| 运动物体速度场估计的新方法 |
| |
| 引用本文： | 陈丽亚,戚飞虎.运动物体速度场估计的新方法[J].上海交通大学学报,1994,28(1):55-61. |
| |
| 作者姓名： | 陈丽亚 戚飞虎 |
| |
| 摘 要： | 本文提出了一种具有一定应用价值的运动物体速度场的计算方法。本算法避开了现有的一些算法所遇到的图像分割问题，用一种类Ｈｏｐｆｉｅｌｄ神经网模型实现局部图像的Ｇａｂｏｒ分解，得到分解系数，对用Ｇａｂｏｒ基函数和式形式表示的图像进行傅立叶变换，求出其频谱不为零的点，并由这些点约束的平面方程求出速度。神经元网络的应用，缩短了优化计算所需时间而不降低计算精度。本算法在ＡＳＴ－３８６微机上进行了实现。
|
| 关 键 词： | 速度场 Gabor分解 运动物体 估计 |
A New Method for Estimating Velocity Fieldof Moving Objects |
| |
| Chen Liya,Qi Feihu..A New Method for Estimating Velocity Fieldof Moving Objects[J].Journal of Shanghai Jiaotong University,1994,28(1):55-61. |
| |
| Authors: | Chen Liya Qi Feihu |
| |
| Institution: | Chen Liya;Qi Feihu. |
| |
| Abstract: | Tail papet presents a new algorithm to estimate the velocity field of moving objects using Gabor transform,which is implemented by a neural network.It is independent of the form of a moving object and its contour. This method costs less of computation with no descent of precision. The algorithm is implemented in micro-computer Ast-386 in C language.All results are satisfactory under ideal, incomplete and general conditions. |
| |
| Keywords: | velocity field Gabor transform Spectrum co-planarity theorem |
| 本文献已被 CNKI 维普 等数据库收录！ |
