论文部分内容阅读
如何解决计算的鲁棒性是摄像机参数估计的线性算法中很重要的问题。本文简要介绍了对基本8点算法的各种改进措施,并在对图像数据进行平移和缩放变换的基础上进一步采用了数据分块和多次选择计算技术。仿真结果表明,新的算法极大的减少了运算时间,并在噪声较小的情况下,计算结果比原算法有了很大的提高,表明这种办法是切实可行的。但当噪声过大时,改进方法的计算结果与原始算法相差不大,尽管计算时间上仍有很大的减少。
How to solve the computational robustness is a very important issue in the linear algorithm of camera parameter estimation. This article briefly introduces the various improvements to the basic 8-point algorithm, and further uses the data block and multiple choice computation techniques based on the image data transformation and scaling. The simulation results show that the new algorithm greatly reduces the computational time, and in the case of smaller noise, the computational result is greatly improved than the original one, which shows that this method is feasible. However, when the noise is too large, the calculation results of the improved method are not much different from the original one, though the computation time is still greatly reduced.