Answers are grounded in this paper and cite their sources.
Video analysis is widely used in undergraduate physics laboratories to determine kinematic quantities. Radial lens distortion can compromise the assumed linear relationship between physical positions and image coordinates, affecting the accuracy of the results. In this work, we investigate how radial lens distortion propagates into the kinematic parameters obtained in video analysis experiments used in physics education. We develop an analytical model based on the radial component of the classical Brown-Conrady model to describe the mapping between the physical position of the object and its observed image coordinate. The analysis shows that radial nonlinearity modifies the quadratic coefficient obtained from the temporal fit of the trajectory, thereby introducing a systematic error into the estimated acceleration. To test this prediction, we performed experimental measurements using the Tracker software on videos recorded with a DSLR camera and a smartphone equipped with different optical modules. The radial distortion of each system was characterised by filming a calibrated ruler, and the resulting radial mappings were inverted numerically. Numerical inversion changed the fitted acceleration by 0.257% for the three-coefficient Ultra-Wide model and by 4.33% for the one-term DSLR model. The mean Ultra-Wide acceleration changed from 9.883 to 9.858 m s⁻², reducing its nominal deviation from conventional standard gravity to 0.521%, although the root-mean-square error (RMSE) of the trajectory fit increased. For the DSLR, the mean acceleration changed from 10.28 to 9.853 m s⁻², reducing its nominal deviation from conventional standard gravity to 0.474%, while the mean trajectory RMSE decreased from 1.496 to 0.385 mm. The results show that radial distortion can introduce systematic errors into acceleration estimates obtained from video analysis and that its correction should be assessed both by its effect on the estimated quantity and by the quality of the trajectory fit.
No takes yet. Share an insight, caveat, or question.
Andrade et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: