A system and method for comparing captured sequences of in-vivo images with (e.g., template) sequences, for example, for computer-automated diagnosis. Captured in vivo images may be divided into a plurality of captured image sequences. For each of the plurality of captured sequences, the processor may assign a first and second set of scores to each frame in the captured sequence and align each frame in the captured sequence with one or more frames in the template sequence for which the comparisons therebetween generate the minimum scores in the first and second sets of scores. The processor may define a match between the template sequence and captured sequences having a combination of scores in the first and second sets of scores for the frames in the captured sets compared with the one or more aligned frames in the template sequence, which are below or above a threshold value.