WO2010045941A2

WO2010045941A2 - Method and device for photographically capturing the image of the field of view of a person

Info

Publication number: WO2010045941A2
Application number: PCT/DE2009/001501
Authority: WO
Inventors: Alfredo Herrera; Frank Weber; Michael Müller; Stefan Christ; Volker Kempert
Original assignee: Cognitec Systems Gmbh
Priority date: 2008-10-21
Filing date: 2009-10-21
Publication date: 2010-04-29
Also published as: WO2010045941A3; DE102008053395A1

Abstract

The invention relates to a method and to a device for photographically capturing the image of the field of view of a particular person in a vertical recording chamber. The invention can be used for automatically detecting, recording, and analyzing the field of view of the person, wherein the face or field of view of the person is localized using optical detecting and recording devices and corresponding device controls, and is compared to prescribed quality features, and wherein the field of view is imaged for further use after analysis and particular corrections, in case of sufficiently good quality features.

Description

Method and device for photographic image capture of the field of view of a person

The invention relates to a method and a device for the photographic image capture of the visual field of a specific person in a virtual recording room.

The invention is applicable to the automatic detection and recording of the visual field of the person in a virtually defined recording room, wherein the visual field of the person taken as an image is used after fulfillment of certain criteria, for example for biometric evaluations.

From DE-OS 195 11 713 Al a method and a device for automatic image recording of a person using an optoelectronic image processing device for directional localization of the person and their face is known, depending on at least one positioning signal of the image processing device, a sensor alignment device, the line of sight Picking camaraderie controls.

From AT-PS 501 370 Bl a device for biometric control of systems, in particular of access barriers, known, in which an adjustable camera for image acquisition of a user with a downstream evaluation unit is connected, which extracts biometric data from the recorded image and from there a control signal are generated for the system, and wherein a memory for receiving a user-specific parameter set and a read-out device are provided which controls the camera setting depending on the parameter set of the respective user for recording. EP 1 280 095 B1 discloses for a person a facial image recognition device with an image input section arranged for input of identification information of the person in a key input section.

From DE-G 9212 628.6 a photo booth with a video camera with a video camera, a video printer and a video screen for reproducing the captured by the camera image is known, wherein the video camera attached to the video screen and with this about the image center axis by at least 90 ° from portrait to landscape format and vice versa rotatably mounted in a substantially vertical, but adjustable to different angles of inclination plane of rotation.

The invention is based on the object to provide a method and a device for detecting and detecting the visual field of a person with which corresponding biometric data of the face are detected by the person, which are suitable to supply the data obtained an ^' evaluation, wherein During the acquisition of appropriate data, behavior-specific information is communicated to the person and the visual field is to be mapped depending on this.

The object is achieved according to the invention with a method according to the features of claim 1 and with a device according to the features of claim 3. Advantageous embodiments of the invention are described in the respective subclaims. The process is carried out according to the invention such that the virtual recording space is permanently searched for faces with the control sensor and the control lens. Once a face is found by the control unit, the capture sensor and the capture lens are activated to capture the face, with the focal length, focus, aperture and exposure time, and sensor gain corresponding to the expected face automatically adjusted.

Simultaneously with these measures, the control unit is an information about the height of the face or field of view of the track, whereupon the slide with the screen as a user display, the recording sensor and the taking lens to the altitude of the face or field of view is adjusted in such a way using a by the eyes formed axis determined, frontal head posture of the person for the taking up position of the camera is based.

When the parameters match, an image is taken with the aid of the exposure sensor and the taking lens, which is analyzed in the corresponding control devices.

If the image meets the predetermined criteria, the process is terminated, otherwise the process continues with the former step until all required criteria are met.

With the method according to the invention, only the recorded face or visual field as an object and criteria derived therefrom, such as the image sharpness, the illumination, the contrast, the homogeneity of the Light distribution in the face, the head posture of the facial expression and partial face covers, such as glasses, hair or a headgear used.

The parameters of the controllable device, such as the setting of the camera and user display height, the exposure times of the camera, the lens focus aperture and lens focal length and the digital sensor gain and additional illumination are combined and processed in the associated evaluation unit.

Of importance here is that the person to be recorded is informed on the basis of the previously recorded image sequences by means of trained as a screen user display on their optical behavior in order to achieve a sufficiently good or optimal recording as quickly as possible.

To do this, the person is instructed accordingly, such as: "Look in front of the camera", "Please step back," "Stand still," Close your mouth, "" Take your glasses off "or" Adopt a neutral face attitude ".

An advantage of the method is that the person to be recorded is not overloaded with behavioral instructions, but is served context-dependent.

For example, if the person is diagnosed with an open mouth, but is unsuitable for a sufficient or optimal image, the person receives the "close their mouth" instruction. The invention will be explained in more detail with reference to an embodiment and a drawings. In the accompanying drawing show:

1 shows a schematically greatly simplified representation of a device according to the invention, and FIG. 2 shows the flowchart of the method according to the invention.

In the figure 1 of the drawing, a device according to the invention is designated by the reference numeral 1.

The device 1 has a arranged in a frame 2 with a housing 3 track 4 with a carriage 5, which is limited to a guide device 6 in a predominantly vertical direction and within a predetermined distance by means of a drive device, not shown movable.

On the guide device 6 is in a suitable

Holder 7 a screen device, such as a monitor 8 is arranged, which is displaceable in height in the region of the predetermined travel path of the guide device 6.

In an area according to the drawing above the monitor 8, a control sensor 9 is arranged in a suitable position on the moving unit 6, which has a control lens 10, a pick-up sensor 11 and a camera 12 with an integrated pick-up lens 13 arranged likewise next to the moving unit 6 is in electrical connection. In the device 1 according to the present invention, it is important that the camera 12 is arranged centrally and at the smallest possible distance above the monitor 8. In an assembly or a part of the device 1, a voice output unit 14 is further provided, which is integrated according to the embodiment described here, for example in the moving unit 4.

The device 1 also has a device control device 15, which is used to control all the components of the device 1 and is therefore electrically connected thereto in a corresponding manner.

As can also be seen from the drawing, a virtual receiving space 16 is provided at a comparatively small distance of, for example, one meter perpendicular to the device 1.

The virtual receiving space 16 is formed by a space of an unspecified height, which according to the drawing extends over an example square and side lengths of about one meter having surface portion 17 on which approximately centrally and opposite to the device 1, a marker 18 is arranged.

The virtual recording room 16 is provided for a comparatively short stay of a particular person 19, which is shown in the drawing very abstract and standing and having a body 20, 21 feet and a head 22 with a face or field of view 23.

The person 19 is shown according to the present embodiment on the surface portion 17 so that the feet 21 at the mark 18 and the face or field of view 23 with a 'computationally determined eye line 24 of the device 1 faces.

The device 1 is further provided with an illumination means 25 for illuminating the image to be recorded and with an output device 26 for the output of a recorded image.

With reference to Figures 1 and 2, the procedure of the method will be briefly described below.

As can be seen from FIGS. 1 and 2, a predetermined person 19 first enters the virtual receiving space 16 to carry out the method, with the aim that the feet 21 are located at the marking 18 and the face or face field 23 is perpendicular to the device 1 and the screen 8 and remains comparatively informal in this position.

In this position, with the aid of the control sensor 9, the position or height of the face or of the field of view 23 is initially determined by means of predetermined value sizes by means of facial pattern detection.

For this purpose, the carriage 5 is moved in such a position along the guide device 6 until the determined from the real eye distance, horizontal eye line 24 of the person is in the same position and perpendicular to the adjusted with the carriage 5 in the same way recording lens 13.

In this position or position of the field of view 23 and determined compliance with the given data, the field of view 23 is received by the camera 12 and the image is stored or applied to a corresponding use.

However, if a discernment of the required quality characteristics is found, the sequence for mapping must be repeated, for which the person 19 receives selective instructions from the device 1 and must act accordingly.

The device 1 is finally equipped with a corresponding illumination means 25, which is controlled in dependence on the respective conditions and requirements. An output unit 26 is finally connected to the device 1 for the output of a respective captured image.

So that not only standing and adult size persons 19 can be detected by the sensors 9 and 11, but also children or, for example, a person sitting in a wheelchair person, the moving unit 4 along the guide means 6 according to these existing situations or sizes in height can be moved ,

List of used reference numbers

1 device for image acquisition

2 frame 3 housing

4 positioning unit

5 sleds

6 guide device

7 bracket 8 screen; monitor

9 control sensor

10 control lens

11 recording sensor

12 Camera 13 Shooting lens

14 voice output unit

15 device control device

16 Virtual Recording Room

17 surface section 18 marking

19 male person

20 bodies

21 feet

22 head 23 face / field of view

24 eye line

25 lighting means

26 output unit A distance

Claims

claims

1. A method of photographic image capture of

Field of view of a specific person in a virtual recording room, with a frame with a screen and a camera for detecting and capturing an image and means for outputting and / or storing the image, characterized in that • the person (19) in the receiving space (16) placed opposite the camera (12) and the screen (8) as a user display device in a first arbitrary arrangement / position such that the field of view (23) is substantially frontal and spaced from the screen (8) and a controllable optic (10) Camera (12) is positioned and held,

• the receiving space (16) is detected by a control lens (10) connected to the camera (12) and the position of the face or field of view (23) is determined by the control unit (15),

The position of the camera (12) is adjusted to the position of the head (22) with respect to a horizontal eye line (24) formed by both eyes and the field of view (23) by a defined adjustment of the camera (12),

If the field of view (23) of the camera (12) is photographically recorded, if the detected field of view (23) and the camera (12) coincide

The image of the visual field (23) is compared with predetermined generalized facial biometric quality features with respect to optically defined criteria, • the result of the comparison is evaluated,

• in which only in the field of view (23) the image sharpness and illumination, the contrast, the homogeneity of the light distribution, the head or face posture, the

Facial expression and also hidden areas of the visual field (23) are considered as quality features,

• If insufficient quality is detected in previous images, the position of the face

(23) and or to correct accordingly the facial expression and / or the head posture with respect to the camera (12) until the

Visual field (23) meets the specified quality characteristics,

• upon fulfillment of the quality characteristics, the face or visual field (23) is photographically photographed definitively and

• the finally scanned field of view (23) or face as an actual image file of a

Evaluation device supplied, evaluated there and stored retrievable.

2. The method according to claim 1, characterized in that provided for the optical criteria, a number of terms or conceptual groups and phonetically stored in a memory array, wherein depending on the determined actual quality characteristics of the visual field (23) a corresponding output or more corresponding outputs to a phonetic output device (speech output unit) and / or to the screen (8) as a user display device as a respective instruction to the person (19) to correct the actual Quality features for predetermined target quality characteristics of the field of view (23) relative to the camera (12) are taught.

3. The method according to claim 1 or 2, characterized in that as quality features hidden areas are valued, such as glasses, deep-seated headgear, even fallen into the face hair, and or other temporarily existing body such as eyepatches or hands are created or incurred.

4. The method according to any one of claims 1 to 3, characterized in that for defining the virtual receiving space (16) of the eye distance of the male person (19) is measured in pixels, the means of real eye distance between 5.8 cm and 7.2 cm is and is converted to real coordinates, wherein it is determined on the basis of the determined conversion parameter for the eye relief and the measured position of the head (22) that the head (22) or the field of view (23) in the receiving space (16).

5. Device for carrying out the method according to

Claim 1, characterized in that in a arranged on a frame holder (7) for the screen (8) a corresponding camera (12) with a taking lens (13) as a receiving device immediately above the screen (8) is arranged, wherein the camera (12) with at least one externally arranged recording sensor (11), an optional illumination means (25) and an optional voice output unit (14) and with a Device control device of a computing and memory arrangement (15) is connected, in which the virtual Aufnähmeräum (16) defined predetermined and stored.

6. Device according to claim 5, characterized in that the voice output unit (14) and the device control or memory arrangement (15) are connected to a trained as a speaker, phonetic output device.

7. Device according to one of claims 5 or 6, characterized in that the receiving space (16) by a provided on the bottom surface portion (17) fixed mark (18) for an optimal / authoritative position of the person (19) relative to the taking lens (13 ) of the camera (12) is defined.

8. Device according to one of claims 5 to 7, characterized in that the holder (7) arranged on a horizontally and vertically movable guide means (6) and the camera (12) is additionally pivotally mounted.

- For this 2 pages drawing -