A system and method for comparing captured sequences of in-vivo images with (e.g., template or model) sequences, for example, for computer-automated recognition of contractions. The size of the opening of an in-vivo lumen passageway represented in each frame in a subset of frames of an image stream captured in vivo may be measured. Frames in the subset of frames of the image stream having a local minimum size of the lumen passageway may be identified. The subset of frames may be divided into segments of sequential frames at frames having local maximum lumen sizes before and after the identified frame having a local minimum size of the lumen passageway to generate contraction sequences. A plurality of the contraction sequences may be compared to template sequences. A plurality of the contraction sequences may be displayed.