Image analysis – Image transformation or preprocessing – Selecting a portion of an image
Reexamination Certificate
2000-05-01
2004-06-29
Boudreau, Leo (Department: 2621)
Image analysis
Image transformation or preprocessing
Selecting a portion of an image
C382S199000, C358S538000
Reexamination Certificate
active
06757444
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an image extraction apparatus and method for extracting a target subject from a background image and a subject image. More particularly, the present invention is directed to a method and apparatus for appropriately generating a mask used for extracting a target subject.
2. Related Arts
Conventionally, as general techniques for realizing image extraction, a chromakey method using a specific color background, a videomatte method for generating a key signal by performing a histogram process, difference (or differential) process, contour enhancement or contour tracking process of an image signal (
The Television Society Technical Report
, Vol. 12, pp. 29-34, 1988), and the like are known.
A technique for performing image extraction based on the difference from the background image is a state-of-the-art one, and for example, Japanese Patent Laid-Open No. 4-216181 discloses a technique for detecting or extracting a target object in a plurality of specific regions in an image by setting a mask image (i.e., a specific processing region) in difference data between the background image and the image to be processed.
Furthermore, Japanese Patent Publication No. 7-16250 discloses a technique for obtaining color-converted data of an original image including a background using a color model of the object to be extracted, and the existence probability distribution of the object to be extracted from brightness difference data between the background image and the original image.
In the difference method from the background image, the luminance level or color component difference between the pixels of the background image and the subject image is normally expressed by a predetermined evaluation function, and the evaluation function is subjected to a thresholding process to extract a region having a difference level equal to or higher than an initial value. As the evaluation function, the correlation between blocks having individual points as centers and a predetermined size (Rosenfeld, A. and Kak, A. C., Digital Picture Processing (2nd ed.), Academic Press, 1982), normalized principal component features (
Journal of the Institute of Electronics, Information and Communication Engineers
, Vol. J74-D-II, pp. 1731-1740), a weighted sum value of a standard deviation and a difference value (
Journal of the Television Society
, Vol. 45, pp. 1270-1276, 1991), a local histogram distance associated with hue and luminance level (
Journal of the Television Society
, Vol. 49, pp. 673-680, 1995), and the like are used.
Japanese Patent Laid-Open No. 4-328689 and Japanese Patent Publication No. 7-31248 disclose a method of extracting a moving object alone by extracting motion vectors or inter-frame difference data from moving images. Japanese Patent Publication Nos. 7-66446, 6-14358, and 4-48030 disclose a method of extracting a moving object based on the difference from the background image. Furthermore, a method of extracting the binocular disparity distribution (i.e., the distance distribution from image sensing means) from images from right and left different view point positions obtained using a binocular image sensing system, and segmenting an object from the background on the basis of the disparity distribution (1995
Information System Society Meeting of the Society of Electronics, Information and Communication Engineers
, pp. 138), or the like is known.
However, of the above-mentioned prior arts, the chromakey method suffers from the following problems:
i: this method cannot be used outdoors due to serious background limitations, and
ii: color omission occurs.
Also, the videomatte method suffers from the following problems:
i: the contour designation must be manually and accurately performed in units of pixels, and
ii: such operation requires much labor and skill.
Furthermore, the difference method from the background image is normally difficult to realize due to the following problems:
i: the background is hard to distinguish from the subject in a partial region of the subject including a portion similar to the background,
ii: the difference method is readily influenced by variations in image sensing condition between the background image and subject image,
iii: a shadow portion formed by the subject is hard to remove, and
iv: in order to faithfully extract the boundary line between the background and subject, the background image and subject image must have considerably different image characteristics (pixel values and the like) in the vicinity of the boundary therebetween.
The technique disclosed in Japanese Patent Publication No. 7-16250 is not suitable for image extraction of an arbitrary unknown object since it requires a color model for the object to be extracted.
In either the method of extracting a moving object from moving images or the method of extracting a subject from the disparity distribution, it is generally hard to extract a subject with high precision independently of the contrast in the boundary portion between the subject and background.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide an image extraction apparatus and method, which can stably extract a subject image in which the background and subject have no distinct difference between their image characteristics.
It is another object of the present invention to provide an image extraction apparatus and method which can obtain a large area of a subject region before region growing by a small number of processing steps, and can extract details of a contour shape.
It is still another object of the present invention to provide an image extraction apparatus and method which can execute a process for equalizing the contour line of a mask after region growing with that of an actual subject without being influenced by the background pattern near the contour line of the subject.
It is still another object of the present invention to provide an image extraction apparatus and method which can stably grow an initial mask only in a subject region independently of variations in the region growing condition, i.e., the tolerance value of a feature difference from a neighboring region.
It is still another object of the present invention to provide an image extraction apparatus and method which can suppress variations in edge intensity distribution caused by a difference in image sensing conditions between the background image and the subject image, noise, or the like, and can accurately extract the contour shape of the subject and the edge of a background portion present in the subject region.
It is still another object of the present invention to provide an image extraction apparatus and method which can stably extract a subject image even when the edge intensity serving as a boundary between the subject and background is small and the subject includes a relatively thin shape.
It is still another object of the present invention to provide an image extraction apparatus and method which can stably extract the contour shape of a subject without being influenced by the edge distribution of a background portion present in the vicinity of the subject.
It is still another object of the present invention to provide an image extraction apparatus and method which can automatically retrieve an incomplete partial shape after region growing on the basis of the condition of shape continuity, and can smooth shape data.
It is still another object of the present invention to provide an image extraction apparatus and method which can stably extract a subject image independently of any specific difference between the image characteristics of the background and subject without being influenced by the background pattern.
It is still another object of the present invention to provide an image extraction apparatus and method which can stably and accurately extract a subject image upon executing extraction based on region growing.
It is still another object of the present invention to provide an image extraction apparatus and method which can obtain an
Hatanaka Koji
Matsugu Masakazu
Takiguchi Hideo
Boudreau Leo
Sherali Ishrat I
LandOfFree
Image extraction apparatus and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image extraction apparatus and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image extraction apparatus and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3363544