Classi cation of Product Inspection Items Using Nonlinear Features

Automated processing and classiication of real-time X-ray images of randomly oriented touching pistachio nuts is discussed. The ultimate objective is the development of a system for automated non-invasive detection of defective product items on a conveyor belt. This approach involves two main steps: preprocessing and classiication. Preprocessing locates individual items and segments ones that touch using a modiied watershed algorithm. The second stage involves extraction of features that allow discrimination between damaged and clean items (pistachio nuts). This feature extraction and classiication stage is the new aspect of this paper. We use a new nonlinear feature extraction scheme called the maximum representation and discriminating feature (MRDF) extraction method to compute nonlinear features that are used as inputs to a classiier. The MRDF is shown to provide better classiication and a better ROC (receiver operating characteristic) curve than other methods.