Image processing method, image processing apparatus, and image processing program

• US 20050152619A1
• Filed: 02/23/2005
• Published: 07/14/2005
• Est. Priority Date: 01/09/2004
• Status: Active Grant

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Motomura, Hideto, Kanamori, Katsuhiro, Sugaya, Takashi, Nakaguchi, Toshiya, Miyake, Yoichi, Tsumura, Norimichi
• Granted Patent: US 7,203,381 B2
• Time in Patent Office: Days
• US Class Current: 382/299
• CPC Class Codes: G06T 3/40 Scaling of whole images or parts thereof, e.g. expanding or contracting

Abstract

An image feature analysis section performs image feature analysis of an input image to generate an image feature quantity independent from image coordinates. An image content densification section information densifies the image feature quantity to generate a densified image feature quantity. An image generation section generates, based on the densified image feature quantity, an image in which the information content of the input image is densified.

Claims

1. An image processing method comprising:

a first step of performing image feature analysis for an original image to obtain an image feature quantity independent from image coordinates;

a second step of subjecting the image feature quantity obtained in the first step to information densification to obtain a densified image feature quantity; and

a third step of generating, based on the densified image feature quantity obtained in the second step, a densified image in which an information content of the original image is densified.

2. The image processing method of claim 1, wherein the second step includes the steps of:

selecting an image feature category to which the image feature quantity belongs from a plurality of image feature categories prepared in advance; and

reading out a basic image feature quantity, of which information content is densified, in the selected image feature category as the densified image feature quantity from a densified image feature database.

3. The image processing method of claim 1, wherein the second step includes the steps of:

calculating respective similarities of the image feature quantity to plural image feature categories prepared in advance; and

summing respective basic image feature quantities in the plural image feature categories with weights according to the respective calculated similarities to generate the densified image feature quantity.

4. The image processing method of claim 1, wherein the second step includes the steps of:

selecting an image feature category to which the image feature quantity belongs from a plurality of image feature categories prepared in advance; and

converting the image feature quantity to the densified image feature quantity by referencing a conversion table database and using a conversion table for feature quantity conversion in the selected image feature category.

5. The image processing method of claim 4, wherein

the plurality of image feature categories are provided for respective material properties of an object shot in an image.

6. The image processing method of claim 1, wherein

spatial resolution or time resolution is used as the image feature quantity.

7. The image processing method of claim 6, wherein

spatial frequency response or time frequency response is obtained using Fourier transform.

8. The image processing method of claim 6, wherein

spatial frequency response or time frequency response is obtained using wavelet transform.

9. The image processing method of claim 6, wherein

spatial frequency response or time frequency response is obtained using a plurality of spatial filters of which at least one of scale, phase, and spatial directionality is different from each other.

10. An image processing apparatus comprising:

an image feature analysis section that performs image feature analysis for an original image to obtain an image feature quantity independent from image coordinates;

an information densification section that performs information densification for the image feature quantity obtained by the image feature analysis section to obtain a densified image feature quantity; and

an image generation section that generates, based on the densified image feature quantity obtained by the information densification section, a densified image in which an information content of the original image is densified.

11. An image processing program that allows a computer to execute:

a first step of performing image feature analysis for an original image to obtain an image feature quantity independent from image coordinates;

a second step of subjecting the image feature quantity obtained in the first step to information densification to obtain a densified image feature quantity; and

a third step of generating, based on the densified image feature quantity obtained in the second step, a densified image in which an information content of the original image is densified.