The interaction between water and fabric is critical in the physiological comfort of garments. However, it is still practically challenging to evaluate textiles with continuous directional liquid moisture transport and outstanding prevention of water penetration in the reverse direction. The traditional liquid moisture transmission test method has its shortcomings. In order to better evaluate the liquid water transport behaviors of fabric, a new method is proposed in this paper. A series of images of fabric wetting are acquired from two different perspectives through a self-made double-sided image acquisition device. At the same time, a suitable segmentation algorithm for the image sequence is proposed in this study. Finally, the index parameters of fabric wettability can be obtained, and the wettability process of fabric can be analyzed. The experimental results show that the method of double-sided imaging can obtain more comprehensive fabric wetting information. Combined with the existing image-processing algorithms, the double-sided imaging image acquisition system proposed in this paper can be used to form a complete evaluation system of fabric liquid moisture transport behaviors.