Optical self-diagnosis of a stereoscopic camera system

• US 8,854,431 B2
• Filed: 12/17/2012
• Issued: 10/07/2014
• Est. Priority Date: 06/18/2010
• Status: Active Grant

• Current Assignee: Xovis Germany GmbH
• Original Assignee: Hella KGaA Hueck & Company
• Inventors: Song, Miao, Lojewski, Bjorn
• Primary Examiner(s): Ustaris, Joseph
• Assistant Examiner(s): BAILEY, FREDERICK D
• Application Number:
  US13/716,430
• Publication Number:
  US 20130120536A1
• Time in Patent Office: 659 Days
• Field of Search: None
• US Class Current: 348/46
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: H04N 13/00 Stereoscopic video systems; Multi-view video systems; Details thereof
• CPC Class Codes: H04N 13/189 Recording image signals; Reproducing recorded image signals
• CPC Class Codes: H04N 13/246 Calibration of cameras
• CPC Class Codes: H04N 13/296 Synchronisation thereof; Control thereof
• CPC Class Codes: H04N 17/002 for television cameras

Abstract

The present invention relates to a method for the optical self-diagnosis of a camera system and to a camera system for carrying out the method. The method comprises recording stereo images obtained from in each case at least two partial images (2, 3) creating a depth image, that is to say a disparity map (5) given by calculated disparity values, determining a number of valid disparity values (6) of the disparity map (5), and outputting a warning signal depending on the number of valid disparity values determined. A device for carrying out such a method comprises a stereo camera (1) having at least two lenses (7,8) and image sensors, an evaluation unit and a display unit.

Claims

1. A method for optical self-diagnosis of a camera system adapted to record a stereo image obtained from at least two partial images, the method comprising:

a) recording a first partial image and a second partial image of an object space with the camera system from different perspectives;

b) producing a depth image from said first partial image and said second partial image on the basis of parallax between mutually corresponding pixels recorded in said first and second partial images, and determining disparity values representing said parallax;

c) determining a number of valid disparity values of the depth image, wherein a valid disparity value is defined as a disparity value which is determined for two pixels, recorded in the first and second partial images, which are identified as being mutually corresponding either unequivocally or at least with a defined probability;

d) producing a warning signal depending on the number of said valid disparity values, wherein a comparison of the number of valid disparity values is performed with a predetermined threshold value, and the warning signal is produced when said number of disparity values falls below the predetermined threshold value, wherein several disparity maps are calculated and compared with each other over a period of time, with the warning signal being produced only when a number of valid disparity values in the calculated disparity maps falls below the predetermined threshold value of valid disparity values within said period of time.

2. A method according to claim 1, wherein said depth image includes a disparity map, which is produced by carrying out a matching method in which said disparity values are determined as a measure for the parallax between the corresponding pixels of the first and second partial images by allocating a pixel recording the first partial image to a corresponding pixel recording the second partial image.

3. A method according to claim 1, further comprising calibrating, in an installation phase, parameters used for optical self-diagnosis of the camera system.

4. A method according to claim 1, wherein said calibrating in an installation phase includes calibrating parameters in such a way that an average number of valid disparity values is determined on the basis of a test object, with a threshold value being calculated by means of the number of valid disparity values determined in this manner.

5. A method according to claim 4, wherein said threshold value is calculated by multiplying said average number of valid disparity values, determined in the installation phase, with a factor which is larger than 0.

6. A method according to claim 1, wherein said recording of the first and second partial images of the object space from different prospective is effectuated simultaneously.

7. A method according to claim 2, wherein said matching method comprises a Fourier transformation.

8. A camera system configured to perform a method according to claim 1, comprising:

a stereo camera with at least two lenses and image sensors for recording the at least two partial images,

an evaluation electronic circuit configured to produce a depth image and calculate the disparity values, and

an output display unit adapted to produce a warning signal.

9. A camera system according to claim 8, wherein said camera system includes one of a person-counting system and a monitoring system with automatic image evaluation.