"

Cookies ussage consent

I agree.

Our site saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing our site without changing the browser settings you grant us permission to store that information on your device.

Computer vision measurement new algorithm and uncertainty evaluation

J. YANG1,* , N. G. LU2

Affiliation

  1. Department of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100081, China
  2. Electronic and Information Engineering, Beijing Information Science and Technology University, Beijing 100192, China

Abstract

Usually, engineers use the least square method to solve the equation group in machine vision measurement task. By algebra principle, the least square method is inefficient for linear correlated variable. In this case, big error exists in the measuring result. In some machine vision task, people only want to know the depth information. In order to solve those problems, this paper proposed new method based on the parametric equation originally. Use parametric equation to define the machine vision system model and define scene points as quaternion. We can calculate the depth information by decomposing the equation group. And evaluation of variance is easier by this equation. Meanwhile, in order to illustrate the variance this paper use Hough transform to the equations. The line in the original coordinate system will change to a point in another coordinate system. Many points fit a line to denote a point in the original coordinate system. The experiment in the end proves the algorithm effective.

Keywords

Machine vision measurement, Three-dimensional reconstruction, Error analysis, Uncertainty.

Citation

J. YANG, N. G. LU, Computer vision measurement new algorithm and uncertainty evaluation, Optoelectronics and Advanced Materials - Rapid Communications, 2, 12, December 2008, pp.758-762 (2008).

Submitted at: Nov. 5, 2008

Accepted at: Dec. 4, 2008