Subpixel edge localization with statistical error compensation

Subpixel Edge Localization (EL) techniques are often affected by an error that exhibits a systematic character. When this happens, their performance can be improved through compensation of the systematic portion of the localization error. In this paper we propose and analyze a method for estimating the EL characteristic of sub-pixel EL techniques through statistical analysis of appropriate test images. The impact of the compensation method on the accuracy of a camera calibration procedure has been proven to be quite significant (44%), which can be crucial especially in applications of low-cost photogrammetry and 3D reconstruction from multiple views.