Abstract: The micro pore structure of wood is an important factor affecting its macro water migration, heat transfer and mechanical properties. In order to quantitatively characterize the influence of pore structure on the above physical parameters, the first task is to obtain accurate pore structure parameter information. However, traditional pore structure detection methods are slightly inadequate in terms of three-dimensional pore structure evaluation and accurate quantitative analysis of parameters. In this paper, X-CT（X-ray computed tomography） technology was used to reconstruct the three-dimensional pore structure of wood, and on this basis, the pore structure parameters were quantified with digital image processing technology, in order to provide a new way for the detection and quantitative analysis of three-dimensional pore structure of wood. This study took larch and oak as the research objects, carried out X-CT detection on them, and performed 3 D volume rendering（3 D stereo images） on the scan results（two-dimensional multi-layered graphics） in Avizo software. On this basis, digital image processing methods such as image cutting, noise filtering, threshold segmentation and marker were used to identify the 3 D pore structure characteristics. Porosity（volume/surface）, pore area and pore volume, pore size distribution and connectivity were quantitative calculated. The results showed that in the 3 mm cubic pine sample, the pore size range was 9.65-79.51 μm, the average pore size was 26.03 μm. the pore volume was 1.58×10¹⁰μm~3, the pore area was 1.65×10~9 μm~2, the volume porosity was 58.46%, and the surface porosity range was 56.61%-59.64 %. The area and volume ratios of axial interconnected pore were about 78.5 % and 80.3%. In the same size oak samples, the pore size range was 11.01-319.68 μm, the average pore size was 51.24 μm, the pore volume was 1.01×10¹⁰ μm~3, the pore area was 1.12×10~9 μm~2, the volume porosity was 37.57%, the surface porosity range was 36.05%-39.07%. The area and volume ratios of axial interconnected pore were about 74.4% and 67.3%. This study qualitatively and quantitatively characterized the wood pore structure based on X-CT technology, which can provide data support for the reconstruction of pore structure of wood and the simulation of thermal conductivity, seepage and diffusion, and provide a new idea for the study of microscopic pores of wood.