In this paper, we investigate the defect detection by making use of pre-made reference image data and classify the defects by using the artificial neural network.
The approach is composed of three main parts. The first step consists of proper generation of two reference image data. For this purpose we expand and shrink the defect free image data to a certain limit respectively by using a low level morphological technique. Then, each reference data is saved for subsequent defect finding process.
The second step proceeds by performing three times logical bit operations between two ready-made reference images and just captured image to be tested, which results in a defects only image.
In the third step, extracting four features from each detected defect, followed by assigning them into the input nodes of already trained artificial neural network gives us corresponding defect kind. The look up table method for classification is also proposed in this paper.
All of the image data are formed in a bit level for the purpose of small data size as well as time saving. Results show that suggested algorithms are available for flexible defect detection, robust classification, and high speed process due to simple logic operation.