Research On Amplitude Measurement Method Based On Machine Vision

Mechanical vibration of objects is a common phenomenon in life,and the amplitude detection is a very important part of its dynamic characteristic test.Through the way of machine vision to measure the value,the equipment used is cheaper,small amount of calculation,convenient for practical measurement,and may be applied in life in the future.Based on the understanding of the existing amplitude measurement methods,this paper discusses the problem of improving the measurement accuracy,and the amplitude detection method of fuzzy image generated by object vibration is studied and analyzed.(1)The corresponding design and selection of the main components of the vibration image acquisition system are carried out,and the lighting scheme of the backside light source is determined;build ultrasonic vibration and low-frequency vibration test platform,design ultrasonic vibration specimens,calibrate the camera,calculate the pixel equivalent,and complete the process design of amplitude detection algorithm.(2)The study of image preprocessing method.The preprocessing process of this paper is given,and the focus of the work is shifted to improving the image quality.Respectively in several filtering is often used to test approach to image noise,and the results were compared.The key is to preserve more tiny details of the image edge,and the corresponding operation is carried out by the bilateral filtering method.On this basis,a mixed bilateral filtering algorithm is proposed to improve the suppression effect of strong noise points in the image and facilitate the image follow-up processing.(3)According to reducing operation complexity and improving detection accuracy,the image edge detection method is studied.The method of extracting amplitude is to detect the edge of the fuzzy region of image vibration.Based on the analysis of the working principle and shortcomings of edge detection algorithms,a new Canny algorithm which can be automatically adapted is proposed.The connected component method is used to evaluate the detection results,and the improved algorithm has higher detection accuracy.In order to improve the detection accuracy,the fractal interpolation method with good edge retention effect was used to make the detection accuracy reach the sub-pixel level.Hough linear detection was used to fit the sub-pixel level edge,so as to obtain amplitude data and complete the detection of amplitude parameters.(4)Based on the vibration test platform and image acquisition system,the fuzzy image processing software program is developed.The vibration images under different conditions are collected and processed,and the amplitude information obtained is compared with the amplitude data detected by the displacement sensor under the same working condition.The stability and applicability of the amplitude detection method are verified.