Object discrimination device, object discrimination method, and program

An object recognition and descriptor technology, applied in the field of object devices, can solve problems such as image error recognition

Active Publication Date: 2015-04-29
NEC CORP
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

With such a recognition method, when the object shown in the input image and the object shown in the reference image are different from each other but the difference between the two objects is subtle, there is an error in the image due to the existence of a large number of corresponding feature points. identified as a question showing the same object

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Object discrimination device, object discrimination method, and program
  • Object discrimination device, object discrimination method, and program
  • Object discrimination device, object discrimination method, and program

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0060] A first embodiment of the present invention will be described with reference to the drawings.

[0061] figure 1 is a block diagram showing the configuration of the object recognition device according to the first embodiment. The object recognition device includes: a local descriptor extraction unit 11 , a local descriptor matching unit 12 , an input image difference region determination unit 13 , an input image difference region descriptor extraction unit 14 , and a descriptor matching unit 15 . For example, the object recognition device can be configured using an information processing device such as a personal computer or a mobile information terminal. Additionally, the functions of the respective units constituting the object recognition device are realized by causing, for example, a processor to expand a program stored in the storage area onto the memory and execute the program. In addition, components of other embodiments described later can be implemented in a s...

no. 2 example

[0096] A second embodiment of the present invention will be described with reference to the drawings. In the second embodiment, the difference region in the reference image is estimated and recognition is performed without registering the difference region in the reference image in the database in advance.

[0097] Figure 12 is a block diagram showing the configuration of an object recognition device according to a second embodiment of the present invention. like Figure 12 As shown in , the object recognition device according to this embodiment includes: a local descriptor extraction unit 11, a local descriptor matching unit 16, an input image difference region determination unit 13, an input image difference region descriptor extraction unit 14, a difference region estimation unit 17. A difference region descriptor extracting unit 18, and a descriptor matching unit 15. As shown, the object recognition apparatus according to the second embodiment differs from the object r...

no. 3 example

[0105] A third embodiment of the present invention will be described with reference to the drawings.

[0106] Figure 16 is a block diagram showing the configuration of an object recognition device according to a third embodiment of the present invention. like Figure 16 As shown in , the object recognition device according to the third embodiment includes: a local descriptor extraction unit 11, a local descriptor matching unit 16, an input image difference region determination unit 13, an input image difference region descriptor extraction unit 14, a descriptor matching unit 15 , difference region estimation unit 19 , and difference region descriptor extraction unit 18 .

[0107] As described above, the object recognition device according to the third embodiment differs from the object recognition device according to the second embodiment in that the difference region estimation unit 17 of the object recognition device has been changed to the difference region estimation un...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The purpose of the invention is to more accurately determine whether an input image contains the same object as an object shown in a reference image. On the basis of geometrical transformation information calculated in an analysis that uses a local feature amount, a difference area in the reference image is transformed to determine a difference area in the input image. A feature amount extracted from the difference area in the input image is compared with a feature amount extracted from the difference area in the reference image. Thereby, a small difference that cannot be recognized by conventional comparison using only local feature amounts can be recognized and only an image showing the same object can be be discriminated.

Description

Background technique [0001] The present invention relates to devices, methods, and programs for accurately recognizing objects in images. [0002] In order to achieve the function of robustly recognizing the subject in the image with respect to changes in photographic size and angle, and with regard to occlusion, a description of detecting a large number of feature points (feature points) in an image and extracting a local area around each feature point is proposed A system of descriptors (local descriptors). As representative systems among them, Patent Document 1 and Non-Patent Document 1 disclose a local descriptor extraction device using a SIFT (Scale Invariant Feature Transform) descriptor. [0003] Traditionally, using a local descriptor extraction device, first uniquely extract brightness-related information from each pixel in the image, detect a large number of feature points (feature points) from the extracted brightness information, and output as The feature point i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & AuthorityApplications(China)
IPC IPC(8): G06T7/00
CPCG06V10/759G06V10/751G06V10/462G06V10/757G06T5/80G06T3/00
Inventor间濑亮太
OwnerNEC CORP