Considering the uneven brightness of the cell images collected by the phase contrast microscope and the low contrast between the cells and background, a fast cell image segmentation method based on dual-Gaussian filtering is proposed according to the characteristics of the sharp changes in the gray level of the cell area and the slow gray changes in the background area. The proposed method constructs a dual-Gaussian filter from the perspective of frequency domain to filter low-frequency information while retaining high-frequency information, which is beneficial to enhance the difference between cells and background and reduce the interference of uneven brightness factors. The cells are segmented using an adaptive threshold, and the morphological closing operation is used to improve their shape. The area constraint is used to remove the influence of impurities on the accuracy of segmentation to a certain extent. The proposed method was tested on a C2C12 dataset, and the accuracy, recall, and F values are 0.9770, 0.9457, and 0.9609, respectively, which were better than the comparison algorithms. The results show that the proposed method has good accuracy and robustness and can achieve fast and accurate cell image segmentation.