Texture analysis for foreign object detection using a single layer neural network

Inspection of food products for quality control to ensure that products are free from impurities (foreign objects) such as stone, glass or metal is a demanding part of a production process. This paper presents a method to detect foreign objects in bags of frozen vegetables and in particular using bags of frozen corn kernels. X-ray imaging is used to view the contents of the bag. We use principal component analysis (PCA) techniques to find the orthogonal vectors in data space that account for as much as possible of the variance of the data. The vectors are then used as the coefficients of the convolution masks. We briefly mention the various texture analysis methods using PCA and describe the artificial neural network texture description method used in this study.<
>