IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD OF IDENTIFYING OBJECT IN IMAGE

• US 20150146991A1
• Filed: 11/19/2014
• Published: 05/28/2015
• Est. Priority Date: 11/28/2013
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Nakano, Shunsuke, Sato, Hiroshi, Kaneda, Yuji, Suzuki, Takashi, Nomoto, Atsuo
• Granted Patent: US 9,633,284 B2
• Time in Patent Office: Days
• US Class Current: 382/219
• CPC Class Codes: G06V 30/242 Division of the character sequences into groups prior to recognition; Selection of dictionaries
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

The degree of similarity between corresponding local feature amounts out of a plurality of local feature amounts of the object in the input image and a plurality of local feature amounts of an object in an image registered in advance is obtained. At least one degree of similarity is selected out of the obtained degrees of similarity based on a capturing condition for the object in the input image and a capturing condition for the object in the registered image, and one degree of similarity is derived from the at least one selected degree of similarity. It is determined based on the one derived degree of similarity whether the object in the input image belongs to the same category as the object in the registered image.

Claims

1. An image processing apparatus for identifying an object in an input image, comprising:

a calculation unit configured to obtain a degree of similarity between corresponding local feature amounts out of a plurality of local feature amounts of the object in the input image and a plurality of local feature amounts of an object in an image registered in advance;

a derivation unit configured to select at least one degree of similarity out of the degrees of similarity obtained by said calculation unit based on a capturing condition for the object in the input image and a capturing condition for the object in the registered image and derive one degree of similarity from the at least one selected degree of similarity; and

a determination unit configured to determine based on the one degree of similarity derived by said derivation unit whether the object in the input image belongs to the same category as the object in the registered image.

2. The apparatus according to claim 1, wherein said derivation unit comprises:

a unit configured to obtain a difference between the capturing condition for the object in the input image and the capturing condition for the object in the registered image; and

a unit configured to select, from the degrees of similarity obtained by said calculation unit, the degrees of similarity in a number according to the difference from a higher rank in descending order of the degree of similarity and derive the one degree of similarity from the selected degrees of similarity.

3. The apparatus according to claim 2, wherein said derivation unit selects a smaller number of degrees of similarity from the higher rank in descending order of the degree of similarity as the difference becomes larger.

4. The apparatus according to claim 1, wherein said derivation unit comprises:

a unit configured to classify the degrees of similarity between the local feature amounts obtained by said calculation unit into groups preset based on the capturing condition for the object in the input image and the capturing condition for the object in the registered image with respect to positions of the respective local feature amounts in the input image and the registered image;

a unit configured to select, out of the degrees of similarity classified into each of the groups, the degrees of similarity in a number according to a ratio preset for the group from a higher rank in descending order of the degree of similarity; and

a unit configured to derive the one degree of similarity from the degrees of similarity selected in respective groups.

5. The apparatus according to claim 1, wherein said derivation unit comprises:

a unit configured to obtain a difference between the capturing condition for the object in the input image and the capturing condition for the object in the registered image;

a unit configured to classify the degrees of similarity between the local feature amounts obtained by said calculation unit into groups according to types of the local feature amounts;

a unit configured to select, from the degrees of similarity classified into each of the groups, the degrees of similarity in a number decided in accordance with the group and the difference from a higher rank in descending order of the degree of similarity; and

a unit configured to derive the one degree of similarity from the degrees of similarity selected in respective groups.

6. The apparatus according to claim 1, wherein said determination unit outputs information representing the category to which, out of objects in a plurality of registered images, the object having the one degree of similarity not less than a threshold and corresponding to a maximum degree of similarity belongs as an identification result of the object in the input image.

7. The apparatus according to claim 1, wherein said derivation unit obtains an average value of the at least one selected degree of similarity as the one degree of similarity.

9. A non-transitory computer-readable storage medium storing a computer program that causes a computer to function as each unit of an image processing apparatus of claim 1.

8. An image processing method performed by an image processing apparatus for identifying an object in an input image, comprising:

a calculation step of obtaining a degree of similarity between corresponding local feature amounts out of a plurality of local feature amounts of the object in the input image and a plurality of local feature amounts of an object in an image registered in advance;

a derivation step of selecting at least one degree of similarity out of the degrees of similarity obtained in the calculation step based on a capturing condition for the object in the input image and a capturing condition for the object in the registered image and deriving one degree of similarity from the at least one selected degree of similarity; and

a determination step of determining based on the one degree of similarity derived in the derivation step whether the object in the input image belongs to the same category as the object in the registered image.