Author/Editor     Derganc, J; Likar, B; Bernard, R; Tomaževič, D; Pernuš, F
Title     Automated visual inspection of color tablets in pharmaceutical blisters
Type     članek
Source     In: QCAV 2001. International Conference on Quality Control by Artificial Vision. Vol 1. Vol 2; 2002 May 21-23; Le Creusot, France. Toulouse: Cepadeues,
Publication year     2001
Volume     str. 415-20
Language     eng
Abstract     In pharmaceutical blister packing, it is today part of the recommended security standard to inspect tablets in each blister before it is sealed. In this paper a machine vision system is described, which detects missing and broken tablets, tablet fragments, as well as the color, size, and shape of individual tablets. The system operates either in "training" or "inspection" mode. In training mode, the image of a defect-free blisters is used to extract the blister model, which is composed of the shading correction function, position of the blisters, positions of tablets inside blisters, labeling function, and the color, position, size and shape of a tablet, and their tolerances. The blister model allows effective and on-line analysis of blisters in the inspection mode. The most important part of the system are the correction of the adverse effects of color shading and color segmentation. A method recently proposed for grayscale shading correction has been modified to correct color images. A novel nonparametric cluster-based segmentation method is proposed. The experimental results indicate that the segmentation method preceded by shading correction accurately extracts clusters with complex shapes. The machine vision system could thus be used in an industrial environment where on-line inspection of color tablets in blisters is required.
Descriptors     TECHNOLOGY, PHARMACEUTICAL
TABLETS
QUALITY CONTROL
PATTERN RECOGNITION