建方式——凸壳插值法,实现图像的特征提取,通过得到的三角网,获得图像结构散点,实现三角剖分,完成监测。实验结果表明,该方法不仅可以增强图像采集效果和分类精度,还能很好地控制图像中的噪声。
At present,when using artificial method to monitor basketball video image,there are some problems such as poor image acquisition effect,large noise and low accuracy of image classification,which can not get high efficiency monitoring results. Basketball video image acquisition is realized by the combination of linear camera,image acquisition card,camera and lighting. The lens model is Nikon Nikkor 28 m mf/2.8 D,with a fixed focal length of 28 mm,which supports automatic adjustment of aperture. The basketball video image can be collected efficiently,and the linear gray level transform operation is carried out,and the mean value,median value and homomorphism filter are used to realize the result. The basketball video image denoising; using the basketball image track skeleton cutting way,the image is connected correctly,then the basketball image parameters are calculated;using the triangular network construction method-convex shell interpolation method,The feature extraction of the image is realized,and the triangulation is realized through the obtained triangulation network and the scattered points of the image structure. The experimental results show that the proposed method can not only enhance the image acquisition effect and classification accuracy,but also control the noise in the image.
引文
1王东京,张宝辉,陈弘原,等.自适应复杂背景干扰的运动目标检测算法[J].红外技术,2017,39(11):1024-1031Wang Dongjing,Zhang Baohui,Chen Hongyuan,et al.Moving-target detection algorithm adapting complex background interference[J].Infrared Technology,2017,39(11):1024-1031
2吴志芳,刘昕.基于图像处理和模式识别的网络图像监控系统[J].计算机测量与控制,2017,25(3):85-87Wu Zhifang,Liu Xin.Network image monitoring system based on image processing and pattern recognition[J].Computer Measurement&Control,2017,25(3):85-87
3丁春晖.关于雾霾天气进行图像监测仿真研究[J].计算机仿真,2016,33(7):301-304Ding Chunhui.The image to fog algorithm based on wavelet noise reduction measure[J].Computer Simulation,2016,33(7):301-304
4兰振平,孙岩国,王智森,等.基于STM32单片机的自适应视频识别[J].大连工业大学学报,2017,36(1):61-66Lan Zhenping,Sun Yanguo,Wang Zhisen,et al.An adaptive video recognition and processing scheme based on STM32 single chip microcomputer[J].Journal of Dalian Polytechnic University,2017,36(1):61-66
5牛世峰,黎莉,任超伟,等.基于CCD摄像技术的公交车跟车距离监测系统研究[J].科学技术与工程,2016,16(11):198-202Niu Shifeng,Li Li,Ren Chaowei,et al.Research on car-following distance detection system of bus based on CCD camera technology[J].Science Technology and Engineering,2016,16(11):198-202
6苗俊.篮球手臂投篮视频图像跟踪提取仿真[J].计算机仿真,2017,3(11):224-227Miao Jun.Basketball arm shot video image tracking extraction simulation[J].Computer Simulation,2017,3(11):224-227
7杨帅,张娄英,朱静.基于LON总线平台的智慧型水质监测管理系统[J].科技通报,2017,33(4):155-158Yang Shuai,Zhang Louying,Zhu Jing.The intelligent water quality monitoringand-management system based on LON fieldbus platform[J].Bulletin of Science and Technology,2017,33(4):155-158
8申红.小波变换域井下视频监控图像改进阈值去噪方法[J].金属矿山,2017,46(7):151-154Shen Hong.Improved threshold denoising method of mine video monitoring image in wavelet transform domain[J].Metal Mine,2017,46(7):151-154
9赵小强,彭红梅,高强.一种鱼体运动特征视频图像检测方法[J].西安邮电大学学报,2017,22(2):38-43Zhao Xiaoqiang,Peng Hongmei,Gao Qiang.A video image detection method for fish motion characteristics[J].Journal of Xi'an University of Posts and Telecommunications,2017,22(2):38-43
10张海娜.不同视角下海量高分辨率视频图像数据挖掘方法[J].科学技术与工程,2017,17(26):257-261Zhang Haina.The data mining method of massive high resolution video images from different perspectives[J].Science Technology and Engineering,2017,17(26):257-261