Design And Development Of Machine Vision Detection System For Foreign Body In Mold Cavity In Plastic Injection Molding Production

With its excellent properties,plastic has become a new basic material that goes hand in handwith steel,wood and cement.Injection molding machine is the most important forming equipment for plastics.During the mold closing stage of injection molding,if foreign matter such as plastic products that remain in the mold cavity does not fall off,it can easily cause safety accidents such as mold damage.The foreign body detection system is of great significance.Based on machine vision technology,this paper designs and develops an embedded system for visual inspection of mold cavity foreign objects.The main research contents of this dissertation are as follows:(1)A fast and accurate image matching algorithm based on global gray and local contour information is adopted.First,the pre-processed image is divided into multiple rectangular windows of equal size,and then based on the overall gray information of the image,the histogram is calculated and the similarity comparison is performed by the Bhattacharyya distance algorithm to determine whether there is a foreign body.Finally,the template matching algorithm based on local contour information can accurately detect the foreign objects in the cavity.The total time for detecting foreign objects in the cavity is less than 200ms,which ensures the real-time performance of the visual inspection system.(2)Establish visual inspection methods such as inverse color contrast,statistical judgment,and threshold self-learning.For abnormal images and uniform images,inverse color contrast algorithm and statistical judgment algorithm based on average and standard deviation are used for processing.The self-learning algorithm is used to determine the template threshold to eliminate the influence of factors such as unstable light intensity and mold jitter on detection accuracy.The accuracy and applicability of the visual inspection system are improved,so that the visual inspection system can accurately identify foreign body of 1 cm~2.(3)The design and development of the embedded cavity foreign body visual inspection system is completed on the basis of comprehensively considering the technical characteristics of machine vision,the actual working conditions of the injection production process and the corresponding process requirements.The visual inspection system is divided into two parts:hardware design and software design.The hardware part includes image capture module,operation recognition module,human-computer interaction module and machine control module.The software part includes image matching algorithm design and QT-based human-computer interaction interface design.The human-computer interaction interface is developed based on QT's embedded version of QT/Embedded human-computer interaction interface development software to realize visual operation.Finally,the developed cavity foreign body visual detection system is tested experimentally.The experimental test results show that the embedded vision detection system developed in this paper can meet the real-time and accuracy requirements of mold cavity foreign body detection in the actual production of injection molding.