SYSTEMS AND METHODS FOR EVALUATING IMAGES

• US 20130022261A1
• Filed: 07/22/2011
• Published: 01/24/2013
• Est. Priority Date: 07/22/2011
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Imai, Francisco, Lin, Andy Lai
• Granted Patent: US 8,670,609 B2
• Time in Patent Office: Days
• US Class Current: 382/154
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

Systems and methods for evaluating images segment a computational image into sub-images based on spectral information in the computational image, generate respective morphological signatures for the sub-images, generate respective spectral signatures for the sub-images, and generate a resulting image signature based on the morphological signatures and the spectral signatures.

Claims

1. A method for evaluating an image, the method comprising:

segmenting a computational image into sub-images based on spectral information in the computational image;

generating respective morphological signatures for the sub-images;

generating respective spectral signatures for the sub-images; and

generating an image signature based on the morphological signatures and the spectral signatures.

2. The method of claim 1, wherein segmenting the computational image into sub-images includes

identifying one or more spectral properties in the computational image corresponding to one or more spectral properties of one or more materials based on the spectral information in the computational image, and

segmenting the computational image into sub-images based on the spectral properties.

3. The method of claim 2, wherein the one or more spectral properties include spectral reflectance.

4. The method of claim 1, further comprising:

searching for one or more images similar to the computational image based on the image signature; and

if one or more similar images are not located by the search,

segmenting the computational image into different sub-images,

generating respective morphological signatures for the different sub-images,

generating respective spectral signatures for the different sub-images, and

generating a different image signature based on the morphological signatures and the spectral signatures of the different sub-images.

5. The method of claim 4, further comprising searching for one or more images similar to the computational image based on the different resulting image signature.

6. The method of claim 1, wherein the morphological signatures are generated using one or more of SIFT and CHoG.

7. The method of claim 1, wherein segmenting the computational image into sub-images is based at least in part on depth information in the computational image.

8. The method of claim 7, wherein generating the respective morphological signatures for the sub-images includes generating a respective three-dimensional histogram of gradients for the sub-images based at least in part on the depth information in the computational image.

9. A system for evaluating an image, the system comprising

a computer readable medium;

a network interface configured to send data to and receive data from one or more other devices or systems; and

one or more processors configured to cause one or more computing devices to

generate sub-images from a computational image based on spectral information in the computational image,

generate respective morphological signatures for the sub-images,

generate respective spectral signatures for the sub-images, and

generate a combined image signature based on the morphological signatures and the spectral signatures.

10. The system of claim 9, wherein the computational image includes spectral information that indicates light not visible to a human eye, and wherein the respective spectral signatures for the sub-images are generated based at least in part on the light not visible to the human eye.

11. The system of claim 9, wherein the computational image includes depth information, and wherein the respective morphological signatures are generated based at least in part on the depth information.

12. The system of claim 11, wherein the depth information is detected by multi-aperture optics in an image capturing device that captured the computational image.

13. The system of claim 9, wherein generating the sub-images includes identifying one or more objects in the computational image;

and generating respective sub-images corresponding to the one or more objects.

14. The system of claim 9, wherein the one or more processors are further configured to cause the one or more computing devices to transmit a search request to one or more search devices, wherein the search request includes the combined image signature.

15. One or more computer-readable media storing instructions that, when executed by one or more computing devices, cause the one or more computing devices to perform operations comprising:

partitioning a computational image into sub-images based at least on spectral information in the computational image;

generating respective spectral signatures for the sub-images; and

generating respective morphological signatures for the sub-images.

16. The computer-readable media of claim 15, wherein the operations further comprise:

searching a plurality of images for spectrally similar images to the computational images based on the spectral signatures; and

searching the spectrally similar images for images with morphology similar to the computational image based on the morphological signatures.

17. The computer-readable media of claim 16, wherein the operations further comprise:

if no images with morphology similar to the computational image are located by searching, generating different respective spectral signatures for the sub-images and generating different respective morphological signatures for the sub-images.

18. The computer-readable media of claim 15, wherein the computational image includes depth information, and wherein partitioning the computational into sub-images is further based at least in part on the depth information.

19. The computer-readable media of claim 15, wherein partitioning the computational image into sub-images based at least on spectral information includes estimating a material in the computational image based on the spectral information, identifying an object based on the material, and partitioning the computational image into a sub-image of the object.