Searching images and identifying images with similar facial features
US-9235782-B1 · Jan 12, 2016 · US
US2016012279A1 · US · A1
| Field | Value |
|---|---|
| Publication number | US-2016012279-A1 |
| Application number | US-201414770528-A |
| Country | US |
| Kind code | A1 |
| Filing date | Feb 14, 2014 |
| Priority date | Feb 28, 2013 |
| Publication date | Jan 14, 2016 |
| Grant date | — |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
The disclosure relates to a device for capturing person-specific data of a person, wherein the person-specific data comprises a facial image of the person, wherein the device has a camera for recording the facial image of the person, a lighting apparatus, and a semi-transparent mirror, wherein the semi-transparent mirror is arranged between the person and the camera, wherein the semi-transparent mirror is oriented such that, on the side of the semi-transparent mirror facing towards the person, the optical path of light incident on the semi-transparent mirror is parallel to the optical path of the portion of this light reflected back by the semi-transparent mirror, wherein the lighting apparatus is used to illuminate the person from the front, wherein the device also has a control unit for capturing an image, by generating light by means of the lighting apparatus in order to illuminate the face of the person and, during the illumination, capturing a first image of the face of the person by means of the camera, capturing a second image of the face of the person by means of the camera without generation of the light, forming a first differential image of the first image and the second image, wherein in the first differential image the face of the person is freed from the background, wherein the person-specific data comprises the first differential image as the facial image.
Opening claim text (preview).
1 . A device for capturing person-specific data of a person, wherein the person-specific data comprises a facial image of the person, wherein the device has a camera for recording the facial image of the person, a lighting apparatus, and a semi-transparent mirror, wherein the semi-transparent mirror is arranged between the person and the camera, wherein the semi-transparent mirror is oriented such that, on the side of the semi-transparent mirror facing towards the person, the optical path of light incident on the semi-transparent mirror is parallel to the optical path of the portion of this light reflected back by the semi-transparent mirror, wherein the lighting apparatus is used to illuminate the person from the front, wherein the device also has a control unit for capturing an image, by: generating light by means of the lighting apparatus in order to illuminate the face of the person and, during the illumination, capturing a first image of the face of the person by means of the camera; capturing a second image of the face of the person by means of the camera without generation of the light; and forming a first differential image of the first image and the second image, wherein in the first differential image the face of the person is freed from the background, wherein the person-specific data comprises the first differential image as the facial image. 2 . The device ( 100 ) according to claim 1 , wherein the generated light is white light, wherein the control unit is also configured to: control the lighting apparatus for generation of coloured or infrared light by the lighting apparatus for illuminating the face of the person and to control the camera in order to capture a third image of the face of the person by the camera during the illumination; capture a fourth image of the face of the person by means of the camera without the generation of light; form a second differential image of the third image and the fourth image, wherein in the second differential image the face of the person is freed from the background, and to calculate the outer contours of the face from the second differential image; and correct the first differential image with the outer contours calculated from the second differential image. 3 . The device according to claim 1 , wherein brightness differences of the pixels used to form the differential image are used to form the differential image, wherein all pixels of which the brightness difference lies below a predetermined threshold value are associated with the background. 4 . The device according to claim 1 , wherein the camera has a lens, wherein the lens is configured to capture exclusively the face of the person in focus and to capture the background out of focus, wherein the control unit is also configured to calculate the outer contours of the face from the sharpness differences of the face of the person and of the background ( 330 ); and correct the first differential image with the calculated outer contours. 5 . The device according to claim 1 , wherein the lighting apparatus is configured to illuminate the person with polarised light, wherein the camera has a polarisation filter, wherein the polarisation direction of the polarised light and the polarisation direction of the light transmitted through the polarisation filter into the camera are rotated by 90 degrees relative to one another. 6 . The device according to claim 1 , wherein the control unit is also configured for the analysis of the facial image of the person, wherein the analysis comprises a determination of the head posture and/or the eye posture, wherein the control unit is also configured, in the event that the head posture and/or the eye posture does/do not meet predefined criteria, to automatically reject the facial image of the person. 7 . The device according to claim 1 , wherein the camera is configured to capture a plurality of images successively in a video stream, wherein the control unit is also configured, when forming the first differential image and/or the second differential image, to use only the images of the video stream that meet predefined quality criteria. 8 . The device according to claim 7 , wherein the quality criteria include a predefined spatial position of the eye area of the person and/or a maximally admissible movement blur and/or a predefined size of the face of the person in the images. 9 . The device according to claim 1 , wherein the camera is height-adjustable by a drive unit, wherein the control unit ( 320 ) is also configured, to capture the spatial position of the eye area of the person relative to the position of the camera; and in the event that the spatial position of the eye area is not located in a predefined region relative to the position of the camera, to actuate the drive unit in order to automatically move the height of the camera via the drive unit, such that the spatial position of the eye area is located in the predefined region relative to the position of the camera. 10 . (canceled) 11 . The device according to claim 1 , wherein the camera is configured to capture infrared light, wherein the control unit ( 320 ) is also configured to capture a fifth image of the face of the person by the camera without illumination of the face of the person by means of the lighting apparatus, wherein the fifth image is captured in the infrared wavelength range; calculate the outer contours of the face from the fifth image; and correct the first differential image with the outer contours calculated from the fifth image. 12 . The device according to claim 1 , wherein the device has a reader for an identification document of the person, wherein the reader is configured, in the activated state, to read out personalised information of the person from the identification document, wherein the person-specific data comprises the personalised information. 13 . The device according to claim 1 , further comprising a fingerprint sensor, wherein the fingerprint sensor is configured, in the activated state, to capture a fingerprint of the person, wherein the person-specific data comprises the captured fingerprint. 14 . The device according to claim 1 , further comprising a control panel, wherein the control panel is configured, in the activated state, to capture an alphanumeric data input of the person and/or an input of a signature of the person, wherein the person-specific data comprises the captured data input and/or signature. 15 . The device according to claim 12 , wherein the control unit is also configured to set the reader and/or the fingerprint sensor and/or the control panel exclusively into the activated state, at most until a facial image of the person is captured by the camera. 16 . The device according to claim 1 , further comprising a network interface, wherein the device is configured to transfer the person-specific data to an apparatus for the automated generation of an identification document from the person-specific data or to an apparatus for verifying the person-specific data. 17 . The device according to claim 1 , wherein the device is a freestanding machine, wherein the machine is free from a back wall arranged behind the face of the person to be captured as viewed from the camera. 18 . The device according to claim 1 , wherein the lighting apparatus is arranged around the semi-transparent mirror. 19 . A method for capturing person-specific data of a person by means of a device, wherein the person-specific data comprises a facial image of the person, wherein the device has a
using acquisition arrangements · CPC title
Combinations of cameras with lighting apparatus; Flash units · CPC title
Illumination specially adapted for pattern recognition, e.g. using gratings · CPC title
Physics · mapped topic
Physics · mapped topic
Related publications grouped by family.
Answers are generated from the same data shown on this page.