Significant difference filter. Significant difference filter changes a pixel to the mean of its surrounding pixel if the object pixel is “significantly different” from its surrounding pixels (shown as red dots in the figure). For the diagonal pixels only, we use an estimated by linear interpolation of the position that is 1-pixel distance to the referred pixel. “Significant difference” is defined as a difference greater than twofold of standard deviation (STD) of surrounding pixels for 2D mode and fivefold for 3D mode. The rationale behind this is that there is no detectable independent actin element as tiny as a 1-pixel element, so a 1-pixel region that is significantly different from its surrounding tends to be noise or bias due to z-axis projection.