Segmentation

The segmentation phase can be stated as follows:

1. Extract a feature vector for an image pixel.
2. Feed into SOM and find the best-matching unit c (Eq. 1 and 2). Show the minimum error as the output.
3. If is larger than the specified limit T, the pixel is classified to a defect.
4. Repeat steps 1-3 for each image pixel.

The limit T is determined by hand, and it depends on features and on the desired accuracy of the segmentation. The segmentation scheme is depicted in Figure 1.

  
Figure 1: The proposed segmentation scheme.

The matching error for the feature vector and the map unit i is defined similarly to the Hamming distance as

 

is defined in Eq. 5 and in Eq. 6. The best-matching unit c is now chosen to be the unit with the minimum error ,

 

If there are several map units with the same minimum error , the one with the minimum Euclidean distance with the feature vector is chosen.