The cell structure uniformity of microcellular polymers significantly impacts material performance, especially for low-porosity microcellular TPU used in chip polishing. The distribution of the cell structure of polishing pads directly affects the removal rate and process repeatability. Despite its importance, no quantitative method for evaluating cell structure uniformity has been reported in the literature. In this study, a digital image processing method that involves morphological operations of scanning electron microscopy (SEM) images, binarization, and cell localization, and the statistical evaluation of cell structure parameters was established to evaluate cell structure uniformity. A quantitative metric, the cell structure uniformity index (CUI), was calculated based on cell structure indices, incorporating the cell size index (Ud), the cell number index (Un), and the cell local spacing index (Ur). By establishing an ideal model and analyzing representative SEM images, the effectiveness and efficiency of the method for evaluating cell structure uniformity of microcellular TPU were successfully validated. The results demonstrated that low-porosity TPU foams exhibited relatively low cell structure uniformity compared to the ideal model. The heterogeneous nucleation process in TPU caused non-uniform cell structures due to the temporal and spatial non-homogeneities during the early cell nucleation process. As the cells grew, they merged and reduced the distance between them, resulting in improved cell structure uniformity.