| 摘要 |
<P>PROBLEM TO BE SOLVED: To appropriately detect a specific object contained in an image. Ž<P>SOLUTION: With respect to a photographed image output from an image acquisition part 231, an image reducing part 232 sequentially reduces and generates a plurality of the images which differ in size. With respect to each of the images output from the image reducing part 232, an imaging part 233 sequentially photographs the image (the image for the object to be determined) contained in the designated area for each of the images. By using a human body detection dictionary stored in a human body detection dictionary storing part 300, an HOG calculating part 234 specifies a plurality of the areas in the object images to be determined, output from the imaging part 233, and calculates the HOG for each of the areas. By using the human body detection dictionary stored in the human body detection dictionary storing part 300, a human body determining part 235 calculates the difference value of the designated class of the calculated HOG, and determined if the human body is contained in the image output from the imaging part 233, based on this difference value. Ž<P>COPYRIGHT: (C)2010,JPO&INPIT Ž
|
