An image input from an image input unit (1) is displayed on a display unit (4) and transmitted to an object extraction unit (2). An initial contour setting unit (5) sets a graphic of a predetermined size at a predetermined position on the screen and the display unit (4) superimposes the graphic on the input image to be displayed. The initial contour can be moved vertically and horizontally by a moving unit (6) and can be magnified/demagnified by a magnification/demagnification unit (7). When the initial contour is moved to a desired position and has a desired size, the contour is confirmed as the initial contour to be used for an object extraction by a decision unit (8). In the object extraction unit (2), an object is extracted by using the input image and the initial contour. The object extracted is recorded onto a recording medium by recording/reproducing unit (3). When this is reproduced, it is displayed on the display unit (4). Thus, it is possible to rapidly set an initial contour by a simple operation without providing a particular input device.