Image recognition apparatus

• US 9,818,038 B2
• Filed: 01/06/2016
• Issued: 11/14/2017
• Est. Priority Date: 01/06/2016
• Status: Active Grant

• Current Assignee: Toshiba Tec Corporation (Toshiba Corporation)
• Original Assignee: Toshiba Tec Corporation (Toshiba Corporation)
• Inventors: Yasunaga, Masaaki
• Primary Examiner(s): Tran, Phuoc
• Application Number:
  US14/989,200
• Publication Number:
  US 20170193319A1
• Time in Patent Office: 678 Days
• CPC Class Codes: G06K 7/10722 Photodetector array or CCD scanning
• CPC Class Codes: G06Q 20/20 Point-of-sale [POS] network systems
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 30/2247 Characters composed of bars, e.g. CMC-7
• CPC Class Codes: G07G 1/0045 with a code reader for reading of an identifying code of the article to be registered, e.g. barcode reader or radio-frequency identity [RFID] reader

Abstract

An image recognition apparatus includes a memory, an interface and a processor. The memory stores identification information obtained from code information attached to the objects and image information on the objects used for object recognition, the storage storing the identification information and the image information for each of objects to be recognized. The processor which controls to: extract a target object region including an object therein from a photographed image; extract code information in the target object region and recognize the identification information based on extracted code information; and recognize the object based on an image of the target object region and image information on each object, if the processor fails to recognize the object based on the code information.

Claims

1. An image recognition apparatus which recognizes objects, comprising:

a memory which stores image information for object recognition for each of a plurality of objects to be recognized;

an interface which acquires a photographed image; and

a processor which performs processing to;

extract a target object region including at least one object therein from the acquired photographed image; and

execute recognition processing a number of times on the target object region until each object in the target object region is recognized, wherein the recognition processing includes;

extracting a local feature for object recognition from the target object region,

recognizing an object based on the local feature compared to the stored image information,

deleting, from the target object region, the local feature related to the recognized object, and

repeating the recognition processing to recognize other objects based on remaining local features in the target object region.

2. The image recognition apparatus according to claim 1, further comprising:

a camera which photographs a placement surface on which the objects are placed,

wherein the interface acquires the image photographed by the camera.

3. The image recognition apparatus according to claim 1, wherein

the memory stores, for each of the plurality of objects to be identified, a feature point and local feature amount as the image information for object recognition, and

the recognition processing further includes:

extracting a feature point and a local feature amount from the target object region as the local feature,

recognizing at least one object based on the feature point and the local feature amount extracted from the target object region compared to the feature point and local feature amount stored in the memory for each object, and

deleting, from the target object region, the feature point and the local feature amount related to the recognized at least one object.

4. The image recognition apparatus according to claim 3, wherein

if at least one object is recognized, the processor executes subsequent recognition processing such that an object is recognized based on a feature point and a local feature amount which are other than the feature amount and local feature amount based on which the at least one object is recognized.

5. The image recognition apparatus according to claim 1, wherein

the memory stores area information on the objects for each of objects to be recognized, and

the processor acquires an area of a recognized object from the memory and ends recognition processing for the target object region if a total area of all objects recognized in the target area is equal to an area of entirety of the target object region.

6. The image recognition apparatus according to claim 5, wherein

the processor determines that the recognition processing for the target object region ends in failure if the area of the entirety of the target object region is less than the total area of all objects recognized in the target area.