Foreign object detection via texture analysis

This paper describes a monitoring system which detects contaminants such as pieces of stone or fragments of glass in food stuffs. The X-ray images of the food stuffs are irregular and highly textured and are thus analysed using an artificial neural network technique for texture recognition. The advantage over the existing methods of inspection is that they achieve a greater capability of reliably locating the contaminants. This paper will show that neural nets comprised of a few nodes can be successfully used in the real-world for industrial applications.

[1]  T. J. Stonham,et al.  Texture image classification and segmentation using RANK-order clustering , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol. III. Conference C: Image, Speech and Signal Analysis,.

[2]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[3]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[4]  Edward Roy Davies,et al.  The use of variance and entropic thresholding methods for image segmentation , 1995, Pattern Recognit..

[5]  F. Ade,et al.  Characterization of textures by ‘Eigenfilters’ , 1983 .

[6]  Li WangDong-Chen He,et al.  Texture classification using texture spectrum , 1990, Pattern Recognit..

[7]  E. R. Davies,et al.  Automatic Detection Of Foreign Objects In Food Products , 1994 .

[8]  E. R. Davies,et al.  Soft contaminant detection using neural networks: techniques and limitations , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[9]  Richard C. Dubes,et al.  Performance evaluation for four classes of textural features , 1992, Pattern Recognit..