Machine Vision Based Beverage Bottle Packaging Inspection And Capture Technology

The modernization of the production line provides a lot of convenience for the production of beverages along with the scientific development,but the emergence of substandard products usually occurs in the beverage filling process,which mainly includes unqualified cap package liquid level and so on.This will seriously affect the quality of the product and brand benefits.The traditional artificial detection method,not only has the large labor intensity,the low detection efficiency and accuracy but also increases the burden on enterprise along with the growth of labor costs.So in order to ensure the quality of the product and reduce the production cost,online detection system based on machine vision has important significance for production line production.In this paper,taking the beverage bottle as the detecting object.the research and design work of bottle packaging quality defect inspection and mechanical arm grasping technology is carried out by means of machine vision technology,robot arm technology and industrial control technology.The specific content of the paper includes:(1)Scheme design of defect detection and mechanical arm grab system for beverage bottle is completed based on machine vision.The hardware selection,software design and the whole system are introduced based on machine vision and robot arm related technology.(2)In this paper,the two problems of bottle cap packaging and beverage filling which are whether qualified are researched.The bottle defects are classified as no cover,crooked cover,high cover three,filling quality is divided into high level and low level in two cases.The bottle cap detection algorithm based on the straight linear fitting method and the liquid level detection algorithm based on the specific region level pixel statistics.Finally,the unqualified beverage bottles can be detected well through the algorithm synthesis.It effectively avoids the misjudgment caused by the shaking of the beverage bottle in the transmission process,the accuracy of defect detection is improved.(3)A vision guided grasping manipulator system is constructed,the ABB manipulator as the main body,the industrial camera,conveyor belt and mechanical arm are combined together through the calibration of camera and hand eye calibration.The bottle of substandard drinks are grabbed out based on the machine vision system.And in accordance with the requirements of the specified location.(4)A human-computer interaction system for beverage bottle defect detection is studied based on VC++,and the system can realize the detection and control function of system.Packaging detection and grabbing system for beverage bottle based on machine vision not only can complete beverage bottle cap and liquid level defect detection but also grabbed out the unqualified product form the production line through the mechanical arm to avoid the traditional way that may make the drink bottle down on the conveyor belt.Therefore,the system improves the production efficiency,reduces the production cost,and has certain reference significance for other production lines of similar products.