Image recognition method

• US 5,838,839 A
• Filed: 06/28/1996
• Issued: 11/17/1998
• Est. Priority Date: 12/01/1995
• Status: Expired due to Term

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.), Ministry of Economy Trade and Industry
• Original Assignee: Hitachi, Ltd., Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.), Real World Computing Partnership
• Inventors: Nagaya, Shigeki, Oka, Ryuichi, Seki, Susumu
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): PATEL, KANJIBHAI B
• Application Number:
  US08/672,759
• Time in Patent Office: 872 Days
• Field of Search: 382/278, 382/276, 382/298, 382/300, 382/299, 382/190, 382/205, 382/114, 382/281, 382/206, 382/240, 382/148, 382/218, 364/728.03, 364/728.07, 345/132, 342/189
• US Class Current: 382/299
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Abstract

The present invention is intended to provide an image recognition method which, in recognizing an input image including an object presumable to have continuity like a human figure, enables simultaneously detecting the shape of the object in the input image and cutting-out the object therefrom, and obtaining information on position of the object at the same time, and which also enables recognition of two objects having the same shape but having different sizes. The processing procedure of image recognition according to the present invention consists of: inputting an input frame image of an object by image data inputting means; preparing scaled-down images 1 to n at different resolutions by image size reducing means; calculating neighborhood correlation values for respective scaled-down images 1 to n by neighborhood correlation processing means and checking relationship between them at the same time; calculating the area and the gravity center of each image using pixel values and pixel coordinate value; identifying respective objects at respective resolution on the basis of their relationship; determining shape correlation of each object; and recognizing the shape of each object by comparing the obtained shape auto-correlation with a previously determined reference value.

Claims

1. An image recognition method for recognizing an image comprising the steps of:

obtaining an input image;

generating, from the input image, a series of scaled-down images each having different resolutions, wherein each scaled-down image is comprised of a plurality of pixels;

performing neighborhood correlation on each of the scaled-down images;

isolating at least one object from the scaled-down images;

detecting a shape auto-correlation of the object; and

comparing the shape auto-correlation of the object to a shape auto-correlation of a reference shape.

2. An image recognition method as defined in claim 1, wherein during the performing step:

storing coordinate information of pixels of the object, and

simultaneously determining a shape feature of the object and position information of the shape feature from the coordinate information.

3. An image recognition method for recognizing an image comprising the steps of:

obtaining an input image;

generating, from the input image, a series of scaled-down images each having different resolutions, wherein each scaled-down image is comprised of a plurality of pixels;

performing neighborhood correlation on each of the scaled-down images;

isolating at least one object from the scaled-down images;

detecting a shape auto-correlation of the object;

adjusting the object'"'"'s shape auto-correlation by taking the logarithm of the object'"'"'s shape auto-correlation;

comparing the object'"'"'s shape auto-correlation to a shape auto-correlation of a reference shape.