Research And Implementation Of A High-speed Imaging And Data Acquisition System Based On Magneto-optical Sensing

Ferromagnetic materials are widely used in petroleum,chemical,transportation and other fields,and their quality is decisive for the safety and reliability of the equipment,while magneto-optical imaging inspection technology is often used for non-destructive testing of ferromagnetic materials due to the advantages of visual inspection and high resolution.Magneto-optical imaging inspection technology combines magnetic leakage detection and Faraday spin effect,converting defect leakage magnetic field signal into light intensity signal through magneto-optical sensors,and displaying the light intensity signal containing defect information in the form of an image,so as to realize the visualization of defect detection.For the magneto-optical image sequences generated by magneto-optical imaging inspection technology at high speed and high resolution detection,it will increase unnecessary inspection costs if each frame of it is transmitted.Therefore,thesis designs a high-speed imaging and data acquisition system based on magneto-optical sensing and on-line processing of magneto-optical images,with the following main research contents:(1)Based on the magneto-optical imaging detection principle,the technical parameters of the magneto-optical film and light source are determined,and the light intensity changes in the reflective optical path structure are analyzed,calculated and simulated.According to the technical requirements of the system,the image sensor is analyzed and selected in combination with the light intensity change,and the technical index of the imaging optical lens is analyzed and calculated based on the technical parameters of the magneto-optical film and the image sensor,and the selection of the optical lens is completed.(2)Analyze and select the data interface,core processor and transmission interface required for the image acquisition module,core processing module and image transmission module according to the technical requirements of the system,design the hardware circuit and logic driver of each module,and build a system hardware platform for magneto-optical image acquisition,processing and transmission.The technical indexes of the system are tested,and the results show that the technical indexes meet the requirements.(3)The magneto-optical image characteristics of ferromagnetic materials are analyzed,and the feature extraction algorithms based on the number of gray-scale histogram maxima and the boundary width of gray-scale histogram distribution are studied.Using the number of grayscale histogram maxima as the statistical law features and the number of grayscale histogram distribution boundary width features as the physical model features,thesis studies and implements the FPGA-based magneto-optical image online processing algorithm.Firstly,the original data input by the image acquisition module is pre-processed by Gaussian filtering to eliminate noise and enhance the detectability of feature information;then,the adaptive threshold iterative magnetooptical image defect online detection algorithm is designed and implemented,which takes row data as the basic processing unit and uses sliding weighting to update the threshold value in real time to achieve adaptive and high-precision magneto-optical image defect online detection.The experimental results show that the magneto-optical image online processing algorithm can achieve the purpose of extracting and outputting images containing defective signals.