Porosity Measurement Method Of Porous Materials Based On Multi-energy Spectrum CT Imaging

Porous materials,as one of the hot materials,are widely used in various fields such as medicine,construction and industry.Porosity,as one of the important indicators to evaluate the performance of porous materials,is of great significance to material quality detection.At present,the calculation method based on CT image is an effective method to measure the porosity of materials,and the threshold segmentation method is mainly used for calculation.The difficulty lies in how to accurately distinguish the porosity from the material in CT image.In order to identify smaller pores,higher resolution is needed,and higher resolution will increase the amount of data,which is more complicated to process.By establishing the sub-pixel model based on the single energy image,the measurement lower than the system resolution can be achieved,but it is difficult to be widely used in practice.Due to the inconsistency between the obtained multi-energy projection data and the reconstruction algorithm based on the single-energy hypothesis,the hardened artifacts appear in the reconstructed image,which affects the recognition of the pores in the porous material.To solve this problem,this paper proposes a method to calculate the porosity of porous materials based on multi-energy spectrum CT imaging,and studies the relationship between image reconstruction resolution and porosity value.In the process of research energy spectrum CT,considering the energy spectrum process is relatively complicated,the thesis first of all,under the condition of the energy spectrum of the unknown,more voltage X-ray projection image decomposition model is established,based on the non-negative matrix decomposition method to solve the energy spectrum more modern,with the nonconvex-project onto convex set(Nonconvex-Project onto Convex Set,NC-POCS for short)algorithm,realization of components(including pore)volume fraction image reconstruction,to get the volume fraction components in CT image,and then calculate the porosity of sample values;It is difficult to measure the attenuation coefficient of complex objects such as alloys and mixtures in practice,so on the basis of this algorithm,the material information is also taken as an unknown variable,and the corresponding linear attenuation coefficient and energy spectrum of the material are solved by using the non-negative matrix decomposition method to achieve decomposition and reconstruction.The effectiveness and feasibility of the algorithm are verified by experiments.Taking the porous material aluminum foam as an example,the changes of porosity under different resolutions are calculated and analyzed.In addition,the differences of the porosity calculated by the algorithm under two different scanning modes of full Angle and energy segment are also compared.Experimental results show that,compared with the traditional algebraic iterative reconstruction algorithm,the proposed algorithm can suppress the beam hardening artifacts,and the porosity values measured in the reconstruction results are less affected by the resolution in the full-angle and segment-energy scanning modes.