Automated visual inspection for medicines in Aluminum-plastic blister Packaging

摘要

This paper presents an automated visual inspection system, which detects defects of medicines including tablets and capsules in Aluminum-plastic blister packaging. The system operates either in training or inspection mode. In training mode, the images of each type defect-free blister pack are used to extract the model. In inspection mode, the type of an inspected blister pack is determined and the defects of the medicines in the blister pack are detected in real-time. In this system, an improved definition of color histogram is adopted to determine the blister pack type correctly, by calculating the distance of the defined color histograms between the inspected image and each model. The eigenvector is designed, which distance could be used to judge the defects of tablet, while the color feature is utilized to detect the defects of capsule. In addition, an effective image skew correction method based on least squares technique is developed. Experiment results show that the system could effectively inspect typical defects of different tablet and capsule.