As the binary image fails to reflect the density distribution of the bubble, this paper proposes an improved new indicator for accurate measurement of spatial uniformity based on gray image analysis: moment. Compared with the inclination angle method in binary image, the test results show that the two methods have significant positive correlation. The experimental results also indicate that the evolution curve of bubble numbers is consistent with its moment evolution curve, and the moment evolution fitting curve can distinguish the high noise in the experiment. Compared with other methods, mixing performance index (t) obtained by moment evolution fitting curve is correlated with heat transfer performance (Pearson correlation coefficienta= -0.94). The variation in characteristic area of the bubble in the direction of bubble growth can characterize the heat transfer performance. The index of dispersion of the global moment and the local moment indicates that the local moment can be used to precisely quantify the spatial uniformity of boiling bubbles. This work could also be applied to study a variety of problems involving spatial uniformity through visualization techniques.