Image analysis – Applications – Surface texture or roughness measuring
Reexamination Certificate
1999-11-30
2004-06-15
Boudreau, Leo (Department: 2621)
Image analysis
Applications
Surface texture or roughness measuring
C345S582000
Reexamination Certificate
active
06751339
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of image processing, and in particular to a method of selecting texture and/or providing texture advice, which can be used for example, by graphics designers, or by graphic design applications.
BACKGROUND OF THE INVENTION
The aspect of texture combination is a necessary feature in any graphic design. For example, selection of a background texture which complements the texture in a picture, which in turn is part of an overall graphic design, is one example of where a graphic artist is required to select an aesthetically pleasing texture combination. Traditionally, texture selection is performed manually by the graphic artist, based upon skill and experience, and also upon subjective assessment criteria. Clearly although this method will, providing the graphic artist is suitably experienced, yield aesthetically pleasing results, the subjective nature of the method does not provide a basis for consistent or repeatable outcomes, or the possibility of automation.
SUMMARY OF THE INVENTION
In accordance with a first aspect of the present invention there is disclosed a method for determining a texture which substantially best matches at least one image from a set of images, the method comprising the steps of:
(a) determining an overall image feature value from the at least one image;
(b) determining, for each member of a set of textures, a texture feature value corresponding to the overall image feature value, so to form a set of corresponding texture feature values;
(c) comparing the overall image feature value with each member of the set of corresponding texture feature values, so to produce a corresponding set of comparison values; and
(d) selecting, from the set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture which substantially best matches the at least one image.
In accordance with a second aspect of the present invention there is disclosed a method for determining a texture which substantially best matches a set of images, the method comprising the steps of:
(a) clustering the set of images into groups dependent upon an image feature classification,
(b) determining a group image feature value for each group;
(c) determining, for each member of a set of textures, a texture feature value corresponding to each group image feature value, so to form a set of corresponding texture feature values;
(d) comparing, for each group of images, the group image feature value with each member of the set of corresponding texture feature values, so to produce, for each group image feature value, a corresponding set of comparison values;
(e) selecting, for each group of images, from the corresponding set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture which substantially best matches the associated group image feature value; and
(f) selecting, from the corresponding textures each of which substantially best matches the group image feature value associated with a corresponding group of images, a best texture according to a second rule, said best texture substantially best matching the set of images.
In accordance with a third aspect of the present invention there is disclosed an apparatus for determining a texture which substantially best matches at least one image from a set of images, the apparatus comprising:
(a) first determination means for determining an overall image feature value from the at least one image;
(b) second determination means for determining, for each member of a set of textures, a texture feature value corresponding to the overall image feature value, so to form a set of corresponding texture feature values;
(c) comparison means for comparing the overall image feature value with each member of the set of corresponding texture feature values, so to produce a corresponding set of comparison values; and
(d) selection means for selecting, from the set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture which substantially best matches the at least one image.
In accordance with a fourth aspect of the present invention there is disclosed an apparatus for determining a texture which substantially best matches a set of images, the apparatus comprising:
(a) clustering means for clustering the set of images into groups dependent upon an image feature classification,
(b) first determination means for determining a group image feature value for each group;
(c) second determination means for determining, for each member of a set of textures, a texture feature value corresponding to each group image feature value, so to form a set of corresponding texture feature values;
(d) comparison means for comparing, for each group of images, the group image feature value with each member of the set of corresponding texture feature values, so to produce, for each group image feature value, a corresponding set of comparison values;
(e) first selection means for selecting, for each group of images, from the corresponding set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture which substantially best matches the associated group image feature value; and
(f) second selection means for selecting, from the corresponding textures each of which substantially best matches the group image feature value associated with a corresponding group of images, a best texture according to a second rule, said best texture substantially best matching the set of images.
In accordance with a fifth aspect of the present invention there is disclosed a computer program product including a computer readable medium having recorded thereon a computer program determining a texture which substantially best matches at least one image from a set of images, the computer program comprising:
(a) first determination steps for determining an overall image feature value from the at least one image;
(b) second determination steps for determining, for each member of a set of textures, a texture feature value corresponding to the overall image feature value, so to form a set of corresponding texture feature values;
(c) comparison steps for comparing the overall image feature value with each member of the set of corresponding texture feature values, so to produce a corresponding set of comparison values; and
(d) selection steps for selecting, from the set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture which substantially best matches the at least one image.
In accordance with a sixth aspect of the present invention there is disclosed a computer program product including a computer readable medium having recorded thereon a computer program for determining a texture which substantially best matches a set of images, the computer program comprising:
(a) clustering steps for clustering the set of images into groups dependent upon an image feature classification,
(b) first determination steps for determining a group image feature value for each group;
(c) second determination steps for determining, for each member of a set of textures, a texture feature value corresponding to each group image feature value, so to form a set of corresponding texture feature values;
(d) comparison steps for comparing, for each group of images, the group image feature value with each member of the set of corresponding texture feature values, so to produce, for each group image feature value, a corresponding set of comparison values;
(e) first selection steps for selecting, for each group of images, from the corresponding set of comparison values, a best comparison value according to a first rule, whereby the selected comparison value is associated with the corresponding texture wh
Windle John Richard
Wu Jing
Boudreau Leo
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
Miller Ryan J.
LandOfFree
Texture advisor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Texture advisor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Texture advisor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3362424