Dynamics Analysis Of Tire Balancing Machine And Study Of Positioning And Recognition Method For Tire Label

Tire is one of the crucial parts of the automobile.Too excess unbalance on the tire will lead to the sustained vibration of the automobile,accelerate the breakdown of the other parts,affect the ride comfort and stability.And even worse it will cause personal injury.Therefore,it is important to test and screen the unbalance on each tire with a balancing machine at the end of tire production.Presently,the test repetition accuracy of the domestic balancing machine for a single tire has been improved to the level of the abroad machine.However there are still deficiencies in the sides of convenience and efficiency with the machine operation.In addition,with the development of information technology,the tire manufacturer puts forth new requirements,which include the data recording and data tracing of the unbalance.In order to improve the automation and informatization of the tire production as well as the competitiveness of domestic tire balancing machine,the dynamics of the tire balancing machine,the conversion method of the influence coefficient of the machine,and the automatic positioning and recognition method of tire label are studied.The work and achievements are as follows:(1)The dynamics of the tire balancing machine is analyzed,and the conversion model of the influence coefficient is put forth.With the model the influence coefficients of all types of general tires can be converted only by measuring the influence coefficient of the calibration tire,which effectively improves the convenience and efficiency of the tire balancing machine operation.The test shows that the maximum relative error of the amplitude of the element of the converted influence coefficient is less than 1.2%,which can meet the unbalance test requirements in the tire production.(2)An automatic tire label positioning method is proposed.According to the characteristics of the tire label,a set of characteristic criteria of the tire inner circle contour and the tire label contour are summarized,the value ranges are determined.The automatic positioning and interception of the tire labels under the condition of complex background and small proportion of label area are achieved.The test shows that the average time-consuming of the method on PC is 0.43 seconds.The processing accuracy rate for 56 sample images is 100%(3)A recognition method for bar code on tire label is proposed.The bar code image is processed by column average,histogram specification and envelope threshold binarization in turn.Finally the bar code is decoded according to Code128 coding rule.The test shows that this method can accurately recognize 56 clear or fuzzy sample bar codes on the tires,and the time-consuming on PC is less than 4ms.(4)The recognition method of the character on the tire label is studied.With 5×5 regional feature template matching of the character image,the character recognition on tire label is realized.The test shows that the recognition accuracy rate of the serial number of 56 sample labels is 94.6%,which may be served as an alternative of the bar code recognition.