Development Of The Recognition System For Output Shaft Type Of Gasoline Engine Crank

Machine vision inspection technology has much virtue such as non-contact,on-line and real-time,rapid, suitable precision and strong scene anti-jamming.It could realize the aim to zero-wasted of mechanical product and has suited for the need of modern manufacturing progress and development. Thus, it has showed wide applied foreground. Combining machine vision technology and image processing technology, a gasoline engine crank automatic diagnosis system is finally designed and manufactured to judge the different model and prevent the wrong assembly which is to fast examine and fast contrast gasoline engine crank.This system suits our country actual situation, satisfies to the domestic Factories'precision need and the bearing of economy capacity.This working process of,this,system,a,process of image recognition,is as followed.Firstly we receive the image signal and process the image,pick-up the diagnostic parameter in the result.After computing each component's parameters aiming at the system precision's demand,we choose the right components to establish the machine vision system.It is composed of SI-M350 CCD, M2514-MP lens,Euresys Picolol image gabber card. Based on Evision platform, This system has established the picture examination and recognition systemframe.It has developed function modules such as image acquiring, image preprocessing, edge detection,image recognition and so on.the overall system is based on the Evision platform so that the usability of the system is enhanced.In this thesis, Roberts,Sobel,Prewitt and Laplacian are used to detect the edges. According to the actual need,we choose the Roberts operator,after analysising and comparing each kind of operator result.Then the image is iteration threholding segmented and binaried to be the best processed condition.In order to confirm the system,large numbers of experimental was made.The result indicated that this system whose efficiency is high works stably.and it reduces the waste in manpower,material and time.Thus it provides an efficient way to recognition motorcycle crank precisely and rapidly in actual production.At the same time, the machine vision technology can be expanded to other applications besides crank.It also provides a new method to other similar components.