PURPOSE: To provide a picture processor which can convert clearly the temporal displacement of an object, etc., in a video into a locus on a still picture (space) even in the case that the motion of camera operation, etc., in contained in the video.
CONSTITUTION: In the picture processor which converts video information obtained by photographing continuously the object into the still picture, a video converting means 105 to similarity-transform the video so that the size and the position of the photographed still object in the video become equal between each frame and another, and a weight coefficient calculating means 106 to calculate a weight coefficient from the video similarity-transformed by the video transforming means 105 by using the feature of the temporal displacement of the video, and a weighted integration means 107 to integrate the transformed video in the time direction by using the weight coefficient calculated by the weight coefficient calculating means 106 are provided.
TONOMURA YOSHINOBU
Next Patent: THREE-DIMENSIONAL SHAPE RESTORING METHOD