Online Detection Method Of Mechanical Error For Bowl Plug Based On Binocular Vision

Bowl plugs are used for plugging auxiliary holes on the engines, the processing precisions are high, required diameter tolerance is micrometer. At present the bowl plugs mechanical errors are mainly manual detections, manual detections precisions are low and can not meet the requirements of modern production online detections. Use machine vision detection will face the following problems:(1)bowl plug has many kinds of diameters, it is difficult to choose suitable lens and cameras to meet different diameters detection requirements;(2)with a single camera can directly obtains the diameter, but the resolution of a single camera is difficult to meet the micron-scale resolution requirements;(3) the impurities on the bowl plug surface have a greater impact on the test results in the image.For these problems, this paper proposes two cameras bowl plug machining errors online visual detection method, the deformation will cause some bowl plugs diameters varies in different directions, the bowl plug on a mechanism for uniform rotation, two parallel cameras acquire multi groups diameter end face images. Use the proposed end face impurity removal method calibration test results, through data processing to obtain bowl plug maximum diameter value, the minimum diameter value, diameter and taper mechanical errors. In addition, using one camera for bowl plug machining errors online visual detection, one camera acquire bowl plug images in different directions and set up a bowl plug gallery, select images through man-machine interface to get the maximum diameter value, the minimum diameter value and the diameters tolerance of bowl plug through edge detection. Set up two complete sets of bowl plug machining errors online visual detection system.Experimental results show that these two plug bowl visual detection system can detect bowl plug eligibility, binocular vision detection system has strong versatility, can meet the demand of different size bowl plug micron level resolution detection, do not need to increase the camera resolution for large size detection, greatly reduce the cost of equipments and suitable for assembly line works, detection speed is much higher than manual detection.