Method for identifying objects, in particular three-dimensional objects

• US 10,491,880 B2
• Filed: 01/27/2017
• Issued: 11/26/2019
• Est. Priority Date: 01/29/2016
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Ketterer, Darno Alexander
• Primary Examiner(s): Hodges, Susan E.
• Application Number:
  US15/417,258
• Publication Number:
  US 20170237966A1
• Time in Patent Office: 1,033 Days
• Field of Search: 348 43
• CPC Class Codes: G06T 7/00 Image analysis
• CPC Class Codes: H04N 13/106 Processing image signals for multi-view video sequence encoding H04N19/597
• CPC Class Codes: H04N 13/204 using stereoscopic image cameras stereoscopic photography G03B35/00
• CPC Class Codes: H04N 13/257 Colour aspects

Abstract

Identification of objects, in particular three dimensional objects, includes digital scanning of at least one image portion of an image of at least one object to be identified for the digital capture of points of the image. The digitally captured points of the at least one image portion of the image are combined to form an image data record, wherein multiple image data records are generated for the object to be identified, and the image data records differ from one another at least in view of at least one image portion characteristic.

Claims

1. A method for identifying three-dimensional objects, the method comprising:

generating a plurality of image data records for an object to be identified, each image data record corresponding to a respective image portion of an image of an object to be identified, each respective image portion being different at least with respect to at least one image portion characteristic, the generating of each image data record comprising;

digitally scanning the respective image portion to digitally capture points of the respective image portion of the image, only a subset of all points in the respective image portion being digitally captured; and

combining the digitally captured subset of all points of the respective image portion of the image to form the image data record;

generating a superposition of the plurality of image data records; and

assigning a transparent color value to the digitally captured subset of all points which have a white color value in the superposition of the plurality of image data records.

2. The method according to claim 1, further comprising:

calculating, based on the superposition of the image data records, at least one new image portion of an image of the object to be identified.

3. The method according to claim 1, the digital scanning further comprising:

digitally capturing the points of the respective image portion at least dependent on a deviation of at least one characteristic of at least two of the individual points, the deviation being at least one of;

a deviation of at least one characteristic of the point captured most recently in relation to a characteristic of at least one other of the points to be captured digitally,

a deviation of at least one color value of the at least two of the points, and

a deviation of at least one brightness value of the at least two of the points.

4. The method according to claim 1, the digital scanning further comprising at least one of:

digitally scanning the image along an image direction of the respective image portion of the image, the image direction extending diagonally; and

digitally capturing the points of the image along the image direction of the respective image portion of the image, the image direction extending diagonally.

5. The method according to claim 1, wherein a grid is for at least one of (i) the digital scanning of the respective image portion of the image and (ii) the digital capture of the points of the respective image portion.

6. The method according to claim 5, wherein at least one of:

the method further comprises modifying at least one characteristic of the grid dependent on at least one characteristic of the respective image portion of the image; and

a pitch of the grid in the vicinity of a central point of the at least one image portion of the image is narrower than in an edge region of the respective image portion of the image.

7. The method according to claim 1, the digital scanning further comprising:

digitally capturing points with a smaller deviation from at least one characteristic in a vicinity of a central point of the respective image portion of the image than in an edge region of the respective image portion of the image.

8. The method according to claim 1, further comprising:

predetermining, by an operator, at least one of a dimension and an arrangement of the respective image portion of the image to be digitally scanned.

9. The method according to claim 1, further comprising:

comparing the digitally captured subset of all points of the image with data of images of objects which are stored in a memory unit, the comparing taking into account at least one of at least one global location characteristic and at least one weather characteristic.

10. The method according to claim 1, wherein an operator input with respect to a type of object to be identified is taken into account in at least one method step.

11. The method according to claim 1, wherein the digital scanning of the respective image portion of the image is performed using of a mobile apparatus which is connected to an external data processing apparatus and configured to transfer digital data to the external data processing apparatus.

12. The method according to claim 11, wherein at least one of an acceleration characteristic and an alignment characteristic of the mobile apparatus is taken into account in at least one method step.

13. The method according to claim 11, further comprising:

preprocessing the respective image portion of the image using a computing unit of the mobile apparatus.

14. The method according to claim 1, further comprising:

outputting instructions to an operator, the instructions at least comprising movement instructions for moving a mobile apparatus configured to capture digital points of the image relative to the object to be identified.

15. A system for identifying three-dimensional objects, the system comprising:

at least one mobile apparatus configured to generate a plurality of image data records for an object to be identified, each image data record corresponding to a respective image portion of an image of the object to be identified, each respective image portion being different at least with respect to at least one image portion characteristic, the at least one mobile apparatus being configured to;

generate each image data record by;

digitally scanning the respective image portion to digitally capture points of the respective image portion of the image, only a subset of all points in the respective image portion being digitally captured; and

combining the digitally captured subset of all points of the respective image portion of the image to form the image data record, the at least one mobile apparatus having at least one collective logic configured to digitally capture the points of the respective image portion;

generate a superposition of the plurality of image data records; and

assign a transparent color value to the digitally captured subset of all points which have a white color value in the superposition of the plurality of image data records; and

at least one data processing apparatus in communication with at least the mobile apparatus, the least one data processing apparatus being configured to identify the object by processing the digitally captured subset of all points of the superposition of the plurality of image data records.

16. The system according to claim 15, the mobile apparatus comprising:

at least one of at least one alignment sensor and an acceleration sensor configured to capture data that is processed by the data processing apparatus in order to identify the object.

17. The system according to claim 15, the mobile apparatus comprising:

a touch-sensitive input unit configured to receive a selection from an operator of the respective image portion for at least one image data record in the plurality of image data records, the input unit having a selection mask configured for selecting the respective image portion of the image, a size of the selection mask being modifiable using at least one finger of the operator.

18. The system according to claim 15, wherein the data processing apparatus is embodied as at least one server.