Image processing apparatus and image processing method

• US 7,982,774 B2
• Filed: 09/06/2005
• Issued: 07/19/2011
• Est. Priority Date: 09/13/2004
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Tsunashima, Nobuhiro
• Primary Examiner(s): Tran; Sinh
• Assistant Examiner(s): Peterson; Christopher K
• Application Number:
  US11/218,613
• Publication Number:
  US 20060056702A1
• Time in Patent Office: 2,142 Days
• Field of Search: 348/208.6, 348/208.14, 375/240.12, 375/240.16, 382/107, 382219-221
• US Class Current: 348/208.6
• CPC Class Codes: G06T 7/254 involving subtraction of images

Abstract

The present invention provides an image processing apparatus detects only an object to be detected by removing a disturbance, such as the shaking of trees by using a very simple process and parameter. The image processing apparatus includes a reference image storing unit for storing a plurality of past image data photographed by a video input unit as reference image data, and an image moving object detecting unit for detecting the difference of pixels or difference in small areas between the image data photographed by the video input unit and each of the plurality of reference image data stored by the reference image storing unit and detecting a moving object in the image data based on the continuity or appearance probability of the difference.

Claims

1. An image processing apparatus comprising:

a reference image storing unit configured to store a plurality of past image data, photographed by a video input unit, as reference image data;

a moving object detecting unit configured to detect an image difference based on one of a difference of each pixel and a difference of each pixel in small image areas between image data photographed by the video input unit and the reference image data; and

an image correcting unit configured to correct a luminance of the reference image data by

determining an average luminance value for the reference image data,

calculating a luminance difference between the average luminance value for the reference image data and an average luminance value for the image data, and

adding the luminance difference to each pixel in the reference image data,

wherein the moving object detecting unit detects a moving object in the image data by

correlating pixel luminance values corresponding to the reference image data and pixel luminance values corresponding to the photographed image data, the pixel luminance values being from an area of the image difference, and

comparing a sequence of image difference values to determine a temporal continuity of the image difference.

2. The image processing apparatus according to claim 1, wherein the image correcting unit corrects a blurring of the image using the image data and the reference image data, the moving object detecting unit extracting an object area using the corrected image.

3. The image processing apparatus according to claim 2, wherein the image correcting unit detects the blurring between the image data and each of the plurality of reference image data, and performs a blur correction.

4. The image processing apparatus according to claim 2, wherein the image correcting unit detects the blurring between data corresponding to two temporally continuous reference images when detecting the blurring between the image data and the reference image data, sets a blur correction reference by accumulating the blurring, and detects the blurring by performing a blur detection of a minute image section in accordance with the blur correction reference.

5. The image processing apparatus according to claim 1, further comprising:

a background image storing unit configured to store a background image obtained by photographing a scene without the moving object;

a background difference processing unit configured to detect a difference between the image data and the background image data; and

a background difference unit configured to compare a moving object area detected by the moving object detecting unit with the background image data, and configured to erase a residual image generated when the moving object is detected by the image moving object detecting unit.

6. The image processing apparatus according to claim 1, wherein the temporal continuity of the image difference is determined as an amount of time during which the image difference exceeds a predetermined threshold, and the moving object detecting unit detects the moving object when the amount of time exceeds a predetermined time threshold.

7. The image processing apparatus of claim 1, wherein the image correcting unit corrects the luminance of the reference image data in response to entire frame luminance differences between the image data and reference image data.

8. A method of causing an image processing apparatus to detect moving objects in an image, comprising:

storing, at a reference image storing unit, a plurality of past image data, photographed by a video input unit, as reference image data;

detecting, at a moving object detecting unit, an image difference based on one of a difference of each pixel and a difference of each pixel in small image areas between image data photographed by the video input unit and the reference image data;

generating, at an image correcting unit, an average luminance value for the reference image data;

calculating, at the image correcting unit, a luminance difference between the average luminance value of the reference image data and an average luminance value corresponding to the image data;

adding, at the image correcting unit, the luminance difference to each pixel of the reference image data;

detecting, at the moving object detecting unit, a moving object in the image data by

correlating pixel luminance values corresponding to the reference image data and pixel luminance values corresponding to the photographed image data, the pixel luminance values being from an area of the image difference, and

comparing a sequence of image difference values to determine a temporal continuity of the image difference.

9. The method according to claim 8, further comprising:

determining the temporal continuity of the image difference as an amount of time during which the image difference exceeds a predetermined threshold; and

detecting a moving object when the amount of time exceeds a predetermined time threshold.