首页> 外国专利> WALKING CYCLE MEASURING DEVICE, PERIODIC IMAGE OBTAINING DEVICE, COMPRESSION PROCESSOR FOR MOVING OBJECT OUTLINE, AND MOVING OBJECT DISCRIMINATION SYSTEM

WALKING CYCLE MEASURING DEVICE, PERIODIC IMAGE OBTAINING DEVICE, COMPRESSION PROCESSOR FOR MOVING OBJECT OUTLINE, AND MOVING OBJECT DISCRIMINATION SYSTEM

机译:行走周期测量设备,周期性图像获取设备,移动物体轮廓的压缩处理器以及移动物体判别系统

摘要

PPROBLEM TO BE SOLVED: To improve the accuracy in discriminating a moving object by utilizing the periodic motion of the moving object. PSOLUTION: This system comprises an image pick-up camera 1 which produces the image data of a moving object; a walk cycle extraction part 23 which produces the data of plural images having the moving object by calculating the difference between the data of plural images having no moving object and that of background images, and then obtains periodic images having periodic motion based on the data of the plural images it earlier produced; an outline extraction part 24 which calculates, based on the periodic images, the P expression vector consisting of a real part and an imaginary part of the complex function having, as the exponent of the exponential function, the curvature function that is expressed with the angle formed by the real axis and a vector between neighboring coordinates; a feature vector extraction part 25 which selects only real and imaginary parts of the complex value function expressing the feature parts of the moving object in relation to the P expression vector, and then compresses the data volume of the P expression vector; and an HMM model producing part 26 which identifies the moving object by applying hidden Markov model to the data-compressed P expression vector. PCOPYRIGHT: (C)2005,JPO&NCIPI
机译:

要解决的问题:通过利用运动物体的周期性运动来提高识别运动物体的准确性。

解决方案:该系统包括一个摄像头1,该摄像头产生运动物体的图像数据。步行周期提取部分23,其通过计算不具有运动对象的多个图像的数据与背景图像的数据之间的差来产生具有运动对象的多个图像的数据,然后基于所述数据来获得具有周期性运动的周期性图像。它先前产生的复数图像;轮廓提取部分24,其基于周期图像计算由复函数的实部和虚部组成的P表达式矢量,该P表达式矢量具有以角度表示的曲率函数作为指数函数的指数由实轴和相邻坐标之间的向量组成;特征向量提取部分25,仅选择表示与P表达向量有关的运动对象的特征部分的复数值函数的实部和虚部,然后压缩P表达向量的数据量; HMM模型产生部分26通过将隐马尔可夫模型应用于数据压缩的P表达矢量来识别运动物体。

版权:(C)2005,JPO&NCIPI

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号