Abstract
A rotation-invariant algorithm for a self-acting inspection system to detect scratches in
grayscale images of ground metallic material will be discussed.
Scratches are darker curves on the image, which can easily be detected with our eyes.
After discussing some methods of texture synthesis, a supposed Gibbs distribution is
investigated.
Then the main algorithm is presented, which is divided into three main parts.
As a first step the grayscale image is transformed into a binary image using local pixel
information.
As a second step small line segments of different angels are searched for by comparison
with the neighbourhood of each pixel. If this correspondence is high, the new pixel at the
center of this neighborhood is set to black.
In the third and last step the discovered scratches are classified. The resulting image is
partitioned into stripes of the same direction as the line segment used in step 2. If the number
of black pixels in one of these stripes is much larger than the number of black pixels in the
other stripes, then with high probability there must be a scratch in this stripe.
If no scratch is found, the material passes the inspection.
| Original language | German (Austria) |
|---|---|
| Publication status | Published - Oct 2004 |
Fields of science
- 101 Mathematics