Development Of Online Visual Inspection System For Ice Cream Bar Contour Defect

Ice-cream bars,also known as snow sticks,are often used as ice-cream handles and can be used to make various handicrafts.By the ice cream bar itself and the impact of processing steps,ice cream bar products often with various types of defects.At present,most of the domestic ice cream bar manufacturers of ice cream bar defects detection methods are mainly manual detection.Manual testing mainly relies on the subjective judgment of the workers,and the results are unstable,so it is easy to detect and miss the defects of ice cream bars.In order to make up for the shortage of artificial detection,it is necessary to study a set of automatic detection method of ice cream bar defect.Aiming at the ice cream bar produced on industrial production line,this paper detects four kinds of common defects,such as Burr,flat head,meat shortage and partial first,by image processing method.The aim of this paper is to establish a set of on-line detection method for ice cream bar with high automatization,fast detection speed,missing rate and low false detection rate.The contents of this paper are mainly focused on the following aspects:First of all,according to the quality requirements of ice cream bar production enterprises to choose the right camera,lens,light source,PLC and other hardware equipment to build machine vision imaging system.By adjusting the parameters and relative position of each equipment,the best imaging effect is obtained,and then the image acquisition is completed and the defect image base is built.Secondly,the paper analyzes the characteristics of the burr,flat head,meat shortage and defect of the ice cream bar,and uses the Halcon image processing software to write the corresponding defect processing algorithm based on a series of operators,such as edge detection,region growth and curve fitting.The defect processing algorithm is tested by the Defects Gallery,the test results are recorded and the errors are detected,and the deficiencies of the defect processing algorithm are compensated according to the test results.Finally,based on the visual Studio 2010 development platform,a good software architecture is designed using the C + + language.It includes Man-machine interaction interface,multithread module,defect detection algorithm module,communication module and motion control module based on MFC.To complete the collaboration between the above modules,and repeatedly debug until the entire software system can work properly.Finally,according to the characteristics of the ice cream bar and the current situation of the ice-cream bar production line,this paper designs an online visual inspection system of ice-cream bar contour defect,which is based on image acquisition,image processing,data display and motion control,and realizes the real-time on-line detection of the defects of the ice cream bar After testing,the system can detect 10 sticks per second,the average missing rate of 0.63,to meet the requirements of the enterprise detection speed and precision,in the industrial site to be applied.