| Terahertz(THz)waves are electromagnetic waves with a frequency between 0.1-10 THz,and its waveband is between microwave and infrared wave,with a wide spectrum,high penetration,low energy,fear of water and other important characteristics,in medical diagnosis,non-destructive testing,heritage protection,radar communication,security and anti-terrorism and many other fields have shown its great scientific significance and application value.Terahertz three-dimensional imaging can not only reflect the surface or overall information of the sample but also obtain the internal structure information of the sample,which has become an important branch in the field of terahertz imaging.THz computed tomography(CT)is a non-destructive,markerless imaging method to detect the internal structure distribution of a sample to be measured in 3D.By detecting the projection of THz waves through the sample at different projection angles and reconstructing the 2D projection data(i.e.,sinogram),the 2D internal structure cross-sectional image of the sample to be measured is obtained,and the 2D cross-sectional reconstructed image at different positions can be obtained by changing the longitudinal position,and then the 3D internal structure information of the sample to be measured is further obtained.The continuous terahertz CT system is simpler and more powerful than the time-domain pulsed terahertz CT system.However,the current continuous THz CT imaging system does not meet the requirements of most applications in terms of real-time data acquisition.The current research status of THz CT imaging is fully investigated,and the current methods to improve the imaging rate and ensure the reconstruction quality in terms of reconstruction algorithm,imaging optical path,and mechanical control are investigated.The acquisition process and reconstruction algorithm of the THz tomography method are described in detail,and the process of the compressive sensing algorithm applied to THz CT imaging is described in detail.The Lab VIEW automatic acquisition system for continuous THz tomography imaging is built,starting from the system requirement analysis,from the overall requirement analysis to the requirement analysis of individual instruments,and designing the front panel and the main framework of the program according to the requirements,clarifying the framework structure,dividing the functional modules required in the system,and explaining the design scheme of each functional module one by one.The system can realize the cooperative acquisition of the rotary stage,translation stage and detector in the THz CT point-scan imaging system,and the system is used in the experiments to acquire the scanned images of water pipe samples as the rotary stage is gradually rotated from 0° to 180°.The galvanometer scanning method was introduced into the continuous THz CT imaging,and by building a theoretical model,the voltage relationship corresponding to the scanning position of the galvanometer was analyzed,and the scanning frequency of the vibration mirror was determined by theoretical and simulation studies.The aberrations of the oscillator in the scan are simulated,and quantitative analysis and correction are performed.A continuous THz CT imaging system based on the scanning of the galvanometer was constructed using a 2.52 THz optically pumped continuous THz waves laser and a THz detector with multiple gold-plated off-axis parabolic mirrors and galvanometer.The scanning area of the system was calibrated,and the 2D sample Siemens was scanned using uncalibrated and calibrated THz beams,and the results of the 2D scanning with and without calibration were compared and analyzed.The 3D reconstruction of the three-dimensional sample polyethylene material plastic water pipe was performed to quantitatively evaluate and analyze the reproduction results of the continuous THz CT imaging method based on galvanometer scanning.The compressed sensing reconstruction algorithm is applied to the continuous THz tomography method with the acquisition of less projection angle data.In the experiment,a self-built Lab VIEW automatic acquisition THz CT imaging system is used to acquire and reconstruct sea urchin and nylon tube samples,and the fidelity of the 3D reconstruction results is quantitatively evaluated and analyzed,and the defects are marked with special locations to verify the feasibility of this automatic acquisition system.The reconstruction results of this algorithm and the FBP filtered inverse projection algorithm were compared by reducing the projection angle data,and it was verified that this algorithm has better noise immunity and higher reconstruction fidelity when there are fewer projection angles.It is also verified that the projection angle data required for reconstruction can be reduced by using the compression-aware reconstruction method to improve the acquisition real time performance. |