| Terahertz wave has attracted much interest in recent years because of its long wavelength and low photon energy,and has strong perspectives and safety characteristics.At the same time,terahertz waves are widely used in medical imaging,nondestructive testing,and component research because terahertz waves can be used by some non-metallic and non-polar substances in industry.Terahertz coherence tomography is a rapid development of detection methods in recent years,with high resolution,non-invasive and non-destructive testing and other characteristics,in line with the current medical testing,safety inspection requirements.However,the current terahertz light source power is low,susceptible to detection sensitivity and electrical noise and other effects,making terahertz imaging technology is currently unable to achieve the desired clarity.At the same time,because the terahertz detection device can only use a single point of the detector,it is also facing the problem of slow scanning speed.This study discusses the above two issues,studied:(1)Using the compressed sensing algorithm to reduce the amount of data collected without affecting the image quality,thus improving the detection speed.And the simulation of the compression algorithm is verified and proved its feasibility.(2)For the collected low-resolution images,there are still some noise-affected,weak detail cannot be identified,need to take further image filtering,image interpolation algorithm to enhance image quality,improve resolution,get high resolution Rate image,extract more image detail.In this paper,image filtering and image interpolation methods are studied.(3)In this paper,several major image interpolation algorithms are studied,and the multi-threshold edge detection-edge-based image interpolation algorithm and covariance adaptive interpolation algorithm are implemented in software.It is applied to terahertz imaging.The advantages and disadvantages of the algorithm and the time complexity are statistically analyzed. |
PDF Full Text Request