Research On Key Technology Of Screw Thread Detection Based On Image Stitching

Screw threads are widely used in the machinery industry because of their good connection and transmission properties.In order to ensure the reliable working performance of the thread,it is very important to ensure the dimensional accuracy of the thread.High-precision and high-efficiency thread detection method is an important means to ensure the accuracy of thread size,and is also an important research content in this field.In view of the shortcomings of the hardware on the current machine vision thread detection equipment,such as the small field of view of the lens,the low precision of the guide rail,and the large installation error of the system components,etc.,on the measurement results,a thread parameter detection based on image stitching is proposed Method,and verify the method through comparative experiments.The research content of this topic mainly includes the following aspects:(1)Based on the principle of machine vision and the principle of gim-80 a vision detection system,this paper proposes a thread parameter detection method based on image splicing,which takes into account the detection speed and accuracy.In this method,only the upper and lower profiles of the thread are scanned to achieve image splicing and improve the detection speed of the system.The method of image mosaic is combined with high-precision grating ruler to improve the precision of image mosaic.(2)The algorithm of image processing and parameter detection is studied.Through the research of image stitching algorithm,high-precision stitching of thread image is realized,and the local image of thread acquired is stitched into high-definition global image of thread.After preprocessing the global image of the stitched thread,the corresponding thread parameters,including major diameter,minor diameter,pitch and profile angle,are obtained by the designed thread parameter detection algorithm.The above algorithm is verified by comparative experiments.The experimental results show that the measurement method meets the requirements of accuracy and stability.(3)Analyze the thread measurement error.The measurement errors in the process of thread detection are analyzed from three aspects: the cleanliness of the thread surface,thehardware and the software of the system.In view of the above factors,the influence of various factors on the measurement results is studied by means of experimental comparison,mathematical calculation and theoretical analysis,and the corresponding methods in the research process are described.In this paper,the cylindrical external thread is taken as the detection object,and the detection method of thread parameters based on image splicing is studied.The whole scheme,related algorithm and error analysis of the detection method proposed in this paper also have good reference significance for other types of thread detection.It has a good reference value for the design of non-contact thread detection system,especially for the improvement of GIM-80 A thread detection system used in the research.