Recognition and measurement of drug-release hole based on machine vision algorithms

It is an important issue to precisely measure and control the size of the drug-release hole in automatic production of osmotic pump controlled-release (OPCR) tablets. In this paper, we focus on the recognition and measurement of drug-release hole based on machine vision algorithms. Firstly, we setup a test-bed which consists of a personal computer, an industrial digital camera and an industrial zoom lens. Based on the test-bed, we develop algorithms to recognize and measure the drug-release hole automatically. The drug-release hole is recognized by using a combined edge detection algorithm, where Canny operator and mathematical morphology are combined to improve the accuracy of edge detection. For measurement purpose, we propose a method to measure the equivalent diameter of the drug-release hole. Experimental results confirm that high measurement precision can be achieved by using our scheme.