A system and method for performing auto-focusing operations for tomosynthetic reconstruction of images are provided. More specifically, embodiments of the present invention provide a system and method for efficiently computing the gradient of one or more depth layers of an object under inspection, wherein such gradients may be used in performing auto-focusing operations to determine a depth layer that includes an in-focus view of a feature that is of interest. In at least one embodiment, a method is provided that comprises capturing detector image data for an object under inspection, and using the detector image data for computing gradient information for at least one depth layer of the object under inspection without first tomosynthetically reconstructing a full image of the at least one depth layer.