Texture evaluation system

Patent metadata

Field	Value
Publication number	US-10535150-B2
Application number	US-201715618222-A
Country	US
Kind code	B2
Filing date	Jun 9, 2017
Priority date	Jun 14, 2016
Publication date	Jan 14, 2020
Grant date	Jan 14, 2020

Abstract

The present disclosure attempts to evaluate how the texture of an object is perceived based on visual features of the topological skeleton of the object. A camera S1 obtains a color image by taking an image of an object, which serves as an evaluation target. Within the image obtained, a visual feature area, which is likely to strike a person's eye when the person claps his/her eyes on the object, and an intensity of a visual stimulus of each pixel of the visual feature area are extracted. Visual skeleton features of each pixel of the image are determined within a contour region which is composed of the visual feature areas extracted. The visual skeleton features determined are shown on a display.

First claim

What is claimed is:

1. A texture evaluation system comprising:
an imager obtaining a color image by taking an image of an object which serves as an evaluation target;
a visual feature area extractor extracting, within the image obtained by the imager, a visual feature area including salient features, and an intensity of a visual stimulus of a pixel of the visual feature area;
a visual skeleton feature determiner determining visual skeleton features of the pixel of the image with regard to a contour region which is composed of a plurality of visual feature areas extracted by the visual feature area extractor; and
a display displaying the visual skeleton features determined by the visual skeleton feature determiner, wherein
the visual skeleton feature determiner sets, for each pixel of the image, a plurality of annular portions which are defined by a plurality of circles having different diameters and centering about each of the pixels, determines an intensity of a visual stimulus within each of the annular portions by calculating an intensity of a visual stimulus of each of the pixels included in each of the annular portions, and determines visual skeleton features for each of the pixels based on the intensity of the visual stimulus within each of the annular portions.

2. A texture evaluation system comprising:
an imager obtaining a color image by taking an image of an object which serves as an evaluation target;
a visual feature area extractor extracting, within the image obtained by the imager, a visual feature area including salient features, and an intensity of a visual stimulus of a pixel of the visual feature area;
a visual skeleton feature determiner determining visual skeleton features of the pixel of the image with regard to a contour region which is composed of a plurality of visual feature areas extracted by the visual feature area extractor; and
a display displaying the visual skeleton features determined by the visual skeleton feature determiner, wherein
kinds of the visual feature areas include at least intensity, chrominance, and orientation, and
the visual skeleton feature determiner sets, for each pixel of the image, a plurality of annular portions which are defined by a plurality of circles having different diameters and centering about each of the pixels, determines an intensity of a visual stimulus within each of the annular portions by calculating an intensity of a visual stimulus of each of the pixels included in each of the annular portions, and determines visual skeleton features for each of the pixels based on the intensity of the visual stimulus within each of the annular portions.

3. A texture evaluation system comprising:
an imager obtaining a color image by taking an image of an object which serves as an evaluation target;
a visual feature area extractor extracting, within the image obtained by the imager, a visual feature area including salient features, and an intensity of a visual stimulus of a pixel of the visual feature area;
a visual skeleton feature determiner determining visual skeleton features of the pixel of the image with regard to a contour region which is composed of a plurality of visual feature areas extracted by the visual feature area extractor; and
a display displaying the visual skeleton features determined by the visual skeleton feature determiner, wherein
kinds of the visual feature areas include at least intensity, chrominance, and orientation,
the visual feature area extractor breaks down the image into a plurality of images for each of the kinds of the visual feature areas, extracts visual feature areas from each of the images, and integrates the visual feature areas extracted from each of the images into one single image, and
the visual skeleton feature determiner sets, for each pixel of the image, a plurality of annular portions which are defined by a plurality of circles having different diameters and centering about each of the pixels, determines an intensity of a visual stimulus within each of the annular portions by calculating an intensity of a visual stimulus of each of the pixels included in each of the annular portions, and determines visual skeleton features for each of the pixels based on the intensity of the visual stimulus within each of the annular portions.

4. The system of claim 1 , wherein
the visual skeleton feature determiner finally determines the visual skeleton features of each of the pixels by a gradual addition of the intensity of the visual stimulus within an inner annular portion to the intensity of the visual stimulus within an outer annular portion, and
in the gradual addition, a first sum is defined as “Dmn K+1 ,” a previous sum as “Dmn K ,” an intensity of a visual stimulus within an annular portion added first as “Smn K ,” and a parameter as “v” (1>v>0), and then the first sum “Dmn K+1 ” is calculated based on Equation (2) where “mn” are coordinates of a center of the annular portion, and “ K ” is a suffix

Equation
⁢

⁢

(
2
)

Dmn

K
+
1

=

Smn
K

+

Dmn
K

1
+

vSmn
K

(
2
)

5. The system of claim 2 , wherein
the visual skeleton feature determiner finally determines the visual skeleton features of each of the pixels by a gradual addition of the intensity of the visual stimulus within an inner annular portion to the intensity of the visual stimulus within an outer annular portion, and
in the gradual addition, a first sum is defined as “Dmn K+1 ,” a previous sum as “Dmn K ,” an intensity of a visual stimulus within an annular portion added first as “Smn K ,” and a parameter as “v” (1>v>0), and then the first sum “Dmn K+1 ” is calculated based on Equation (2) where “mn” are coordinates of a center of the annular portion, and “ K ” is a suffix

Equation
⁢

⁢

(
2
)

Assignees

• Mazda Motor

Inventors

• Watanabe Masayuki
• Goda Naokazu
• Komatsu Hidehiko
• Michida Nanae
• Kishi Atsuhide
• Nishikawa Kazuo
• Nouzawa Takahide

Classifications

• G06T7/49 Primary

  based on structural texture description, e.g. using primitives or placement rules · CPC title

• G06V20/59 Primary

  inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions · CPC title

• G06T7/73

  using feature-based methods · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• G06T2207/20221

  Image fusion; Image merging · CPC title