Oblique Incident Visual Inspection Method And System For Inner Wall Of Hot Rolled Steel Pipe End

Due to the advantages of high strength,good pressure resistance,earthquake resistance,energy saving and environmental protection,and recyclability,steel pipes are widely used in all aspects of industrial production.When the steel pipe passes through the automatic nondestructive testing equipment,it will leave a detection blind area at the pipe end.Therefore,it is of great significance to study the non-destructive testing method and equipment of the pipe end.First,an oblique incident visual inspection method is proposed to obtain the image of the inner wall of the steel pipe and extract the inner wall defects.According to the inspection parameters,the inspection principle of the pipe end inner wall defect inspection system is introduced,and the layout of the inspection equipment system,the inspection process and the design of the photoelectric system are completed.The required camera,lens and light source are selected and laid out.After completing the hardware selection and layout of the system,the method to improve quality under this detection method is studied.Firstly,calibrate the internal parameters of the camera according to the camera imaging model.Secondly,in order to deal with the geometric distortion problem that exists when the image of the inner wall of the pipe end is collected when the camera is installed at an angle,an image distance correction algorithm and an algorithm for the concave surface to be corrected to a plane are proposed.The MATLAB software is used to simulate and verify,and a good correction effect is obtained,which is convenient for subsequent image processing.Then,for the problem of camera exposure time and image motion blur under the rotating motion of the steel pipe during image acquisition,a blur kernel based on geometric distortion correction and rotating motion is established.After a variety of deblurring algorithms,compare the processed images and image quality evaluation indicators,the constrained least square method is selected for motion blur removal.Finally,the method of combining the CLAHE algorithm and the homomorphic filtering is used to reduce the influence of uneven illumination of the image caused by the side lighting of the dark field.Combined with the effect of image quality improvement,according to the local gray and gradient changes of the image,determine the presence or absence of defects and the location of the defects.Furthermore,through image shading processing,Roberts gradient operator,expansion and corrosion methods,the image contrast is enhanced,the defects are segmented,and the defect edge features are extracted.It is verified that the processing flow has a good effect on the extraction of defects such as inner bulges and pits.Synthesize the above research,a set of detection software suitable for this system is designed,and each functional module and software operation process are introduced in detail.Finally,an experimental platform is built and preliminary experimental verification is carried out.Compared with other methods,the method is easy to realize high-speed automatic detection and has the characteristics of no contact,no coupling agent,no blind zone,etc.Therefore,it has broad application prospects.