The Phase Only Transform for unsupervised surface defect detection

Dror Aiger (1) , Hugues Talbot (1)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge

We present a simple, fast, and effective method for detecting defects on textured surfaces. Our method is unsupervised and contains no learning stage or information on the texture being inspected. The method is based on the Phase Only Transform (PHOT) which correspond to the Discrete Fourier Transform (DFT), normalized by the magnitude. The PHOT removes any regularities, at arbitrary scales, from the image while preserving only irregular patterns considered to represent defects. The localization is obtained by the inverse transform followed by adaptive thresholding using a simple standard statistical method. The main computational requirement is thus to apply the DFT on the input image. The method is also easy to implement in a few lines of code. Despite its simplicity, the methods is shown to be effective and generic as tested on various inputs, requiring only one parameter for sensitivity. We provide theoretical justification based on a simple model and show results on various kinds of patterns. We also discuss some limitations of the approach.

Mots clés: Fourier transform; texture

Domaines: Traitement des images [eess.IV]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Dror Aiger, Hugues Talbot. The Phase Only Transform for unsupervised surface defect detection. C. H. Chen. Emerging Topics in Computer Vision and its Applications, World Scientific, pp.215-232, 2011, 978-981-4340-99-1. ⟨hal-00622514⟩