Image segmentation apparatus

• US 6,813,379 B2
• Filed: 08/21/2001
• Issued: 11/02/2004
• Est. Priority Date: 08/23/2000
• Status: Active Grant

• Current Assignee: KDDI R&D Laboratories Inc. (KDDI Corporation)
• Original Assignee: KDDI R&D Laboratories Inc. (KDDI Corporation)
• Inventors: Matsumoto, Shuichi, Kawada, Ryoichi, Koike, Atsushi
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/933,073
• Publication Number:
  US 20020041711A1
• Time in Patent Office: 1,169 Days
• Field of Search: 382/173, 382/108, 382/232, 382/233, 382/236, 382/238, 382/243, 382/239, 375/240, 375/240-1-2, 375/240.08, 375/240-2-2, 375/240-8-2, 375/240.27, 375/240.28, 375/240.24, 375/240.06, 348/554, 709/247
• US Class Current: 382/173
• CPC Class Codes: H04N 19/537 Motion estimation other than block-based
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression

Abstract

A shape entropy, a motion vector entropy, and a texture entropy of a certain area i of an input image are derived by a shape entropy calculation section, a motion vector entropy calculation section, and a texture entropy calculation section, respectively. Among them, the texture entropy is derived by taking an allowed value n of encoding degradation given according to a limit (encoding rate) of an entropy which can transmitted into consideration. Therefore, a generated entropy obtained by adding up the shape entropy, the motion vector entropy, and the texture entropy becomes a value according to the encoding rate. In such optimizing segmentation encoding as to minimize the generated entropy, therefore, it becomes possible to conduct segmentation according to the encoding rate. According to the present invention, there is provided an image segmentation apparatus capable of conducting such optimizing segmentation as to minimize the generated entropy while taking the limit (encoding rate) of the entropy which can be transmitted into consideration.

Claims

1. An image segmentation apparatus comprising:

first means for estimating a shape entropy of a certain area of an image signal;

second means for estimating a motion vector entropy of the region;

third means for estimating a texture entropy based on a signal variance of the area and an allowed value of encoding degradation determined according to an encoding rate; and

means for adding up the entropies derived by the first to third means and thereby estimating a generated entropy of the area.

2. An image segmentation apparatus according to claim 1, wherein the third means derives an estimated texture entropy E according to the following equation,

3. An image segmentation apparatus according to claim 1, wherein segmentation of an image signal is conducted so as to minimize a generated entropy of the image signal derived based on the generated entropy of the area.

4. An image segmentation apparatus according to claim 3, wherein the third means derives an estimated texture entropy E according to the following equation,