The Research On Visual Inspection System For Quality Of Product In Healthcare Wine Production Line

With the rapid development of automatic production in alcohol, beverage and medicine manufacturing, people are not only satisfied with product function, but also its high quality. However, the traditional detection method of canned solution product made by human can not meet the need of the current product standard and production rate. Hence, a straight-line visual inspection system for healthcare wine is proposed in this paper, which broke out the design of the rotation-pause mode. The robot deals with the difficulty of detecting solution in irregular bottles, which has great effect on economy and society.First of all, the prospect of alcohol, beverage and medicine markets is analyzed. Additionally, the requirement of canned product detection is investigated. Machine vision technology and its application in all the fields are presented. Meanwhile, the components and the development of detection systems based on machine vision are described in the paper. Taking some canned medicine products for example, automatic devices at home and abroad are researched and the features of the current technology are concluded.Secondly, owing to the analysis of lighting methods, camera lens, camera and the classification of machine vision system, the machine vision platform for healthcare wine detection is built up. Based on this, a ’straight-line structure’ is put forward to resolve the problem of non-cylindrical bottle detection and key points of healthcare wine detection system are studied. Besides, the paper pays attention to the design of rotation device, classification device, detecting device and electrical control.Furthermore, a tracking-detection method through series of images is developed to deal with the difficulty of moving target inspection. A new orientation method is raised through region growing algorithm. The static background is set up by mutual variance parameter between frames. Combining centroid iteration algorithm with Kalman filter realizes tracking motional particles. Foreign particles in the solution are recognized by comparing the direction of the track.In the end, some development matters and the performance of machine vision software are introduced. The need of healthcare wine inspection system is analyzed. The module design of healthcare wine visual detection software is brought forward to satisfy users. Besides, functions of information management, production inspection, data inquisition modules in the software are exhibited.