Direct visual servoing using ZNCC criterion

Amaury Dame (1) , Eric Marchand (1)

1 Lagadic - Visual servoing in robotics, computer vision, and augmented reality

This paper proposes a direct visual scheme. In direct visual servoing approaches, the goal is to consider all the image as a whole. Therefore we do no rely on a feature extraction or tracking process. In nominal condition, these approaches have proved to be robust and very precise. In this paper, we proposed to consider a new metric that is the Zero mean Normalized Cross-Correlation function (ZNCC). This correlation criterion is known to be more robust that the classical SSD to linear brightness variations. This paper gives the expressions required to perform a ZNCC-based visual servoing task. Experimental results demonstrates the efficiency of the proposed method.

Domaines: Robotique [cs.RO]

Amaury Dame, Eric Marchand. Direct visual servoing using ZNCC criterion. 2012. ⟨hal-01683329⟩