(19)
(11)EP 3 343 896 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
15.07.2020 Bulletin 2020/29

(21)Application number: 17207879.2

(22)Date of filing:  16.12.2017
(51)Int. Cl.: 
H04N 5/225  (2006.01)
H04N 1/028  (2006.01)
G06K 7/10  (2006.01)
H04N 5/235  (2006.01)
H04N 1/193  (2006.01)

(54)

A METHOD AND SYSTEM FOR SYNCHRONIZING ILLUMINATION TIMING IN A MULTI-SENSOR IMAGER

VERFAHREN UND SYSTEM ZUM SYNCHRONISIEREN DES BELEUCHTUNGS-TIMINGS IN EINEM MULTISENSORBILDGEBER

PROCÉDÉ ET SYSTÈME PERMETTANT DE SYNCHRONISER L'ÉCLAIRAGE DANS UN IMAGEUR À CAPTEURS MULTIPLES


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 28.12.2016 CN 201611235124

(43)Date of publication of application:
04.07.2018 Bulletin 2018/27

(73)Proprietor: Hand Held Products, Inc.
Fort Mill, SC 29707 (US)

(72)Inventors:
  • CHEN, Feng
    Morris Plains, NJ New Jersey 07950 (US)
  • REN, Jie
    Morris Plains, NJ New Jersey 07950 (US)
  • QU, Haiming
    Morris Plains, NJ New Jersey 07950 (US)
  • ZHANG, Qing
    Morris Plains, NJ New Jersey 07950 (US)

(74)Representative: Haseltine Lake Kempner LLP 
Lincoln House, 5th Floor 300 High Holborn
London WC1V 7JH
London WC1V 7JH (GB)


(56)References cited: : 
EP-A1- 3 197 145
US-A1- 2006 061 680
WO-A1-2016/031359
US-A1- 2016 373 652
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    FIELD OF THE INVENTION



    [0001] The present invention relates to multi-sensor imaging systems, and in particular to the illumination techniques in such systems.

    BACKGROUND



    [0002] Generally speaking with the development of new imaging technologies and customer demand for increasingly better imaging products, multi-sensor imaging systems with dedicated central processing units (CPUs) are necessary. For better user experience and power efficiency, illumination systems for the sensor must be synchronous to cover a common field of view (FOV) area and to improve illumination strength Some illuminations must be isolated from the synchronous illumination to avoid interfering with other sensors. However, in most cases the illumination pulse is controlled by sensors, such as global shutter sensors. It is difficult to synchronize the operation of the sensors and the illumination pulses.

    [0003] For example, one CPU includes one or two imaging sensor interfaces. In a bi-optics image scanner, there are at least four sides (bottom side, top side, left side, and right side) to capture the image. Two to four CPUs are required to support four image sensors. Generally, each image sensor has its own illumination source. If illumination timing from each of the image sensors isn't controlled, illumination from the top side can enter into the bottom side sensor directly, for example. Usually, the timing of the image sensors is adjusted to avoid illumination conflict and improve illumination overlap according to the instant requirements. However, in various CPU systems, it is difficult to adjust the sensor timings to get the appropriate illumination time, whether it be overlap timing or non-overlap timing.

    [0004] Therefore, a need exists for a multi-sensor illumination timing control which is simple to implement and provides stable illumination with no flickering.

    [0005] Reference is made to US 2016/373652 A1, which discloses: an imaging apparatus including a light source that, in operation, emits first pulsed light and second pulsed light, an image sensor that includes at least one pixel including a photodiode, a first charge accumulator and a second charge accumulator, the first charge accumulator and the second charge accumulator, in operation, accumulating signal charge from the photodiode, and a control circuit that, in operation, controls the image sensor. The control circuit, in operation, causes the first charge accumulator to begin to accumulate the signal charge a period of time after the light source begins to emit the first pulsed light. The control circuit, in operation, causes the second charge accumulator to begin to accumulate the signal charge the period of time after the light source begins to emit the second pulsed light.

    SUMMARY



    [0006] Accordingly, the present invention embraces a system according to claim 1.

    [0007] In an exemplary embodiment, the system is comprised of N number of image sensors with active pulse illumination, N being a natural number > 2. Each sensor is configured to generate a frame synchronous signal (FENN) for image capture, where subscript N indicates the image sensor corresponding to the frame synchronous signal. The system further comprises at least two CPUs. Each of the CPUs controls at least 1 of the N number of image sensors. The system further includes an illumination control block communicatively linked to each of the CPUs. The illumination control block is configured to generate illumination pulses for each of N image sensors. The illumination pulse for each of the N image sensors being set to have the same pulse period (Tp) and active pulse width (Wp). The active width pulse for each illumination pulse is set to have maximum exposure time for each of N image sensors.

    [0008] The illumination control block is further configured to communicate pulse period (Tp) and active pulse width (Wp) for the illumination pulses to each of the CPUs. Each of the CPUs is configured to ensure during initialization of image capture that time to capture a frame (Tfr) plus a time interval between subsequent image captures (Twait) is equal to the illumination pulse period Tp communicated by illumination control block, and is therefore the same for each sensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN. Each of the CPUs is further configured to monitor the frame synchronous signal (FENN) and corresponding illumination pulse for the image sensor under the control of the particular CPU. Further, each of the CPUs is configured to calculate offset periods (TdN and TdN+1), the offset period being between the negative edge of the illumination pulse and the negative edge of the frame synchronous signal corresponding to each of N image sensors. Each of the CPUs is further configured to adjust the TwaitN to TwaitNR, where TwaitNR = TwaitN + TdN. Finally, each of the CPUs is configured to align the negative edge of FENN and the corresponding illumination pulse.

    [0009] In another exemplary aspect of the system, each of the CPUs corresponds on a one-to-one basis with the N number of sensors.

    [0010] In another exemplary aspect of the system, each of the CPUs controls 2 of the N number of sensors.

    [0011] In another exemplary aspect of the system, the illumination control block is selected from a central processing unit (CPU), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA).

    [0012] In another exemplary aspect of the system, each of the CPUs is configured to integrate the corresponding image sensor image capture with an active integration time.

    [0013] In another exemplary aspect of the system, each of the CPUs is configured to limit the active integration time of the corresponding image sensor according to the pulse width (Wp).

    [0014] In another exemplary aspect of the system, the active integration time is Wp based upon TintN > Wp, TintN being the pixel integration time for exposure, set by the image sensor N.

    [0015] In another exemplary aspect of the system, the active integration time is TintN based upon TintN < Wp, where TintN is the pixel integration time for exposure, set by the sensor N.

    [0016] In another exemplary aspect of the system, each illumination pulse has a frequency that is equal to a frame rate of the corresponding image sensor.

    [0017] In another exemplary aspect of the system, each illumination pulse has a frequency that is equal to twice a frame rate of the corresponding image sensor.

    [0018] The present invention further embraces a method according to claim 8.

    [0019] In an aspect, the method comprises the steps of: (a) providing an illumination pulse for each of N sensors from an illumination control block; (b) setting each of N illumination pulses to have the same pulse period (Tp) and active pulse width (Wp); (c) setting the active width pulse for each of N illumination pulses to have maximum exposure time for each of N image sensors; (d) ensuring during initialization of image capture that the time to capture a frame (Tfr) plus the time interval between subsequent image captures (Twait) is the same for each sensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, and is equal to the pulse period (Tp) ; (e) monitoring the frame synchronous signal (FENN) and corresponding illumination pulse for each sensor; (f) obtaining the offset periods (TdN and TdN+1) between FENN and FENN+1 from the monitoring step, the offset period being between the negative edge of the illumination pulse and the negative edge of the frame synchronous signal; (g) adjusting the TwaitN to TwaitNR, where TwaitNR = TwaitN + TdN, the TdN being determined in the monitoring step for the next frame capture; and (h) aligning the negative edge of FENN and the corresponding illumination pulse.

    [0020] In another exemplary aspect, the method further comprises the step of (i) integrating the sensors image capture with an active integration time.

    [0021] In another exemplary aspect, embodiment, the method further comprises the step of limiting the active integration time according to the pulse width (Wp).

    [0022] In another exemplary aspect of the method, the active integration time is Wp based upon TintN > Wp, TintN being the pixel integration time for exposure, set by the sensor N.

    [0023] In another exemplary aspect of the method, the active integration time is TintN based upon TintN < Wp, where TintN is the pixel integration time for exposure, set by the sensor N.

    [0024] In another exemplary aspect of the method, the illumination pulse frequency is twice the frame rate of the corresponding sensor.

    [0025] In another exemplary aspect of the method, the illumination pulse frequency the same as the frame rate of the corresponding sensor.

    [0026] In another exemplary aspect of the method, each of the N sensors is controlled by a corresponding CPU. The monitoring step is accomplished by the CPU corresponding to the image sensor generating the FENN.

    [0027] In yet another exemplary aspect, the method further comprises the step of (j) communicating the pulse period (TP) and active pulse width (Wp) to the corresponding CPU by the illumination control block.

    [0028] In another exemplary aspect of the method, the obtaining, adjusting, and aligning steps are accomplished by the corresponding CPU.

    [0029] The foregoing illustrative summary, as well as other exemplary objectives and/or advantages of the invention, and the manner in which the same are accomplished, are further explained within the following detailed description and its accompanying drawings.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0030] 

    Figure 1 schematically depicts a block diagram of the system in accordance with an exemplary embodiment of the invention.

    Figure 2 schematically depicts in a flow diagram, the functions of the components of the system in accordance with an exemplary embodiment of the present invention.

    Figure 3 schematically depicts the signal flow for the system in an exemplary embodiment of the present invention.

    Figure 4 is a flowchart illustrating a method for synchronizing active illumination pulses in a multi-sensor imager in accordance with an exemplary embodiment of the present invention.


    DETAILED DESCRIPTION



    [0031] The present invention embraces a system for synchronizing illumination timing in a multi-sensor imager.

    [0032] In an exemplary embodiment, depicted in Figure 1, the system (200) comprises an illumination control block (210), image sensors (222, 232, and 234), and CPUs (220 and 230). The system (200) illustrated depicts only 2 CPU's (220 and 230) and only three imaging sensors (222, 232, and 234), however more imaging sensors and corresponding CPU's are possible. In accordance with the present invention, and depicted in the Figure, each CPU controls either one or two imaging sensors. The "X" designation identifies the CPU. The "N" designation identifies the imaging sensor and pulses and signals with the "N" subscript correspond to the imaging sensor with the same "N" designation.

    [0033] The illumination control block (210) is communicatively linked to the CPUs (220 and 230). Illumination control block (210) may be a central processing unit (CPU), a complex programmable logic device (CPLD), a field programmable gate array (FPGA), or the like.

    [0034] The imaging sensors (222, 232, and 234) may be Jade sensors; that is, the imaging software is Java Agent Development Framework-based. In the alternative, the sensors may be global shutter sensors.

    [0035] Referring now to Figure 2, the system (200) of Figure 1, is illustrated showing the functions of the illumination control block (210) and the CPUs (220 and 230). The illumination control block (210) which generates illumination pulses for each sensor in functions (213 and 214) first sets the illumination pulse period (Tp) and the active pulse width (Wp) in function (211). The Tp and Wp are set by the illumination control block (210) to be the same for each illumination pulse. The illumination control block (210) communicates the information about the illumination Tp and Wp to the CPUs (220 and 230) in function (212).

    [0036] The CPUs (220 and 230) control the sensors (222, 232, and 234 respectively) and compel the sensors to initialize image capture in steps (240a, 240b, 240c). The CPU's also are each configured to ensure during initialization of image capture that time to capture a frame (Tfr) plus a time interval between subsequent image captures (Twait) is the same for each sensor and is equal to the illumination pulse period (Tp) communicated by the illumination control block (210). Thus, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN = Tp, or in the present embodiment as shown in the Figure 2 as function (241a, 241b, and 241c), Tfr1 + Twait1 = Tp, Tfr2 + Twait2 = Tp, and Tfr3 + Twait3 = Tp.

    [0037] The CPUs (220 and 230) are configured to monitor the frame synchronous signal (FENN) and corresponding illumination pulse for the image sensor under the control of the particular CPU. For example, in the Figure, CPU1 monitors the frame synchronous signal (FEN1) and Illumination Pulse 1 in function (242a). CPU2 (230) monitors the frame synchronous signal (FEN2) and Illumination Pulse 2 in function (242b). Because CPU2 (230) controls Sensor 2 (232) and Sensor 3 (234), CPU2 (230) also monitors the frame synchronous signal (FEN3) and Illumination Pulse 3 in function (242c).

    [0038] In a similar way, CPU1 (220) and CPU2 (230) calculate offset periods, functions (243a, 243b, 243c), (TdN, TdN+1 and TdN+2 respectively.) The offset period being that between the negative edge of the corresponding illumination pulse and the negative edge of the frame synchronous signal corresponding to each of N image sensors. For example, the offset period Td1 is between the negative edge of illumination pulse 1 and the negative edge of FEN1. In general, the negative edge of a pulse is considered to be the high to low transition of a square wave signal.

    [0039] In functions 244a, 224b, and 244c, the CPUs (220 and 230) are configured to adjust the time interval between image captures, or the TwaitN. Thus, TwaitN is changed to TwaitNR, where TwaitNR = TwaitN + TdN, that is, the previous interval between image captures plus the offset period for a given sensor. Thus, for Sensor 1 (222) Twait1 is adjusted to Twait1R which is Twait1 + Td1. The CPU's (220 and 230) make this adjustment of Twait based upon calculating the offset period, TdN.

    [0040] In functions 245a, 245b, and 245c, the CPU's (220 and 230) are configured to align the negative edge of FENN and the corresponding illumination pulse. For example, CPU1 (220) is configured to align the negative edge of FEN1 and illumination pulse 1.

    [0041] Referring now to Figure 3, the illumination pulse and the frame synchronous signal for only Sensors 1 and 2 (222 and 232) are shown in comparison to each other. Figure 3, is discussed in conjunction with Figures 1 and 2.

    [0042] The active illumination pulse (500) is shown to have a pulse period Tp (501) and a pulse width Wp (502).

    [0043] FEN1 (300) has an offset period Td1 (301) as FEN2 (400) has an offset period Td2 (401). Note that the offset periods TdN ensure that the illumination pulse (500) does not illuminate the same portion of Tfr1 (303) and Tfr2 (403), the frame capture periods. At the same time, the interval between frame captures (TwaitN) should be the same for each sensor. So, in the Figure 3, it can be seen that Twait1 (302) + Tfr1 (303) = Twait2 (402) + Tfr2 (403) at the same time as being offset from each other.

    [0044] It can also be seen from the signal flow in Figure 3 that a new interval between frame captures TwaitNR is equal to TwaitN + TdN. In particular, Twait1R (304) = Twait1 (302) + Td1 (301) and Twait2R (404) = Twait2 (402) + Td2 (401).

    [0045] Also shown in Figure 3 are the pixel integration times: Tint1 (305) for Sensor 1 (222) and Tint2 for Sensor 2 (232) . The active integration time for each sensor may be equal to the pixel integration time, TintN, when TintN < Wp (502), the illumination pulse width. Alternatively, the active integration time may be equal to the illumination pulse width, Wp (502) when the pixel integration time, TintN is greater than or equal to the illumination pulse width, Wp (502).

    [0046] In Figure 3, the illumination pulse (500) has a frequency which is equal to the frame rate of Sensor 1 (222) and Sensor 2 (232). Alternatively, the illumination pulse (500) may have a frequency which is twice the frame rate of the sensors (222 and 232) (not shown in Figure).

    [0047] The present invention also embraces a method for synchronizing active illumination pulses in a multi-sensor imager. The method which will be described in conjunction with the system illustrated in Figures 1, 2, and 3.

    [0048] Referring to Figure 4, in an exemplary embodiment, the method comprises the steps of: (601) providing an illumination pulse for each of N sensors from an illumination control block; (602) setting each of N illumination pulses to have the same pulse period (Tp) and active pulse width (Wp); (603) setting the active width pulse for each of N illumination pulses to have maximum exposure time for each of N image sensors; (604) ensuring during initialization of image capture that the time to capture a frame (Tfr) plus the time interval between subsequent image captures (Twait) is the same for each sensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, and is equal to the pulse period (Tp); (605) monitoring the frame synchronous signal (FENN) and corresponding illumination pulse for each sensor; (606) obtaining the offset periods (TdN and TdN+1) between FENN and FENN+1 from the monitoring step, the offset period being between the negative edge of the illumination pulse and the negative edge of the frame synchronous signal; (607) adjusting the TwaitN to TwaitNR, where TwaitNR = TwaitN + TdN, the TdN being determined in the monitoring step for the next frame capture; and (608) aligning the negative edge of FENN and the corresponding illumination pulse.

    [0049] The method may further comprise the step of (609) communicating the pulse period (TP) and active pulse width (Wp) to the corresponding CPU by the illumination control block after the (602) setting step.

    [0050] The method may further comprise the steps of (610) integrating the sensors image capture with an active integration time and (611) of limiting the active integration time according to the pulse width (Wp).

    [0051] For step (611), the active integration time is set to be the illumination pulse width, Wp, based upon the pixel integration time, TintN being greater than or equal to Wp. TintN being set by the corresponding sensor N.

    [0052] For step (611), the active integration time is set to be the pixel integration time, TintN, based upon the pixel integration time being less than the illumination pulse width Wp.

    [0053] In the method, each of the N sensors is controlled by a corresponding CPU. The monitoring step (605) is accomplished by the CPU corresponding to the image sensor generating the FENN. The obtaining (606), adjusting (607), and aligning (608) steps are accomplished by the corresponding CPU.

    [0054] In the specification and/or figures, aspects of the invention have been disclosed. The use of the term "and/or" includes any and all combinations of one or more of the associated listed items. The figures are schematic representations and so are not necessarily drawn to scale. Unless otherwise noted, specific terms have been used in a generic and descriptive sense and not for purposes of limitation.


    Claims

    1. A system (200) for, for each of multiple imaging sensors in a multi-sensor imager, aligning a start of a time to capture a frame with a corresponding illumination pulse, comprising:

    N number of imaging sensors (222, 232, 234) with active pulse illumination, N being a natural number ≥ 2, each imaging sensor being configured to generate a frame synchronous signal, FENN (300, 400), wherein a time when the frame synchronous signal of the imaging sensor becomes low is the start of the time to capture a frame (TfrN), and a time for which the frame synchronous signal of the imaging sensor is high is a time (TwaitN) including the pixel integration time (TintN) of the imaging sensor, the pixel integration time for the imaging sensor ending at a negative edge of the frame synchronous signal (FENN), where subscript N indicates the imaging sensor corresponding to the frame synchronous signal;

    at least 2 CPUs (220, 230), each of the CPUs controlling at least 1 of the N number of imaging sensors;

    an illumination control block (210) communicatively linked to each of the CPUs;

    the illumination control block being configured to generate (213, 214) illumination pulses for each of N imaging sensors, the illumination pulse for each of N imaging sensors being set (211) to have the same pulse period, Tp, (501) and active pulse width, Wp, (502) the active pulse width setting a maximum period for active integration time for each of the N imaging sensors, the active integration time for an imaging sensor being the active pulse width if the pixel integration time of the imaging sensor is equal to or greater than the active pulse width, and the active integration time for the imaging sensor being the pixel integration time if the pixel integration time is less than the active pulse width;

    the illumination control block being configured to communicate (212) pulse period, Tp, and active pulse width, Wp, for the illumination pulses to each of the CPUs; each of the CPUs being configured to ensure (241a, 241b, 241c) during initialization of image capture that time to capture a frame, Tfr, plus a time interval between subsequent image captures, Twait, is equal to the illumination pulse period Tp, and is the same for each imaging sensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN;

    wherein the time to capture the frame, Tfr, starts at a Tfr negative edge and ends at a Tfr positive edge and the time interval between subsequent image captures, Twait, starts at a Twait positive edge and ends at a Twait negative edge;

    each of the CPUs being configured to monitor (242a, 242b, 242c) the frame synchronous signal, FENN, and corresponding illumination pulse for the imaging sensor under the control of the CPU;

    each of the CPUs being further configured to calculate (243a, 243b, 243c) offset periods, TdN and TdN+1, the offset period being between the negative edge of the illumination pulse and the negative edge of the frame synchronous signal corresponding to each of N imaging sensors, the negative edge of the frame synchronous signal being the Twait negative edge;

    each of the CPUs being further configured to adjust (244a, 244b, 244c) the TwaitN to TwaitNR, where TwaitNR = TwaitN + TdN such that the negative edge of the adjusted TwaitN and the negative edge of the corresponding illumination pulse are aligned (245a, 245b, 245c).


     
    2. The system of claim 1, wherein each of the CPUs correspond on a one-to-one basis with the N number of imaging sensors.
     
    3. The system of claim 1, wherein each of the CPUs control 2 of the N number of imaging sensors.
     
    4. The system of claim 1, wherein the illumination control block is selected from a CPU, a CPLD, and a FPGA.
     
    5. The system of claim 1, wherein each of the CPUs is configured to integrate the corresponding imaging sensor image capture with an active integration time.
     
    6. The system of claim 1, wherein each illumination pulse has a frequency that is equal to a frame rate of the corresponding imaging sensor.
     
    7. The system of claim 1, wherein each illumination pulse has a frequency that is equal to twice a frame rate of the corresponding imaging sensor.
     
    8. A method (600) for , for each of multiple imaging sensors in a multi-sensor imager, aligning a start of a time to capture a frame with a corresponding illumination pulse, comprising the steps of:

    a) setting (211), by an illumination control block (210) communicatively linked to each of at least 2 CPUs (220, 230), each of the CPUs controlling at least 1 of N imaging sensors (222, 232, 234), N being a natural number ≥ 2, each of N illumination pulses for each of the N imaging sensors to have the same pulse period, T , and active pulse width, Wp, the active pulse width setting a maximum period for active integration time for each of the N imaging sensors, the active integration time for an imaging sensor being the active pulse width if the pixel integration time (TintN) of the sensor is equal to or greater than the active pulse width, and the active integration time for the imaging sensor being the pixel integration time if the pixel integration time is less than the active pulse width;

    b) communicating (212), by the illumination control block, pulse period, Tp, and active pulse width, Wp, for the illumination pulses to each of the CPUs;

    c) generating (213, 214), by the illumination control block, illumination pulses for each of the N imaging sensors, each imaging sensor being configured to generate a frame synchronous signal, FENN (300, 400), wherein a time when the frame synchronous signal of the imaging sensor becomes low is the start of the time to capture a frame (TfrN), and a time for which the frame synchronous signal of the imaging sensor is high is a time (TwaitN,) including the pixel integration time (TintN) of the imaging sensor, the pixel integration time ending at a negative edge of the frame synchronous signal (FENN), where subscript N indicates the imaging sensor corresponding to the frame synchronous signal;

    d) ensuring (241a, 241b, 241c), by the at least 2 CPUs, during initialization of image capture that the time to capture a frame, Tfr, plus the time interval between subsequent image captures, Twait, is the same for each imaging sensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, and is equal to the pulse period, TP, wherein the time to capture the frame, Tfr, starts at a Tfr negative edge and ends at a Tfr positive edge and the time interval between subsequent image captures, Twait, starts at a Twait positive edge and ends at a Twait negative edge;

    e) monitoring (242a, 242b, 242c), by the at least 2 CPUs, the frame synchronous signal, FENN, and corresponding illumination pulse for each imaging sensor;

    f) calculating (243a, 243b, 243c), by the at least 2 CPUs, offset periods, TdN and TdN+1, between FENN and FENn+1 from the monitoring step, the offset period being between the negative edge of the illumination pulse and the negative edge of the frame synchronous signal, the negative edge of the frame synchronous signal being the Twait negative edge;

    g) adjusting (244a, 244b, 244c), by the at least 2 CPUs, the TwaitN to TwaitNR, where TwaitNR = TwaitN + TdN, the TdN being determined in the monitoring step for the next frame capture, such that the negative edge of the adjusted TwaitN and the negative edge of the corresponding illumination pulse are aligned (245a, 245b, 245c).


     
    9. The method of claim 8, further comprising the step of integrating (610) the imaging sensors image capture with an active integration time.
     


    Ansprüche

    1. System (200) zum Angleichen eines Zeitbeginns zum Aufnehmen eines Einzelbilds an einen entsprechenden Beleuchtungsimpuls für jeden von mehreren Bildgebungssensoren in einem Multisensorbildgeber, umfassend:

    eine Anzahl N von Bildgebungssensoren (222, 232, 234) mit aktiver Impulsbeleuchtung, wobei N eine natürliche Zahl ≥ 2 ist, wobei jeder Bildgebungssensor konfiguriert ist, um ein Einzelbild-synchrones Signal FENN zu erzeugen (300, 400), wobei eine Zeit, zu der das Einzelbild-synchrone Signal des Bildgebungssensors niedrig wird, der Beginn der Zeit zum Aufnehmen eines Einzelbilds (TfrN) ist und eine Zeit, zu der das Einzelbild-synchrone Signal des Bildgebungssensors hoch ist, eine Zeit (TwaitN) ist, die die Pixelintegrationszeit (TintN) des Bildgebungssensors enthält, wobei die Pixelintegrationszeit für den Bildgebungssensor an einer negativen Flanke des Einzelbild-synchronen Signals (FENN) endet, wobei der Index N den dem Einzelbild-synchronen Signal entsprechenden Bildgebungssensor angibt;

    mindestens 2 CPU (220, 230), wobei jede der CPU mindestens 1 der Anzahl N von Bildgebungssensoren steuert;

    einen Beleuchtungssteuerblock (210), der kommunikativ mit jeder der CPU verbunden ist;

    wobei der Beleuchtungssteuerblock konfiguriert ist, um Beleuchtungsimpulse für jeden der N Bildgebungssensoren zu erzeugen (213, 214), wobei der Beleuchtungsimpuls für jeden der N Bildgebungssensoren eingestellt ist (211), um die gleiche Impulsperiode, Tp, (501) und die gleiche aktive Impulsbreite, Wp, (502) aufzuweisen, wobei die aktive Impulsbreite eine maximale Periode für die aktive Integrationszeit für jeden der N Bildgebungssensoren einstellt, wobei die aktive Integrationszeit für einen Bildgebungssensor die aktive Impulsbreite ist, wenn die Pixelintegrationszeit des Bildgebungssensors gleich oder größer als die aktive Impulsbreite ist, und die aktive Integrationszeit für den Bildgebungssensor die Pixelintegrationszeit ist, wenn die Pixelintegrationszeit kleiner als die aktive Impulsbreite ist;

    wobei der Beleuchtungssteuerblock konfiguriert ist, um die Impulsperiode, Tp, und die aktive Impulsbreite, Wp, für die Beleuchtungsimpulse an jede der CPU zu kommunizieren (212);

    jede der CPU während der Initialisierung der Bilderfassung konfiguriert ist, um sicherzustellen (241a, 241b, 241c), dass die Zeit zum Aufnehmen eines Einzelbilds, Tfr, plus ein Zeitintervall zwischen aufeinanderfolgenden Bilderfassungen, Twait, gleich der Beleuchtungsimpulsperiode Tp ist und gleich für jeden Bildgebungssensor, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, ist;

    wobei die Zeit zum Aufnehmen des Einzelbilds, Tfr, an einer negativen Tfr-Flanke beginnt und an einer positiven Tfr-Flanke endet und das Zeitintervall zwischen aufeinanderfolgenden Bilderfassungen, Twait, an einer positiven Twait-Flanke beginnt und an einer negativen Twait-Flanke endet;

    jede der CPU konfiguriert ist, um das Einzelbild-synchrone Signal, FENN, und den entsprechenden Beleuchtungsimpuls für den Bildgebungssensor unter der Steuerung der CPU zu überwachen (242a, 242b, 242c);

    jede der CPU ferner konfiguriert ist, um Versatzperioden, TdN und TdN+1, zu berechnen (243a, 243b, 243c), wobei die Versatzperiode zwischen der negativen Flanke des Beleuchtungsimpulses und der negativen Flanke des Einzelbild-synchronen Signals, das jedem von N Bildgebungssensoren entspricht, liegt, wobei die negative Flanke des Einzelbild-synchronen Signals die negative Twait-Flanke ist;

    jede der CPU ferner konfiguriert ist, um die TwaitN auf TwaitNR anzupassen (244a, 244b, 244c), wobei TwaitNR = TwaitN + TdN, sodass die negative Flanke der angepassten TwaitN und die negative Flanke des entsprechenden Beleuchtungsimpulses angeglichen sind (245a, 245b, 245c).


     
    2. System nach Anspruch 1, wobei jede der CPU auf einer Eins-zu-eins-Basis der Anzahl N von Bildgebungssensoren entspricht.
     
    3. System nach Anspruch 1, wobei jede der CPU 2 der Anzahl N von Bildgebungssensoren steuert.
     
    4. System nach Anspruch 1, wobei der Beleuchtungssteuerblock aus einer CPU, einer CPLD und einem FPGA ausgewählt ist.
     
    5. System nach Anspruch 1, wobei jede der CPU konfiguriert ist, um die entsprechende Bilderfassung des Bildgebungssensors mit einer aktiven Integrationszeit zu integrieren.
     
    6. System nach Anspruch 1, wobei jeder Beleuchtungsimpuls eine Frequenz aufweist, die gleich einer Bildrate des entsprechenden Bildgebungssensors ist.
     
    7. System nach Anspruch 1, wobei jeder Beleuchtungsimpuls eine Frequenz aufweist, die der doppelten Bildrate des entsprechenden Bildgebungssensors entspricht.
     
    8. Verfahren (600) zum Angleichen eines Zeitbeginns zum Aufnehmen eines Einzelbilds an einen entsprechenden Beleuchtungsimpuls für jeden von mehreren Bildgebungssensoren in einem Multisensorbildgeber, umfassend die folgenden Schritte:

    a) Einstellen (211), durch einen Beleuchtungssteuerblock (210), der kommunikativ mit jeder von mindestens 2 CPU (220, 230) verbunden ist, wobei jede der CPU mindestens 1 der N Bildgebungssensoren (222, 232, 234) steuert, wobei N eine natürliche Zahl ≥ 2 ist, wobei jeder der N Beleuchtungsimpulse für jeden der N Bildgebungssensoren die gleiche Impulsperiode, Tp, und die gleiche aktive Impulsbreite, Wp, aufweist, wobei die aktive Impulsbreite eine maximale Periode für die aktive Integrationszeit für jeden der N Bildgebungssensoren einstellt, wobei die aktive Integrationszeit für einen Bildgebungssensor die aktive Impulsbreite ist, wenn die Pixelintegrationszeit (TintN) des Sensors gleich oder größer als die aktive Impulsbreite ist, und die aktive Integrationszeit für den Bildgebungssensor die Pixelintegrationszeit ist, wenn die Pixelintegrationszeit kleiner als die aktive Impulsbreite ist;

    b) Kommunizieren (212), durch den Beleuchtungssteuerblock, der Impulsperiode, Tp, und der aktiven Impulsbreite, Wp, für die Beleuchtungsimpulse an jede der CPU;

    c) Erzeugen (213, 214), durch den Beleuchtungssteuerblock, von Beleuchtungsimpulsen für jeden der N Bildgebungssensoren, wobei jeder Bildgebungssensor konfiguriert ist, um ein Einzelbild-synchrones Signal, FENN, zu erzeugen (300, 400), wobei eine Zeit, zu der das Einzelbild-synchrone Signal des Bildgebungssensors niedrig wird, der Beginn der Zeit zum Aufnehmen eines Einzelbilds (TfrN) ist und eine Zeit, zu der das Einzelbild-synchrone Signal hoch ist, eine Zeit (TwaitN) ist, die die Pixelintegrationszeit (TintN) des Bildgebungssensors enthält, wobei die Pixelintegrationszeit an einer negativen Flanke des Einzelbild-synchronen Signals (FENN) endet, wobei der Index N den Bildgebungssensor angibt, der dem Einzelbild-synchronen Signal entspricht;

    d) Sicherstellen (241a, 241b, 241c), durch die mindestens 2 CPU während der Initialisierung der Bilderfassung, dass die Zeit zum Aufnehmen eines Einzelbilds, Tfr, plus das Zeitintervall zwischen aufeinanderfolgenden Bilderfassungen, Twait, für jeden Bildgebungssensor gleich, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, ist und gleich der Impulsperiode, Tp, ist, wobei die Zeit zum Aufnehmen des Einzelbilds, Tfr, bei einer negativen Tfr-Flanke beginnt und bei einer positiven Tfr-Flanke endet und das Zeitintervall zwischen aufeinanderfolgenden Bilderfassungen, Twait, bei einer positiven Twait-Flanke beginnt und bei einer negativen Twait-Flanke endet;

    e) Überwachen (242a, 242b, 242c), durch die mindestens 2 CPU, des Einzelbild-synchronen Signals, FENN, und des entsprechenden Beleuchtungsimpulses für jeden Bildsensor;

    f) Berechnen (243a, 243b, 243c), durch die mindestens 2 CPU, der Versatzperioden, TdN und TdN+1, zwischen FENN und FENN+1 aus dem Überwachungsschritt, wobei die Versatzperiode zwischen der negativen Flanke des Beleuchtungsimpulses und der negativen Flanke des Einzelbild-synchronen Signals liegt, wobei die negative Flanke des Einzelbild-synchronen Signals die negative Twait-Flanke ist;

    g) Anpassen (244a, 244b, 244c), durch die mindestens 2 CPU, der TwaitN auf TwaitNR, wobei TwaitNR = TwaitN + TdN, wobei die TdN in dem Überwachungsschritt für die nächsten Einzelbilderfassung bestimmt wird, sodass die negative Flanke der angepassten TwaitN und die negative Flanke des entsprechenden Beleuchtungsimpulses angeglichen sind (245a, 245b, 245c).


     
    9. Verfahren nach Anspruch 8, ferner umfassend den Schritt des Integrierens (610) der Bilderfassung der Bildgebungssensoren mit einer aktiven Integrationszeit.
     


    Revendications

    1. Système (200) permettant, pour chacun des plusieurs capteurs d'imagerie dans un imageur à capteurs multiples, de synchroniser un début d'un temps pour capturer une trame avec une impulsion d'éclairage correspondante, comprenant :

    un nombre N de capteurs d'imagerie (222, 232, 234) avec éclairage par impulsion active, N étant un nombre naturel ≥ 2, chaque capteur d'imagerie étant configuré pour générer un signal synchrone de trame, FENN (300, 400), un temps où le signal synchrone de trame du capteur d'imagerie devient faible étant le début du temps de capture d'une trame (TfrN), et un temps pendant lequel le signal synchrone de trame du capteur d'imagerie est élevé étant un temps (TwaitN) incluant le temps d'intégration de pixels (TintN) du capteur d'imagerie, le temps d'intégration de pixels pour le capteur d'imagerie se terminant au niveau d'un front négatif du signal synchrone de trame (FENN), l'indice N indiquant le capteur d'imagerie correspondant au signal synchrone de trame ;

    au moins 2 CPU (220, 230), chacun des CPU contrôlant au moins 1 du nombre N de capteurs d'imagerie ;

    un bloc de commande d'éclairage (210) lié en communication à chacun des CPU ;

    le bloc de commande d'éclairage étant configuré pour générer (213, 214) des impulsions d'éclairage pour chacun des N capteurs d'imagerie, l'impulsion d'éclairage pour chacun des N capteurs d'imagerie étant réglée (211) pour avoir la même période d'impulsion, Tp, (501) et largeur d'impulsion active, Wp, (502) la largeur d'impulsion active définissant une période maximale pour le temps d'intégration actif pour chacun des N capteurs d'imagerie, le temps d'intégration actif pour un capteur d'imagerie étant la largeur d'impulsion active si le temps d'intégration de pixels du capteur d'imagerie est supérieur ou égal à la largeur d'impulsion active, et le temps d'intégration actif pour le capteur d'imagerie étant le temps d'intégration de pixels si le temps d'intégration de pixels est inférieur à la largeur d'impulsion active ;

    le bloc de commande d'éclairage étant configuré pour communiquer (212) la période d'impulsion, Tp, et la largeur d'impulsion active, Wp, pour les impulsions d'éclairage à chacun des CPU ;

    chacun des CPU étant configuré pour garantir (241a, 241b, 241c) pendant l'initialisation de la capture d'image que le temps pour capturer une trame, Tfr, plus un intervalle de temps entre les captures d'image suivantes, Twait, est égal à la période d'impulsion d'éclairage Tp, et est le même pour chaque capteur d'imagerie, Tfr1 + Twait1 = Ttr2 + Twait2 = TfrN + TwaitN ;

    le temps pour capturer la trame, Tfr, commençant au niveau d'un front négatif de Tfr et se terminant au niveau d'un front positif de Tfr et l'intervalle de temps entre les captures d'image suivantes, Twait, commençant au niveau d'un front positif de Twait et se terminant au niveau d'un front négatif de Twait ;

    chacun des CPU étant configuré pour surveiller (242a, 242b, 242c) le signal synchrone de trame, FENN, et l'impulsion d'éclairage correspondante pour le capteur d'imagerie sous le contrôle du CPU ;

    chacun des CPU étant en outre configuré pour calculer (243a, 243b, 243c) des périodes de décalage, TdN et TdN+1, la période de décalage étant entre le front négatif de l'impulsion d'éclairage et le front négatif du signal synchrone de trame correspondant à chacun des N capteurs d'imagerie, le front négatif du signal synchrone de trame étant le front négatif de Twait ;

    chacun des CPU étant en outre configuré pour ajuster (244a, 244b, 244c) le TwaitN à TwaitNR, où TwaitNR = TwaitN + TdN, de telle sorte que le front négatif de TwaitN ajusté et le front négatif de l'impulsion d'éclairage correspondante sont alignés (245a, 245b, 245c).


     
    2. Système selon la revendication 1, chacun des CPU correspondant sur une base un à un au nombre N de capteurs d'imagerie.
     
    3. Système selon la revendication 1, chacun des CPU commandant 2 du nombre N de capteurs d'imagerie.
     
    4. Système selon la revendication 1, le bloc de commande d'éclairage étant sélectionné parmi un CPU, un CPLD et un FPGA.
     
    5. Système selon la revendication 1, chacun des CPU étant configuré pour intégrer la capture d'image de capteur d'imagerie correspondante avec un temps d'intégration actif.
     
    6. Système selon la revendication 1, chaque impulsion d'éclairage ayant une fréquence qui est égale à une fréquence de trame du capteur d'imagerie correspondant.
     
    7. Système selon la revendication 1, chaque impulsion d'éclairage ayant une fréquence qui est égale à deux fois une fréquence de trame du capteur d'imagerie correspondant.
     
    8. Procédé (600) permettant, pour chacun des plusieurs capteurs d'imagerie dans un imageur à capteurs multiples, de synchroniser un début d'un temps pour capturer une trame avec une impulsion d'éclairage correspondante, comprenant les étapes suivantes :

    a) régler (211), à l'aide d'un bloc de commande d'éclairage (210) lié de manière communicative à chacun d'au moins 2 CPU (220, 230), chacun des CPU contrôlant au moins 1 parmi N capteurs d'imagerie (222, 232, 234), N étant un nombre naturel ≥ 2, chacune des N impulsions d'éclairage pour chacun des N capteurs d'imagerie ayant la même période d'impulsion, Tp, et la largeur d'impulsion active, Wp, la largeur d'impulsion active fixant une période maximale pour le temps d'intégration active pour chacun des N capteurs d'imagerie, le temps d'intégration active pour un capteur d'imagerie étant la largeur d'impulsion active si le temps d'intégration de pixel (TintN) du capteur est supérieur ou égal à la largeur d'impulsion active, et le temps d'intégration active pour le capteur d'imagerie étant le temps d'intégration de pixels si le temps d'intégration de pixels est inférieur à la largeur d'impulsion active ;

    b) communiquer (212), à l'aide du bloc de commande d'éclairage, la période d'impulsion, Tp, et la largeur d'impulsion active, Wp, pour les impulsions d'éclairage à chacun des CPU ;

    c) générer (213, 214), à l'aide du bloc de commande d'éclairage, des impulsions d'éclairage pour chacun des N capteurs d'imagerie, chaque capteur d'imagerie étant configuré pour générer un signal synchrone de trame, FENN (300, 400), un temps où le signal synchrone de trame du capteur d'imagerie devient faible étant le début du temps de capture d'une trame (TfrN), et un temps pendant lequel le signal synchrone de trame du capteur d'imagerie est élevé étant un temps (TwaitN) incluant le temps d'intégration de pixels (TintN) du capteur d'imagerie, le temps d'intégration de pixels se terminant au niveau d'un front négatif du signal synchrone de trame (FENN), l'indice N indiquant le capteur d'imagerie correspondant au signal synchrone de trame ;

    d) garantir (241a, 241b, 241c), à l'aide desdits au moins 2 CPU, lors de l'initialisation de la capture d'image que le temps de capture d'une trame, Tfr, plus l'intervalle de temps entre les captures d'image suivantes, Twait, est le même pour chaque capteur d'imagerie, Tfr1 + Twait1 = Tfr2 + Twait2 = TfrN + TwaitN, et est égal à la période d'impulsion, Tp, le temps pour capturer la trame, Tfr, commençant au niveau d'un front négatif de Tfr et se terminant au niveau d'un front positif de Tfr et l'intervalle de temps entre les captures d'image suivantes, Twait, commençant au niveau d'un front positif de Twait et se terminant au niveau d'un front négatif de Twait ;

    e) surveiller (242a, 242b, 242c), à l'aide desdits au moins 2 CPU, le signal synchrone de trame, FENN, et l'impulsion d'éclairage correspondante pour chaque capteur d'imagerie ;

    f) calculer (243a, 243b, 243c), à l'aide desdits au moins 2 CPU, des périodes de décalage, TdN et TdN+1, entre FENN et FENN+1 à partir de l'étape de surveillance, la période de décalage courant entre le front négatif de l'impulsion d'éclairage et le front négatif du signal synchrone de trame, le front négatif du signal synchrone de trame étant le front négatif de Twait ;

    g) ajuster (244a, 244b, 244c), à l'aide desdits au moins 2 CPU, le TwaitN à TwaitNR, où TwaitNR = TwaitN + TdN, le TdN étant déterminé à l'étape de surveillance de la capture de la trame suivante, de telle sorte que le front négatif de TwaitN ajusté et le front négatif de l'impulsion d'éclairage correspondante sont alignés (245a, 245b, 245c).


     
    9. Procédé selon la revendication 8, comprenant en outre l'étape d'intégration (610) de la capture d'image des capteurs d'imagerie avec un temps d'intégration active.
     




    Drawing















    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description