Research On Phased Array Ultrasound Image Restoration Method Based On Constrained Quadratic Programming

Phased array ultrasonic detection technology is widely used in the field of nondestructive testing(NDT)due to its advantages of large detection area,fast detection speed,and high detection accuracy.However,in the detection process,due to the effects of acoustic wave diffraction,transducer and electronic system electrical response and environmental noise,the resolution of the detected image is reduced,and the quality of the detected image is degraded.In order to solve the above problems,this thesis studies the method of improving the resolution of the phased array ultrasonic image by constrained quadratic programming.Firstly,based on the spatial impulse response(SIR)theory,the acoustic field characteristics of the phased array ultrasonic transducer are studied,and a simulation experiment of the phased-array linear array transducing acoustic field is performed.The influence of the number of elements,the spacing of the array element and the angle of the array element on the acoustic field of the phased array ultrasonic transducer is analyzed.The experimental results show that the acoustic field calculation method using the SIR theory can better reflect the acoustic field of the phased array ultrasonic transducer.By changing the parameters of transducer,the acoustic field can be controlled,which lays the foundation for the next step in the study of phased array ultrasound image restoration methods.Based on the study of spatial impulse response(SIR)theory and the characteristics of the phased array ultrasonic acoustic field,a phased array ultrasonic imaging discrete linear model combining the effects of acoustic wave diffraction,electronic response of transducers and electronic systems,and environmental noise was established.On this basis,the minimum mean squared error(MMSE)restoration method based on Bayesian estimation theory is studied,analyzed the recovery process to improve convergence constraint conditions,combined with the linear phased array ultrasonic imaging discrete model,established restoration objective function and restoration method based on constrained quadratic programming optimization.Finally,the proposed method is verified by simulation and experiment.The simulation results show that the constrained quadratic programming method and the MMSE method are better than the synthetic aperture focusing technique(SAFT)in the resolution of the reconstructed image.At the same time,it can be seen from the experimental results that the resolution improvement ability and noise suppression ability of the constrained quadratic programming method are better than the MMSE method.Experiments were conducted by collecting actual data through the Multi2000 ultrasound platform and using SAFT,MMSE,and constrained quadratic programming methods to image single and multiple defect experimental data,respectively,the imaging results were evaluated by lateral resolution and peak signal noise ratio(PSNR).The experimental results show that the lateral resolution of constrained quadratic programming is less than 1.5mm for the recovery of single and multiple defects,and the PSNR has a different degree in the range of 12.92%~20.12% compared to SAFT,and the effectiveness and feasibility of constrained quadratic programming in the restoration of phased array ultrasound images are verified.