Application of the Zernike moment in intravenous infusion needle´s image description

Zernike moments are useful tools in pattern recognition and image description due to their orthogonality and rotation invariance property. In this paper, an image description method which could recognize the position of intravenous infusion needles is developed. Firstly, preprocessing and normalizing the character image are introduced; secondly, the image´s rotation invariant features and the rotation phase using Zernike moments are gotten. The result shows that the Zernike moment has better invariant properties, and can describe robustly the rotation needle-image which advances on the higher assembly accuracy.