Design Of Image Detection System For Pipe Inner Wall

With the rapidly development of pipe manufacturing technology,various types of pipes are widely used in metallurgy,petrochemical,military industries and so on.In particular,in the military field,pipe is a very crucial component of artillery weapons.Further more,the inner wall of pipe not only affects the technical specifications of velocity and the precision of weapons,but also affects the safety and reliability of weapons over the course of its whole life.Therefore,whether the detection of chrome quality of pipe inner wall in the manufacturing process or the regular inspection of the inner wall quality of pipe during its using process,it is very important in the military and engineering.In order to obtain images of pipe inner wall which is with high field of view and high resolution,an improved image detection system for pipe inner wall based on rotating scan imaging principle was designed in this thesis.Real-time collection of sequence images was realized,also,integrated design and auto-control of the measuring equipment were completed.The main work of this thesis are shown as follows:1.The overall design scheme of image detection system for pipe inner wall was completed.Including imaging module,measuring-car module,drive vehicle module,and the relevant control module.2.The hardware design and debugging of image detection system were completed,which included the mechanical structure design and assembly of measuring-car and drive vehicle,selection of relevant devices such as motors and camera,the synchronous control scheme and circuit debugging of these devices.3.The software design and debugging of image detection system were completed.The software of the system includes epigynous computer software and hypogynous computer software.The functions of the former are parameter setting,camera call,data saving and communication with the hypogynous computer.The functions of the latter are command receiving,parameter transmission and command execution.4.An improved image stitching method was proposed.It segments the sub-region that possesses the most abundant anomalous information by detecting the edge of the overlap region.Then it extracts SIFT feature points of the sub-region and matches them.After that,a novel fusion method based on the weight of Sigmoid function was used to realize the seamless fusion between sequence images.5.Experiment system is established to test the performance of drive vehicle,the function of imaging module and to analyze the results of image processing.The experimental results show that,the axial displacement accuracy of drive vehicle reaches 0.5mm/300mm;the measurement resolution is 50?m;computational efficiency of the proposed stitching method has improved about 35% in the whole registration process,the fusion results by this fusion method can satisfy application requirements.