A new strategy for improving vision based tracking accuracy based on utilization of camera calibration information

Abstract— Camera calibration is one of the essential
components of a vision based tracking system where the objective
is to extract three dimensional information from a set of two
dimensional frames. The information extracted from the
calibration process is significant for examining the accuracy of the
vision sensor, and thus further for estimating its effectiveness as a
tracking system in real applications. This paper introduces
another use for this information in which the proper location of
the camera can be predicted. Anew mathematical formula based
on utilizing the extracted calibration information was used for
finding the optimum location for the camera, which provides the
best detection accuracy. Moreover, the calibration information
was also used for selecting the proper image Denoising filter. The
results obtained proved the validity of the proposed formula in
finding the desired camera location where the smallest detection
errors can be produced. Also, results showed that the proper
selection of the filter parameters led to a considerable
enhancement in the overall accuracy of the camera, reducing the
overall detection error by 0.2 mm.