Research on bottom detection in Intelligent Empty Bottle Inspection System

Abstract

Intelligent empty bottle inspection system is an important inspection equipment of empty bottle before filling beer, and it is a blend of machine vision, precision machine and real-time control. They need to cooperate perfectly to achieve the desired effect. In the design of the empty bottle inspection system, one of the key technologies is the bottle bottom detection which affects the speed and accuracy of the system. It includes positioning and defect recognition of bottle bottom. For the problems such as the slow detection speed and low detection precision of bottle bottom detection, some new methods are proposed in this paper. The positioning algorithm of the bottle bottom in images is studied after preprocessing the obtained images, and the accurate positioning is achieved by improving the Randomized Hough transform. In the defect recognition of bottle bottom, a method of calculating optimum radius in Fourier spectrum is used to solve the problem of the detection accuracy being influenced by the antiskid veins of bottle bottom. It can improve the recognition accuracy effectively. Experiments show the methods proposed in this paper can effectively improve the precision and speed of the bottle bottom detection. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.3341