Research On Gear Defect Detection Technique Of The Shape Of Small Modulus Plastic Gear Based On Image Fusion

With the rapid development of computer vision technology and photovoltaic technology,the technology of visual detection is developing rapidly and used commonly.As a means of non-contact detection,visual detection technology is attracting increasing attention.Because plastic gear does not only have excellent features such as lighter mass,cheaper price and lower transmission noise,but its strength and stiffness are also close to that of metal materials that make it able to replace some non-ferrous metals and alloys,so it has been widely used in a variety of micro-mechanical,electrical products and precision instruments.Plastic gear has become a very important part in modern manufacturing.So the manufacture precision of gears are of great importance on their performance.Ningbo is a base of our country's bearing manufacturing,and there are more than 1,000 bearing enterprises there.In recent years,many producing enterprises have accomplished the automatic manufacture of bearing through the technological revolution.However,it is still the manpower visual measurement to detect to the surface defect of gears and to reject the nonconformity ones.The method is not only much workload and inefficiently,but also low reliable.It affects products quality when they leave factory,for which users often returned.This has not only caused economic losses to the enterprises,but also damaged their reputation.Making use of visual inspection and computer technology in gear quality detection can improve the detection rate of gear products and reduce production costs,what is very important to the rapid development of enterprises and improving the competitiveness of enterprises in the market.The research and application of international computer vision system has begun from 1980s.In recent years,computer vision inspection is booming.New conception,new theory and new technology come forth continuously,which have been applied on vision inspection and automatic recognition of all kinds of products.In China,industry vision system is still in a time of conception leading.Although the theory and experience of computer vision inspection have got some results,it has just started and dropped behind some countries.Detection of metallic gear is an important link in modern gear technology.Some colleges and scientific research organizations have gradually developed the theory researches of the defect detection and parameter detection based on computer vision,they also have achieved some progresses and realized some applications,which provide good foundation for the quality detection of small modulus plastic gear.Combining the project of industrial science and technology of Ningbo city(2005B100014)and the cooperative project of Ningbo academy-corporation,the structural characteristics and defect characteristics of small modulus plastic gears were studied deeply,and a virtual instrument based on computer vision inspection system was introduced.The system completes visual inspection based on LabVIEW,with industrial camera and 1394 image acquisition cards and using certain image processing technology.The short development cycle and fast test speed of the system give full play to the characteristics of formidable function and strong extension of virtual instrument technology.What this article examined is the small modulus plastic gear.This kind of gear is divided into the upper wheel piece and the lower wheel piece.The sizes of the two wheel pieces are different.The upper piece is smaller and it falls adding on the lower wheel piece,so the outline contrast of the upper wheel piece in image is not very intense,this makes it difficult to exam.The two wheel pieces both need examination,which causes the common examination used back-light unsuitable.This paper systematic discussed the overall plan of the examination system,and carried careful theoretical analysis of hardware demand as well as the selection principle of the system,and finally determined the hardware of every part.Image Pre-processing is an important part of target detection and identification,and it is a key step of the small plastic gear defect detection system.The noise types of digital image were analyzed,several common image filter were introduced,the limitations of each filter algorithm were pointed out through the analysis of characteristics of small modulus gear digital images,and finally the edge maintain filter was selected to conduct the image pre-processing.Image registration and image fusion are necessary preconditions for making the testing template.Therefore,the definition of image registration,the mathematical model and several common image geometric transformations were introduced,and several current used image registration methods were compared,and finally the method of image registration was decided to use in this paper.This paper also studied the multi-focusing image fusion algorithm that based on wavelet decomposition,carried on the wavelet decomposition to the gear image,used different wavelet bases to carry out the image fusion,and carried the quantitative analysis on the fused gear image through evaluating indicators such as the image entropy,the average value,the standard deviation and the average gradient and so on to choose the suitable system wavelet base and the best decomposition layers.This paper proposed the research technique of the examination of the small modulus plastic gear tooth profile flaw.This technique based on the unique feature of the small modulus plastic gear and the multi-focus imaging principle,and applied the multi-focus image fusion technology and LabVIEW-based virtual instrument technology.And combined with Virtual Instrument Development Platform LabVIEW and image processing software Vision Builder AI, the paper developed the automatic defect detection procedure of plastic double-gear flaw.This paper also evaluated the performance of the detection system,and verified the practicality and feasibility of the system through the sample image analysis of a group of small-modulus plastic gears.At present,because we can't realize the automatic registration and fusion of the small plastic gear modulus image that we gathered,so the automatic on-line detection can not be achieved.This paper used Virtual Instrument Development Platform to complete the visual detection system,this method fully displayed the formidable function and strong extension characteristics of virtual instrument technology.And through the open characteristics , the high degree of programming flexibility and the excellent interactive interface of LabVIEW,the high complex application development became simple and flexible,and thereby,and this saved development costs and improved the cost-effective of the system.Therefore,the machine vision system based on virtual instrument will be a hot topic in the field of machine vision technology for a long period.