WO2021238477A1 - 深度相机、电子设备及控制方法 - Google Patents

深度相机、电子设备及控制方法 Download PDF

Info

Publication number
WO2021238477A1
WO2021238477A1 PCT/CN2021/087277 CN2021087277W WO2021238477A1 WO 2021238477 A1 WO2021238477 A1 WO 2021238477A1 CN 2021087277 W CN2021087277 W CN 2021087277W WO 2021238477 A1 WO2021238477 A1 WO 2021238477A1
Authority
WO
WIPO (PCT)
Prior art keywords
signals
taps
signal
normalized
tap
Prior art date
Application number
PCT/CN2021/087277
Other languages
English (en)
French (fr)
Inventor
张学勇
Original Assignee
Oppo广东移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Oppo广东移动通信有限公司 filed Critical Oppo广东移动通信有限公司
Publication of WO2021238477A1 publication Critical patent/WO2021238477A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/06—Systems determining position data of a target
    • G01S17/08—Systems determining position data of a target for measuring distance only
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G06T7/521—Depth or shape recovery from laser ranging, e.g. using interferometry; from the projection of structured light
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof

Definitions

  • This application relates to the field of depth measurement technology, in particular to a depth camera, electronic equipment, and a control method.
  • Pulse modulation I-TOF Indirect-Time of Flight, I-TOF
  • three-dimensional ranging technology uses indirect measurement of the reflected laser pulse delay to obtain depth data.
  • the pulse modulation I-TOF has the advantages of low duty cycle and power saving.
  • the embodiments of the present application provide a depth camera, an electronic device, and a control method.
  • the depth camera of the embodiment of the present application includes a light emitting module, a light receiving module, and a processing circuit.
  • the light emitting module is configured to send one light pulse signal in one sub-period, and send multiple light pulse signals in one frame period, and the frame period includes multiple consecutive sub-periods.
  • the light receiving module includes a pixel, the pixel includes a plurality of taps, the plurality of taps acquire a plurality of electrical signals generated by the light pulse signal and/or the ambient light signal, and any two of the taps are turned on The time does not overlap, and the turn-on frequency of any two of the taps in the frame period is different.
  • the processing circuit is used to calculate the depth information of the shot scene according to the multiple electrical signals.
  • the electronic device of the embodiment of the present application includes a housing and a depth camera.
  • the depth camera is combined with the housing.
  • the depth camera includes a light emitting module, a light receiving module and a processing circuit.
  • the light emitting module is configured to send one light pulse signal in one sub-period, and send multiple light pulse signals in one frame period, and the frame period includes multiple consecutive sub-periods.
  • the light receiving module includes a pixel, the pixel includes a plurality of taps, the plurality of taps acquire a plurality of electrical signals generated by the light pulse signal and/or the ambient light signal, and any two of the taps are turned on The time does not overlap, and the turn-on frequency of any two of the taps in the frame period is different.
  • the processing circuit is used to calculate the depth information of the shot scene according to the multiple electrical signals.
  • the control method of the embodiment of the present application is used for a depth camera.
  • the depth camera includes a light emitting module and a light receiving module.
  • the light receiving module includes pixels, and the pixels include a plurality of taps.
  • the control method includes: the light emitting module emits one optical pulse signal in one sub-period, and transmits a plurality of the optical pulse signals in one frame period, and the frame period includes a plurality of consecutive sub-periods.
  • the multiple taps acquire multiple electrical signals generated by the light pulse signal and/or the ambient light signal, the opening times of any two of the taps do not overlap, and any two of the taps are in the frame period
  • the opening frequencies within are different; and the depth information of the shot scene is calculated according to the multiple electrical signals.
  • Fig. 1 is a schematic diagram of a depth camera according to some embodiments of the present application.
  • FIG. 2 is a schematic diagram of the structure of pixels in a depth camera according to some embodiments of the present application.
  • FIG. 3 is a schematic diagram of the structure of pixels in a depth camera according to some embodiments of the present application.
  • FIG. 4 is a schematic diagram of the structure of pixels in a depth camera according to some embodiments of the present application.
  • FIG. 5 is a schematic diagram of the structure of pixels in a depth camera according to some embodiments of the present application.
  • FIG. 6 is a schematic diagram of a working sequence of a depth camera according to some embodiments of the present application.
  • Fig. 7 is a schematic diagram of an electronic device according to some embodiments of the present application.
  • FIG. 8 is a schematic flowchart of a control method of some embodiments of the present application.
  • the depth camera 100 includes a light emitting module 10, a light receiving module 20 and a processing circuit 30.
  • the light emitting module 10 is used to send one optical pulse signal in one sub-period T1, and send multiple optical pulse signals in one frame period T2, and the frame period T2 includes multiple consecutive sub-periods T1.
  • the light receiving module 20 includes pixels 21.
  • the pixels 21 include multiple taps 23.
  • the multiple taps 23 are used to obtain multiple electrical signals generated by light pulse signals and/or ambient light signals.
  • the opening time of any two taps 23 is different. They overlap, and the turn-on frequencies of any two taps 23 in the frame period T2 are different.
  • the processing circuit 30 is used to calculate the depth information of the shot scene according to a plurality of electrical signals.
  • the pixels 21 each include a photoelectric conversion element 22, each tap 23 includes a transfer gate 231, the photoelectric element 22 is used to receive light pulse signals and/or ambient light The signal is used to obtain an electrical signal, and the transfer gate 231 is used to transfer the electrical signal.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is a polygon, and one or more taps 23 are distributed on each side of the polygon.
  • the light-receiving surface 221 of the photoelectric conversion element 22 is circular, and a plurality of taps 23 are distributed around the center of the light-receiving surface 221.
  • the multiple taps 23 have the same sampling duration, and the sampling duration is the same as the pulse width T0 of the optical pulse signal.
  • the closing moment of the previous tap 23 is the opening moment of the next tap 23.
  • a plurality of taps 23 are sequentially turned on.
  • the tap 23 that is turned on earlier in the first sub-period T1 has a lower turn-on frequency in the frame period, and the tap 23 that is turned on later in the first sub-period T1 is used in the frame period T2. The higher the opening frequency within.
  • the processing circuit 30 may be used to calculate that each tap 23 is in a frame period based on all the electrical signals obtained by each tap 23 in a frame period T2. The obtained total electrical signal; and calculating the depth information of the shot scene based on the multiple total electrical signals.
  • the processing circuit 30 may be used to determine a predetermined number of target signals from a plurality of total signals, and the predetermined number of target signals is determined by the predetermined number adjacent to each other in the turn-on sequence.
  • One tap 23 is obtained; and depth information is calculated according to a predetermined number of target signals.
  • the predetermined number has a value of three
  • determining the predetermined number of target signals from the multiple total signals includes: normalizing the multiple total signals to obtain multiple normalized signals; Select the normalized signal with the largest amplitude from the multiple normalized signals as the target signal; and select two normalized signals adjacent to the normalized signal with the largest amplitude from the multiple normalized signals As the target signal.
  • an embodiment of the present application also provides an electronic device 1000.
  • the electronic device 1000 includes a housing 200, and the depth camera 100 is combined with the housing 200.
  • the depth camera 100 includes a light emitting module 10, a light receiving module 20 and a processing circuit 30.
  • the light emitting module 10 is used to send one optical pulse signal in one sub-period T1, and send multiple optical pulse signals in one frame period T2, and the frame period T2 includes multiple consecutive sub-periods T1.
  • the light receiving module 20 includes pixels 21.
  • the pixels 21 include multiple taps 23.
  • the multiple taps 23 are used to obtain multiple electrical signals generated by light pulse signals and/or ambient light signals.
  • the opening time of any two taps 23 is different. They overlap, and the turn-on frequencies of any two taps 23 in the frame period T2 are different.
  • the processing circuit 30 is used to calculate the depth information of the shot scene according to a plurality of electrical signals.
  • the pixels 21 each include a photoelectric conversion element 22, each tap 23 includes a transfer gate 231, the photoelectric element 22 is used to receive light pulse signals and/or ambient light Signal to obtain an electrical signal, and the transfer gate 231 is used to transfer the electrical signal
  • the light-receiving surface 221 of the photoelectric conversion element 22 is a polygon, and each side of the polygon is distributed with one or more taps 23.
  • the light-receiving surface 221 of the photoelectric conversion element 22 is circular, and a plurality of taps 23 are distributed around the center of the light-receiving surface 221.
  • the multiple taps 23 have the same sampling duration, and the sampling duration is the same as the pulse width T0 of the optical pulse signal.
  • the closing moment of the previous tap 23 is the opening moment of the next tap 23.
  • a plurality of taps 23 are sequentially turned on.
  • the tap 23 that is turned on earlier in the first sub-period T1 has a lower turn-on frequency in the frame period, and the tap 23 that is turned on later in the first sub-period T1 is used in the frame period T2. The higher the opening frequency within.
  • the processing circuit 30 may be used to calculate that each tap 23 is in a frame period based on all the electrical signals obtained by each tap 23 in a frame period T2. The obtained total electrical signal; and calculating the depth information of the shot scene based on the multiple total electrical signals.
  • the processing circuit 30 may be used to determine a predetermined number of target signals from a plurality of total signals, and the predetermined number of target signals is determined by the predetermined number adjacent to each other in the turn-on sequence.
  • One tap 23 is obtained; and depth information is calculated according to a predetermined number of target signals.
  • the predetermined number has a value of three
  • determining the predetermined number of target signals from the multiple total signals includes: normalizing the multiple total signals to obtain multiple normalized signals; Select the normalized signal with the largest amplitude from the multiple normalized signals as the target signal; and select two normalized signals adjacent to the normalized signal with the largest amplitude from the multiple normalized signals As the target signal.
  • control method of the embodiment of the present application can be used for the depth camera 100 described in any one of the above embodiments.
  • Control methods include:
  • the optical transmitting module 10 transmits one optical pulse signal in one sub-period, and transmits multiple optical pulse signals in one frame period, and the frame period includes multiple consecutive sub-periods;
  • Multiple taps 23 acquire multiple electrical signals generated by light pulse signals and/or ambient light signals, the turn-on times of any two taps 23 do not overlap, and the turn-on frequencies of any two taps 23 in the frame period are different;
  • step 03 calculates the depth information of the shot scene based on multiple electrical signals, including: calculating the total electrical signal obtained by each tap 23 in one frame period; and according to the multiple total electrical signals.
  • the electrical signal calculates the depth information.
  • the step of calculating depth information based on multiple total electrical signals includes: calculating each tap 23 based on all electrical signals obtained by each tap 23 within a frame period T2 The total electrical signal obtained in one frame period T2; and the depth information is calculated based on the multiple total electrical signals.
  • the step of calculating depth information based on multiple total electrical signals includes: determining a predetermined number of target signals from the multiple total signals, and the predetermined number of target signals are switched on in order A predetermined number of adjacent taps 23 are obtained; and depth information is calculated based on the predetermined number of target signals.
  • the predetermined number is three
  • the step of determining the predetermined number of target signals from the multiple total signals includes: normalizing the multiple total signals to obtain multiple normalized signals; Selecting the normalized signal with the largest amplitude from the plurality of normalized signals as the target signal; and selecting the normalized signal with the largest amplitude from the plurality of normalized signals Two adjacent normalized signals are used as the target signal.
  • the predetermined number is three
  • the step of determining the predetermined number of target signals from the multiple total signals includes: normalizing the multiple total signals to obtain multiple normalized signals Calculate the sum value of any two adjacent normalized signals according to the turn-on sequence of multiple taps 23; select the largest sum value from the multiple sum values, and normalize the two that make up the sum value
  • the normalized signal is determined as the target signal; the normalized signal adjacent to any one of the two target signals among the plurality of normalized signals is determined as the target signal according to the turn-on sequence of the plurality of taps 23.
  • the depth camera 100 includes a light emitting module 10, a light receiving module 20 and a processing circuit 30.
  • the light emitting module 10 is used to send one optical pulse signal in one sub-period T1, and send multiple optical pulse signals in one frame period T2, and the frame period T2 includes multiple consecutive sub-periods T1.
  • the light receiving module 20 includes pixels 21.
  • the pixels 21 include multiple taps 23.
  • the multiple taps 23 are used to obtain multiple electrical signals generated by light pulse signals and/or ambient light signals.
  • the opening time of any two taps 23 is different. They overlap, and the turn-on frequencies of any two taps 23 in the frame period T2 are different.
  • the processing circuit 30 is used to calculate the depth information of the shot scene according to a plurality of electrical signals.
  • the light pulse signal emitted by the light emitting module 10 may be invisible light, such as infrared light, ultraviolet light, and the like. As shown in FIG. 6, the pulse width of the light pulse signal emitted by the light emitting module 10 is T0.
  • the light emitting module 10 emits an optical pulse signal with a pulse width of T0 in a sub-period T1, and the sub-period T1 is expressed as the interval between the emission moments of any two adjacent optical pulse signals and the difference between one optical pulse signal The sum of pulse width T0.
  • a plurality of consecutive sub-periods T1 constitute a frame period T2, and the frame period T2 is expressed as a period for the depth camera 100 to obtain a frame of depth image (consisting of multiple depth information), wherein the light emitting module 10 is in a frame period T2
  • a plurality of optical pulse signals with a pulse width of T0 will be emitted, and the interval between the emission moments of any two adjacent optical pulse signals is the same.
  • the pixels 21 in the light receiving module 20 can collect light pulse signals and/or ambient light signals. Specifically, when the brightness of the shot scene is low, the pixels 21 in the light receiving module 20 may only be able to collect light pulse signals. When the photographed scene has a certain brightness, the light receiving module 20 may collect the light pulse signal and the ambient light signal at the same time, or it may only collect the ambient light signal.
  • pixels 21 with multiple taps 23 can be used to collect light.
  • the intensity of the light reflected back by the subject collected by the tap 23 will be stronger, and signal saturation is prone to occur; and when the distance of the subject is relatively long, the tap 23 collects The intensity of the light returned by the subject will be weaker.
  • the part generated by the light pulse signal accounts for a relatively small proportion, and the signal-to-noise ratio is low. Use this electrical signal for depth information Calculations will affect the measurement accuracy of depth information (it can also be understood as depth resolution, which is used to measure the depth camera's ability to distinguish distances).
  • the pixels 21 in the light receiving module 20 are provided with multiple taps 23, and the multiple taps 23 are turned on at different turn-on frequencies within the frame period.
  • the saturation of the tap 23 used to measure short-distance subjects can be avoided to the greatest extent, and at the same time, it can be ensured that the long-distance subjects can be maximized.
  • the light is collected ground to obtain an electrical signal with a high signal-to-noise ratio. The calculation of depth information based on electrical signals with high signal-to-noise ratio is beneficial to improve the measurement accuracy of depth information.
  • the number of pixels 21 in the light receiving module 20 may be one or more. In the specific embodiment of the present application, the number of pixels 21 is multiple.
  • Each pixel 21 includes a photoelectric conversion element 22 and a plurality of taps 23, wherein the number of taps 23 can be four, five, six, seven, eight, ten, twenty, fifty. Wait, there is no restriction here.
  • the photoelectric conversion element 22 is used to receive light pulse signals and/or ambient light signals to obtain electrical signals.
  • Each tap 23 includes a charge transfer gate 231 (Transfer gate, TG), a charge storage gate 233 (Storage gate, SG), and a floating diffusion region 235 (Floating Diffusion, FD).
  • the charge transfer gate 231 is used to transfer the charge generated by the photoelectric conversion element 22 after receiving light to the charge storage gate 233 for storage.
  • the floating diffusion 235 is connected to the readout circuit (not shown) of the pixel 21.
  • the charge storage gate The charge in the pole 233 can be read by the readout circuit through the floating diffusion 235.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is a polygon, and one or more taps 23 are distributed on each side of the polygon.
  • the polygon can be, for example, a triangle, a quadrilateral, a pentagon, a hexagon, an octagon, a twelve-sided, a hexadecagon, etc., which is not limited here.
  • the polygon can be a regular polygon or a non-regular polygon, and there is no restriction here.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is a regular polygon.
  • the electric field distribution of the photoelectric conversion element 22 whose light-collecting surface 221 is a regular polygon is relatively symmetric, which can increase the charge transfer rate.
  • the light-receiving surface 221 of the photoelectric conversion element 22 is square. Among the four sides of the square, two taps 23 are distributed on each side, and the two taps 23 distributed on each side can be symmetrically arranged with respect to the vertical bisector of the side. The eight taps 23 have a symmetrical structure up and down and left and right.
  • the electric field distribution of the photoelectric conversion element 22 is relatively symmetric, which can increase the charge transfer rate.
  • the number of taps 23 on each side can also be one, three, five, etc., which is not limited here.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is a regular hexagon. Among the six sides of the regular hexagon, a tap 23 is distributed on each side, and each tap 23 can be located at the center of the corresponding side.
  • the six taps 23 have a symmetrical structure up and down and left and right.
  • the light-receiving surface 221 of the pixel 21 of FIG. The square-shaped photoelectric conversion element 22 has a high degree of symmetry of electric field distribution and a high charge transfer rate. It should be noted that, in the pixel 21 shown in FIG. 3, the number of taps 23 on each side can also be two, three, five, etc., which is not limited here.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is a regular octagon.
  • a tap 23 is distributed on each side, and each tap 23 can be located at the center of the corresponding side.
  • the eight taps 23 have a symmetrical structure up and down and left and right.
  • the light-receiving surface 221 of the pixel 21 of FIG. The symmetry of the electric field distribution of the photoelectric conversion element 22 having a regular hexagon is high, and the charge transfer rate is also high.
  • the number of taps 23 on each side can also be two, three, five, etc., which is not limited here.
  • the light-collecting surface 221 of the photoelectric conversion element 22 is circular.
  • a plurality of taps 23 are distributed around the center of the light receiving surface 221.
  • a plurality of taps 23 are evenly distributed around the center of the light receiving surface 221.
  • the symmetry of the electric field distribution of the photoelectric conversion element 22 with a circular light-receiving surface 221 is higher than the symmetry of the electric field distribution of the photoelectric conversion element 22 with a polygonal light-receiving surface 221, and the charge transfer rate is also higher.
  • the multiple taps 23 have the same sampling duration, and the sampling duration is the same as the pulse width T0 of the optical pulse signal.
  • the closing moment of the previous tap 23 is the opening moment of the next tap 23.
  • a plurality of taps 23 are sequentially turned on.
  • the tap 23 that is turned on earlier in the first sub-period T1 has a lower turn-on frequency in the frame period, and the tap 23 that is turned on later in the first sub-period T1 is used in the frame period T2. The higher the opening frequency within.
  • the pixel 21 has eight taps 23, which are tap 1, tap 2, tap 3, tap 4, tap 5, tap 6, tap 7, and tap 8.
  • the frame period of the optical pulse signal emitted by the light emitting module 10 is T2, and the sub period is T1.
  • the pulse width of one optical pulse signal is T0.
  • the sampling duration of each tap 23 is T0.
  • each sub-period T1 constitutes a frame period T2.
  • the closing time of the previous tap 23 is the opening time of the next tap 23.
  • taps 1 to 8 are sequentially turned on.
  • the opening time of the tap 1 is the same as the time when the light emitting module 10 starts to emit the light pulse signal
  • the closing time of the tap 1 is the same as the time when the light emitting module 10 stops emitting the light pulse signal.
  • the opening moment of tap 2 is the same as the closing moment of tap 1.
  • the opening moment of tap 3 is the same as the closing moment of tap 2.
  • the opening moment of tap 4 is the same as the closing moment of tap 3.
  • the opening moment of tap 8 is the same as the closing moment of tap 7.
  • tap 2 and tap 8 are turned on sequentially.
  • the opening moment of the tap 2 is the same as the moment when the light emitting module 10 stops emitting light pulse signals.
  • the opening moment of tap 3 is the same as the closing moment of tap 2.
  • the opening moment of tap 4 is the same as the closing moment of tap 3.
  • the opening moment of tap 8 is the same as the closing moment of tap 7.
  • the turn-on time of the n-th tap 23 in the N-th sub-period T1 satisfies the following rule: the turn-on time of the n-th tap 23 is relative to the emission time of the optical pulse signal of the N-th sub-period T1. (n-1) Delay of T0.
  • the number of opened taps 23 is 8, and the eight opened taps 23 are tap 1 to tap 8, respectively.
  • the number of taps 23 that are turned on is seven, and the seven taps 23 that are turned on are tap 2 to tap 8, respectively.
  • the number of opened taps 23 is 6, and the opened 6 taps 23 are tap 3 to tap 8, respectively.
  • the number of opened taps 23 is five, and the five opened taps 23 are tap 4 to tap 8, respectively.
  • the number of opened taps 23 is 4, and the 4 opened taps 23 are tap 5 to tap 8, respectively.
  • the number of opened taps 23 is three, and the three opened taps 23 are tap 6 to tap 8, respectively.
  • the number of opened taps 23 is two, and the two opened taps 23 are tap 7 and tap 8, respectively.
  • the number of taps 23 opened is one, and one tap 23 opened is tap 8. That is to say, in each frame period T2, the tap 23 with a shorter delay time at the turn-on time relative to the transmission timing of the optical pulse signal of the sub-period T1, the lower the turn-on frequency in the frame period T1. The longer the delay time of the tap 23 at the turn-on time relative to the transmission timing of the optical pulse signal of the sub-period T1, the higher the turn-on frequency in the frame period T2.
  • the taps 1 to 8 are turned on in sequence.
  • the tap 23 that is turned on earlier can be used to measure the object that is closer to the depth camera 100, and the tap 23 that is turned on later can be used to measure the distance to the depth camera 100.
  • the energy of the light will attenuate more, resulting in the higher the energy of the light that can be received by the tap 23 of the subject that measures the closer distance, and the signal is likely to saturate.
  • the lower the energy of the light that can be received by the tap 23 of the object the lower the signal-to-noise ratio of the obtained electrical signal.
  • the depth camera 100 can minimize the saturation of the tap 23 used to measure a short-distance subject through such a depth-adaptive tap 23 opening frequency allocation operation, and at the same time, it can ensure that the tap 23 is used to measure far distances.
  • the subject at a distance can maximize the collection of light, so as to obtain an electrical signal with a high signal-to-noise ratio.
  • the calculation of depth information based on electrical signals with high signal-to-noise ratio is beneficial to improve the measurement accuracy of depth information.
  • the processing circuit 30 may be used to calculate that each tap 23 is in a frame period based on all the electrical signals obtained by each tap 23 in a frame period T2. Obtain the total electrical signal, and calculate the depth information of the shot scene based on the multiple total electrical signals.
  • the electrical signal obtained by tap 8 in one frame period includes Q81 (obtained in the first sub-period T1), Q82 (obtained in the second sub-period T1), and Q83 (obtained in the third sub-period T1).
  • the processing circuit 30 may be used to determine a predetermined number of target signals from a plurality of total signals, and calculate depth information according to the predetermined number of target signals.
  • a predetermined number of target signals are obtained from a predetermined number of taps 23 adjacent to each other in the turn-on sequence.
  • the predetermined number has a value of three, and the three target signals are obtained by three adjacent taps 23 in the turn-on sequence.
  • the processing circuit 30 When the processing circuit 30 is used to determine three target signals from multiple total signals, the processing circuit 30 is mainly used to normalize the multiple total signals to obtain multiple normalized signals, and according to the multiple taps 23 Calculate the sum of any two adjacent normalized signals in the opening sequence, then select the largest sum from multiple sums, and determine the two normalized signals that make up the sum as the target signal. Finally, the normalized signal adjacent to any one of the two target signals among the plurality of normalized signals is determined as the target signal according to the turn-on sequence of the plurality of taps 23.
  • the processing circuit 30 first performs normalization processing on a plurality of total signals to obtain a plurality of normalized signals.
  • the normalized signal Q1' of tap 1 Q1/1.
  • the normalized signal Q2' of tap 2 Q2/2.
  • the normalized signal Q3' of tap 3 Q3/3.
  • the signal-to-noise ratio of the normalized signal of any tap 23 is greater than the signal-to-noise ratio of the electrical signal obtained by the tap 23 in any sub-period T1.
  • the processing circuit 30 selects three normalized signals from the eight normalized signals Q1', Q2', Q3', Q4', Q5', Q6', Q7', and Q8' as the target signal.
  • the processing circuit 30 can calculate the sum of any two adjacent normalized signals (the sum of the amplitudes of the normalized signals), where the adjacent normalized signals indicate that the two normalized signals are switched on in order Two adjacent taps 23 are obtained.
  • the processing circuit 30 calculates the sum value S12 of Q1' and Q2', calculates the sum value S23 of Q2' and Q3', calculates the sum value S34 of Q3' and Q4', calculates the sum value S45 of Q4' and Q5', and calculates The sum of Q5' and Q6' is S56, the sum of Q6' and Q7' is calculated S67, and the sum of Q7' and Q8' is calculated S78. Finally, the processing circuit 30 selects the largest sum value from the 6 sum values of S12, S23, S34, S45, S56, S67, and S78.
  • the processing circuit 30 selects the normalized signal adjacent to any one of the two target signals as the target signal, and the normalized signal adjacent to the normalized signal Q3' is Q2', and the normalized signal The normalized signal adjacent to Q4' is Q5', and the processing circuit 30 can select the normalized signal Q2' or the normalized signal Q5' as the target signal.
  • the number 2 in parentheses represents the coefficient corresponding to the delay time existing at the time when the tap 3 is turned on relative to the time when the light pulse signal is emitted.
  • the processing circuit 30 selects the normalized signal Q5' as the target signal, the depth information Among them, the number 2 in parentheses represents the coefficient corresponding to the delay time existing at the time when the tap 3 is turned on relative to the time when the light pulse signal is emitted.
  • the farthest distance that the depth camera 100 can measure The ranging range is [0, 3c ⁇ T0]. Compared with pixels with only three taps (the ranging range of pixels with three taps is ), a greater number of taps 23 are provided in the depth camera 100 of the embodiment of the present application, so that the ranging range of the depth camera 100 is improved. In addition, the depth camera 100 of the embodiment of the present application subtracts the electrical signals obtained by the multiple taps 23 to eliminate the interference of inherent noise and the ambient light signal, and there is no need to introduce an additional sampling window for the ambient light signal.
  • the processing circuit 30 may also select a normalized signal with the largest amplitude from the multiple normalized signals, for example The normalized signal with the largest amplitude is Q(n+1)', then the processing circuit 30 first determines the two normalized signals Qn' and the normalized signal adjacent to the normalized signal Q(n+1)' The magnitude of the signal Q(n+2)'.
  • a larger number of taps 23 are provided in the depth camera 100 of the embodiment of the present application, so that the ranging range of the depth camera 100 is improved.
  • the depth camera 100 of the embodiment of the present application subtracts the electrical signals obtained by the multiple taps 23 to eliminate the interference of inherent noise and the ambient light signal, and there is no need to introduce an additional sampling window for the ambient light signal.
  • the predetermined number of values may also be four, and the four target signals are obtained by four taps 23 adjacent to each other in the turn-on sequence.
  • the processing circuit 30 is mainly used to normalize the multiple total signals to obtain multiple normalized signals, and according to the multiple taps 23 Calculate the sum of any two adjacent normalized signals in the opening sequence, then select the largest sum from multiple sums, and determine the two normalized signals that make up the sum as the target signal. Finally, two normalized signals adjacent to the two target signals among the plurality of normalized signals are determined as target signals according to the turn-on sequence of the plurality of taps 23.
  • the processing circuit 30 first performs normalization processing on a plurality of total signals to obtain a plurality of normalized signals.
  • the normalized signal Q1' of tap 1 Q1/1.
  • the normalized signal Q2' of tap 2 Q2/2.
  • the normalized signal Q3' of tap 3 Q3/3.
  • the signal-to-noise ratio of the normalized signal of any tap 23 is greater than the signal-to-noise ratio of the electrical signal obtained by the tap 23 in any sub-period T1.
  • the processing circuit 30 selects three normalized signals from the eight normalized signals Q1', Q2', Q3', Q4', Q5', Q6', Q7', and Q8' as the target signal.
  • the processing circuit 30 can calculate the sum of any two adjacent normalized signals (the sum of the amplitudes of the normalized signals), where the adjacent normalized signals indicate that the two normalized signals are switched on in order Two adjacent taps 23 are obtained.
  • the processing circuit 30 calculates the sum value S12 of Q1' and Q2', calculates the sum value S23 of Q2' and Q3', calculates the sum value S34 of Q3' and Q4', calculates the sum value S45 of Q4' and Q5', and calculates The sum of Q5' and Q6' is S56, the sum of Q6' and Q7' is calculated S67, and the sum of Q7' and Q8' is calculated S78. Finally, the processing circuit 30 selects the largest sum value from the 6 sum values of S12, S23, S34, S45, S56, S67, and S78.
  • the processing circuit 30 selects the two normalized signals adjacent to the two target signals as the target signal, the normalized signal adjacent to the normalized signal Q3' is Q2', and the normalized signal Q4' The adjacent normalized signal is Q5', and the processing circuit 30 can select the normalized signal Q2' and the normalized signal Q5' as the target signal.
  • the number 2 in parentheses represents the coefficient corresponding to the delay time existing at the time when the tap 3 is turned on relative to the time when the light pulse signal is emitted.
  • the method of selecting four target signals for depth information calculation when the part generated by the ambient light signal in the normalized signal Q3' of tap 3 is eliminated, the reduction What goes is the normalized signal Q2' of tap 2 instead of the normalized signal Q5' of tap 5. It is understandable that tap 2 and tap 3 are two taps 23 adjacent to each other in the turn-on sequence, and the normalized signal of the two is more similar in the amount of the part generated by the ambient light signal, so that tap 3 can be removed more accurately.
  • the ambient light component in the normalized signal Q3' helps to obtain more accurate depth information.
  • the processing circuit 30 may also select a normalized signal with the largest amplitude from the multiple normalized signals as the target signal. , And select the other three adjacent to this normalized signal (including the two directly adjacent and the indirectly adjacent one, between the indirectly adjacent normalized signal and the normalized signal with the largest amplitude A normalized signal is spaced apart) the normalized signal is used as the target signal. For example, the normalized signal with the largest amplitude is Q(n+1)', then the processing circuit 30 first judges the two normalized signals Qn' and the normalized signal directly adjacent to the normalized signal Q(n+1)' The size of the signal Q(n+2)' is changed.
  • the processing circuit 30 determines that the normalized signal Q(n+2)' is generated by the ambient light signal, and the normalized signal Qn' and the normalized signal Q( n+1)' is jointly generated by the light pulse signal and the ambient light signal. Subsequently, the processor 30 determines the normalized signal Qn', the normalized signal Q(n+1)', and the normalized signal Q(n+2)' as the target signal, and then selects the target signal from the remaining normalized signal The normalized signal Q(n-1)' directly adjacent to the normalized signal Qn' is selected.
  • the processor 30 determines the normalized signal Qn', the normalized signal Q(n+1)', and the normalized signal Q(n+2)' as the target signal, and then selects the target signal from the remaining normalized signal Select the normalized signal Q(n+3)' directly adjacent to the normalized signal Q(n+2)'.
  • the part generated by the ambient light signal in the normalized signal Qn' of the n-th tap 23 is eliminated
  • the normalized signal Q(n-1)' of the n-1th tap 23 instead of the normalized signal Q(n+2)' of the n+2th tap 23.
  • the n-1th tap 23 and the nth tap 23 are two adjacent taps 23 in the turn-on sequence, and the normalized signal of the two is more similar in quantity generated by the ambient light signal. Therefore, the ambient light component in the normalized signal Qn′ of the n-th tap 23 can be removed more accurately, which is beneficial to obtain more accurate depth information.
  • the normalized signal of the n+3 tap 23 is subtracted Q(n+3)' instead of the normalized signal Qn' of the n-th tap 23.
  • the n+2th tap 23 and the n+3th tap 23 are two adjacent taps 23 in the turn-on sequence, and the normalized signal of the two has a larger amount of the part generated by the ambient light signal. Therefore, the ambient light component in the normalized signal Q(n+2)′ of the n+2th tap 23 can be removed more accurately, which is beneficial to obtain more accurate depth information.
  • the value of the predetermined number can also be five, six, seven, eight, ten, etc., which is not limited here. Except for the two target signals generated by the ambient light signal and the light pulse signal at the same time among the multiple target signals, the rest of the target signals are all generated by the ambient light signal, and the processing circuit 30 can respond to the multiple target signals generated by the ambient light signal. The average value is taken, and the average value can be used to remove the ambient light component in the two target signals generated by the ambient light signal and the light pulse signal at the same time.
  • an embodiment of the present application also provides an electronic device 1000.
  • the electronic device 1000 includes a housing 200 and the depth camera 100 described in any one of the above embodiments.
  • the depth camera 100 is combined with the housing 200.
  • the depth camera 100 may be arranged on the side of the housing 200 where the display screen is installed; in another example, the depth camera 100 may be arranged on the other side of the housing 200 that is opposite to the side where the display screen is installed.
  • the electronic device 1000 may be a mobile phone, a tablet computer, a notebook computer, a smart wearable device (smart watch, smart bracelet, smart glasses, smart helmet, etc.), a virtual reality device, etc., which is not limited here.
  • the electronic device 1000 of the embodiment of the present application is equipped with a depth camera 100.
  • the pixels 21 in the depth camera 100 are provided with multiple taps 23, and the multiple taps 23 are turned on at different turn-on frequencies within a frame period.
  • the saturation of the tap 23 used to measure short-distance subjects can be avoided to the greatest extent, and at the same time, it can be ensured that the long-distance subjects can be maximized.
  • the light is collected ground to obtain an electrical signal with a high signal-to-noise ratio.
  • the calculation of depth information based on electrical signals with high signal-to-noise ratio is beneficial to improve the measurement accuracy of depth information.
  • control method of the embodiment of the present application can be used for the depth camera 100 described in any one of the above embodiments.
  • Control methods include:
  • the optical transmitting module 10 transmits one optical pulse signal in one sub-period, and transmits multiple optical pulse signals in one frame period, and the frame period includes multiple consecutive sub-periods;
  • Multiple taps 23 acquire multiple electrical signals generated by light pulse signals and/or ambient light signals, the turn-on times of any two taps 23 do not overlap, and the turn-on frequencies of any two taps 23 in the frame period are different;
  • step 03 calculates the depth information of the shot scene according to multiple electrical signals, including:
  • the step of calculating depth information based on multiple total electrical signals includes:
  • the step of calculating depth information based on multiple total electrical signals includes:
  • a predetermined number of target signals are determined from a plurality of total signals, and the predetermined number of target signals are obtained by a predetermined number of taps 23 adjacent to each other in the turn-on sequence;
  • the depth information is calculated according to a predetermined number of target signals.
  • the predetermined number is three
  • the step of determining the predetermined number of target signals from the plurality of total signals includes:
  • Two normalized signals adjacent to the normalized signal with the largest amplitude are selected from the plurality of normalized signals as the target signal.
  • the predetermined number is three
  • the step of determining the predetermined number of target signals from a plurality of total signals includes:
  • the normalized signal adjacent to any one of the two target signals among the plurality of normalized signals is determined as the target signal.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Optics & Photonics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Optical Radar Systems And Details Thereof (AREA)
  • Measurement Of Optical Distance (AREA)
  • Studio Devices (AREA)

Abstract

一种深度相机(100)、电子设备(1000)及控制方法。深度相机(100)包括光发射模组(10)、光接收模组(20)及处理电路(30)。光发射模组(10)用于在一个子周期(T1)内发送一个光脉冲信号。光接收模组(20)中的像素(21)包括多个抽头(23)，多个抽头(23)获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头(23)的开启时间不重叠，且任意两个抽头(23)在帧周期(T2)内的开启频率不同。处理电路(30)用于根据多个电信号计算被摄场景的深度信息。

Description

深度相机、电子设备及控制方法
优先权信息
本申请请求2020年5月29日向中国国家知识产权局提交的、专利申请号为202010472386.1的专利申请的优先权和权益，并且通过参照将其全文并入此处。
技术领域
本申请涉及深度测量技术领域，特别涉及一种深度相机、电子设备及控制方法。
背景技术
脉冲调制式I-TOF(Indirect-Time of Flight,I-TOF)三维测距技术是利用对反射激光脉冲延迟的间接测量来获取深度数据的。与连续调制式I-TOF相比，脉冲调制式I-TOF具有占空比低、节省功耗等优势。
发明内容
本申请实施方式提供了一种深度相机、电子设备及控制方法。
本申请实施方式的深度相机包括光发射模组、光接收模组及处理电路。所述光发射模组用于在一个子周期内发送一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所述帧周期包括多个连续的所述子周期。所述光接收模组包括像素，所述像素包括多个抽头，所述多个抽头获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同。所述处理电路用于根据所述多个电信号计算被摄场景的深度信息。
本申请实施方式的电子设备包括壳体及深度相机。所述深度相机与所述壳体结合。所述深度相机包括光发射模组、光接收模组及处理电路。所述光发射模组用于在一个子周期内发送一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所述帧周期包括多个连续的所述子周期。所述光接收模组包括像素，所述像素包括多个抽头，所述多个抽头获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同。所述处理电路用于根据所述多个电信号计算被摄场景的深度信息。
本申请实施方式的控制方法用于深度相机。所述深度相机包括光发射模组及光接收模组。所述光接收模组包括像素，所述像素包括多个抽头。所述控制方法包括：所述光发射模组在一个子周期内发射一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所 述帧周期包括多个连续的所述子周期；所述多个抽头获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同；及根据所述多个电信号计算被摄场景的深度信息。
本申请实施方式的附加方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实践了解到。
附图说明
本申请的上述和/或附加的方面和优点可以从结合下面附图对实施方式的描述中将变得明显和容易理解，其中：
图1是本申请某些实施方式的深度相机的示意图；
图2是本申请某些实施方式的深度相机中像素的结构示意图；
图3是本申请某些实施方式的深度相机中像素的结构示意图；
图4是本申请某些实施方式的深度相机中像素的结构示意图；
图5是本申请某些实施方式的深度相机中像素的结构示意图；
图6是本申请某些实施方式的深度相机中的工作时序示意图；
图7是本申请某些实施方式的电子设备的示意图；
图8是本申请某些实施方式的控制方法的流程示意图。
具体实施方式
下面详细描述本申请的实施方式，所述实施方式的示例在附图中示出，其中，相同或类似的标号自始至终表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施方式是示例性的，仅用于解释本申请的实施方式，而不能理解为对本申请的实施方式的限制。
请参阅图1、图2和图6，本申请实施方式提供一种深度相机100。深度相机100包括光发射模组10、光接收模组20及处理电路30。光发射模组10用于在一个子周期T1内发送一个光脉冲信号，在一个帧周期T2内发送多个光脉冲信号，帧周期T2包括多个连续的子周期T1。光接收模组20包括像素21，像素21包括多个抽头23，多个抽头23用于获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头23的开启时间不重叠，且任意两个抽头23在帧周期T2内的开启频率不同。处理电路30用于根据多个电信号计算被摄场景的深度信息。
请参阅图1和图2，在某些实施方式中，像素21均包括一个光电转换元件22，每个抽头23包括一个转移栅极231，光电元件22用于接收光脉冲信号和/或环境光信号以获得电 信号，转移栅极231用于转移电信号。
请参阅图2至图4，在某些实施方式中，光电转换元件22的收光面221为多边形，多边形的每条边上分布有一个或多个抽头23。
请参阅图5，在某些实施方式中，光电转换元件22的收光面221为圆形，多个抽头23环绕收光面221的中心分布。
请参阅图2和图6，在某些实施方式中，多个抽头23具有相同的采样时长，且采样时长与光脉冲信号的脉冲宽度T0相同。在一个子周期T1内，前一个抽头23的关闭时刻为下一个抽头23的开启时刻。
在某些实施方式中，在每个帧周期T2内的第一个子周期T1中，多个抽头23依次序开启。在每个帧周期T2内，在第一个子周期T1中越先开启的抽头23在该帧周期内的开启频率越低，在第一个子周期T1中越晚开启的抽头23在该帧周期T2内的开启频率越高。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于根据每个抽头23在一个帧周期T2内获得的所有电信号计算每个抽头23在一个帧周期内获得的总电信号；及根据多个总电信号计算被摄场景的深度信息。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于从多个总信号中确定预定数量个目标信号，预定数量个目标信号由开启顺序相邻的预定数量个抽头23获得；及根据预定数量个目标信号计算深度信息。
在某些实施方式中，预定数量的取值为三，从多个总信号中确定预定数量个目标信号，包括：对多个总信号进行归一化处理以获得多个归一化信号；从多个归一化信号中选出幅值最大的归一化信号作为目标信号；及从多个归一化信号中选出与幅值最大的归一化信号相邻的两个归一化信号作为目标信号。
请参阅图2和图7，本申请实施方式还提供一种电子设备1000。电子设备1000包括壳体200，深度相机100与壳体200结合。深度相机100包括光发射模组10、光接收模组20及处理电路30。光发射模组10用于在一个子周期T1内发送一个光脉冲信号，在一个帧周期T2内发送多个光脉冲信号，帧周期T2包括多个连续的子周期T1。光接收模组20包括像素21，像素21包括多个抽头23，多个抽头23用于获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头23的开启时间不重叠，且任意两个抽头23在帧周期T2内的开启频率不同。处理电路30用于根据多个电信号计算被摄场景的深度信息。
请参阅图1和图2，在某些实施方式中，像素21均包括一个光电转换元件22，每个抽头23包括一个转移栅极231，光电元件22用于接收光脉冲信号和/或环境光信号以获得电信号，转移栅极231用于转移电信号
请参阅图2至图4，在某些实施方式中，光电转换元件22的收光面221为多边形，多 边形的每条边上分布有一个或多个抽头23。
请参阅图5，在某些实施方式中，光电转换元件22的收光面221为圆形，多个抽头23环绕收光面221的中心分布。
请参阅图2和图6，在某些实施方式中，多个抽头23具有相同的采样时长，且采样时长与光脉冲信号的脉冲宽度T0相同。在一个子周期T1内，前一个抽头23的关闭时刻为下一个抽头23的开启时刻。
在某些实施方式中，在每个帧周期T2内的第一个子周期T1中，多个抽头23依次序开启。在每个帧周期T2内，在第一个子周期T1中越先开启的抽头23在该帧周期内的开启频率越低，在第一个子周期T1中越晚开启的抽头23在该帧周期T2内的开启频率越高。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于根据每个抽头23在一个帧周期T2内获得的所有电信号计算每个抽头23在一个帧周期内获得的总电信号；及根据多个总电信号计算被摄场景的深度信息。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于从多个总信号中确定预定数量个目标信号，预定数量个目标信号由开启顺序相邻的预定数量个抽头23获得；及根据预定数量个目标信号计算深度信息。
在某些实施方式中，预定数量的取值为三，从多个总信号中确定预定数量个目标信号，包括：对多个总信号进行归一化处理以获得多个归一化信号；从多个归一化信号中选出幅值最大的归一化信号作为目标信号；及从多个归一化信号中选出与幅值最大的归一化信号相邻的两个归一化信号作为目标信号。
请参阅图1、图2及图8，本申请实施方式还提供一种控制方法。本申请实施方式的控制方法可以用于上述任意一个实施方式所述的深度相机100。控制方法包括：
01：光发射模组10在一个子周期内发射一个光脉冲信号，在一个帧周期内发送多个光脉冲信号，帧周期包括多个连续的子周期；
02：多个抽头23获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头23的开启时间不重叠，且任意两个抽头23在帧周期内的开启频率不同；及
03：根据多个电信号计算被摄场景的深度信息。
请参阅图2，在某些实施方式中，步骤03根据多个电信号计算被摄场景的深度信息，包括：计算每个抽头23在一个帧周期内获得的总电信号；及根据多个总电信号计算深度信息。
请参阅图2和图6，在某些实施方式中，根据多个总电信号计算深度信息的步骤，包括：根据每个抽头23在一个帧周期T2内获得的所有电信号计算每个抽头23在一个帧周期T2内获得的总电信号；及根据多个总电信号计算深度信息。
请参阅图2和图6，在某些实施方式中，根据多个总电信号计算深度信息的步骤，包括：从多个总信号中确定预定数量个目标信号，预定数量个目标信号由开启顺序相邻的预定数量个抽头23获得；及根据预定数量个目标信号计算深度信息。
在某些实施方式中，预定数量为三，从多个总信号中确定预定数量个目标信号的步骤，包括：对多个所述总信号进行归一化处理以获得多个归一化信号；从所述多个归一化信号中选出幅值最大的归一化信号作为所述目标信号；及从所述多个所述归一化信号中选出与所述幅值最大的归一化信号相邻的两个归一化信号作为所述目标信号。
请参阅图2，在某些实施方式中，预定数量为三，从多个总信号中确定预定数量个目标信号的步骤，包括：对多个总信号进行归一化处理以获得多个归一化信号；根据多个抽头23的开启顺序计算出任意两个相邻的归一化信号的和值；从多个和值中选取最大的和值，并将组成该和值的两个归一化信号确定为目标信号；根据多个抽头23的开启顺序确定多个归一化信号中的与两个目标信号中的任意一个相邻的归一化信号为目标信号。
下面结合附图做进一步说明。
请参阅图1、图2和图6，本申请实施方式提供一种深度相机100。深度相机100包括光发射模组10、光接收模组20及处理电路30。光发射模组10用于在一个子周期T1内发送一个光脉冲信号，在一个帧周期T2内发送多个光脉冲信号，帧周期T2包括多个连续的子周期T1。光接收模组20包括像素21，像素21包括多个抽头23，多个抽头23用于获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头23的开启时间不重叠，且任意两个抽头23在帧周期T2内的开启频率不同。处理电路30用于根据多个电信号计算被摄场景的深度信息。
其中，光发射模组10发射的光脉冲信号可以是不可见光，例如红外光、紫外光等。如图6所示，光发射模组10发射的光脉冲信号的脉冲宽度为T0。光发射模组10在一个子周期T1内会发射一个脉冲宽度为T0的光脉冲信号，子周期T1表示为任意两个相邻的光脉冲信号的发射时刻之间的间隔与一个光脉冲信号的脉冲宽度T0之和。多个连续的子周期T1组成一个帧周期T2，帧周期T2表示为深度相机100获取一帧深度图像(由多个深度信息组成)的周期，其中，光发射模组10在一个帧周期T2内会发射多个脉冲宽度为T0的光脉冲信号，任意两个相邻的光脉冲信号的发射时刻之间的间隔相同。
光接收模组20中的像素21可以采集光脉冲信号和/或环境光信号。具体地，当被摄场景的亮度较低时，光接收模组20中的像素21可能仅能够采集到光脉冲信号。当被摄场景具有一定亮度时，光接收模组20中可能同时采集到光脉冲信号和环境光信号，也可能仅采集到环境光信号。
可以理解，在使用基于飞行时间的深度相机100进行深度信息的测量时，可以采用具 有多个抽头23(也称为采样窗口)的像素21进行光线的采集。而当被摄物体的距离较近时，抽头23采集到的由被摄物体反射回的光线的强度会较强，容易发生信号饱和；而当被摄物体的距离较远时，抽头23采集到的由被摄物体返回的光线的强度会较弱，抽头23采集到的电信号中，由光脉冲信号生成的部分的占比较小，信噪比较低，使用这一电信号进行深度信息的计算，会影响深度信息的测量精度(也可理解为深度分辨率，深度分辨率用于衡量深度相机对距离的分辨能力)。
本申请实施方式的深度相机100中，光接收模组20内的像素21设置了多个抽头23，且多个抽头23在帧周期内以不同的开启频率开启。通过这种深度自适应的抽头23的开启频率分配操作，可以最大程度地避免用于测量近距离的被摄物体的抽头23的饱和，同时可以保证用于测量远距离的被摄物体能够最大化地采集到光线，从而获得信噪比较高的电信号。基于信噪比较高的电信号进行深度信息的计算，有利于提升深度信息的测量精度。
请继续参阅图1和图2，在某些实施方式中，光接收模组20中的像素21可以是一个或多个。在本申请的具体实施例中，像素21的数量为多个。每个像素21均包括一个光电转换元件22及多个抽头23，其中，抽头23的数量可以是四个、五个、六个、七个、八个、十个、二十个、五十个等等，在此不作限制。光电转换元件22用于接收光脉冲信号和/或环境光信号以获得电信号。每个抽头23包括电荷转移栅极231(Transfer gate，TG)、电荷存储栅极233(Storage gate，SG)及浮动扩散区235(Floating Diffusion，FD)。电荷转移栅极231用于将光电转换元件22接收光线后生成的电荷转移到电荷存储栅极233上进行存储，浮动扩散区235与像素21的读出电路(图未示)连接，电荷存储栅极233中的电荷可以通过浮动扩散区235被读出电路读出。
请参阅图2至图4，在某些实施方式中，光电转换元件22的收光面221为多边形，多边形的每条边上分布有一个或多个抽头23。多边形例如可以为三角形、四边形、五边形、六边形、八边形、十二边、十六边形等等，在此不作限制。多边形可以为正多边形或非正多边形，在此也不作限制。在本申请的一个实施例中，光电转换元件22的收光面221为正多边形。收光面221为正多边形的光电转换元件22的电场分布较为对称，可以提高电荷转移速率。
如图2所示，在一个例子中，光电转换元件22的收光面221为正方形。正方形的四条边中，每条边均分布有两个抽头23，每条边上分布的两个抽头23可以关于该条边的垂直平分线对称设置。八个抽头23呈上下及左右对称结构。图2所示的像素21中，由于光电转换元件22的收光面221为正方形，光电转换元件22的电场分布较为对称，可以提高电荷转移速率。需要说明的是，图2所示的像素21中，每条边上的抽头23的数量也可以是一个、三个、五个等，在此不作限制。
如图3所示，在另一个例子中，光电转换元件22的收光面221为正六边形。正六边形的六条边中，每条边均分布有一个抽头23，每个抽头23均可以位于对应边的中心位置。六个抽头23呈上下及左右对称结构。与图2的像素21相比，图3的像素21的收光面221为正六边形，收光面221为正六边形的光电转换元件22的电场分布的对称度相较于收光面221为正方形的光电转换元件22的电场分布的对称度来得高，电荷转移速率也较高。需要说明的是，图3所示的像素21中，每条边上的抽头23的数量也可以是两个、三个、五个等，在此不作限制。
如图4所示，在又一个例子中，光电转换元件22的收光面221为正八边形。正八边形的八条边中，每条边均分布有一个抽头23，每个抽头23均可以位于对应边的中心位置。八个抽头23呈上下及左右对称结构。与图3的像素21相比，图4的像素21的收光面221为正八边形，收光面221为正八边形的光电转换元件22的电场分布的对称度相较于收光面221为正六边形的光电转换元件22的电场分布的对称度来得高，电荷转移速率也较高。需要说明的是，图4所示的像素21中，每条边上的抽头23的数量也可以是两个、三个、五个等，在此不作限制。
请参阅图5，在某些实施方式中，光电转换元件22的收光面221为圆形。多个抽头23环绕收光面221的中心分布。在图5所示实施例中，多个抽头23环绕收光面221的中心均匀分布。收光面221为圆形的光电转换元件22的电场分布的对称度相较于收光面221为多边形的光电转换元件22的电场分布的对称度来得高，电荷转移速率也较高。
请参阅图2和图6，在某些实施方式中，多个抽头23具有相同的采样时长，且采样时长与光脉冲信号的脉冲宽度T0相同。在一个子周期T1内，前一个抽头23的关闭时刻为下一个抽头23的开启时刻。在每个帧周期T2内的第一个子周期T1中，多个抽头23依次序开启。在每个帧周期T2内，在第一个子周期T1中越先开启的抽头23在该帧周期内的开启频率越低，在第一个子周期T1中越晚开启的抽头23在该帧周期T2内的开启频率越高。
具体地，请结合图1、图2及图6，假设像素21具有八个抽头23，分别为抽头1、抽头2、抽头3、抽头4、抽头5、抽头6、抽头7及抽头8。光发射模组10发射光脉冲信号的帧周期为T2，子周期为T1，在每个子周期T1中，一个光脉冲信号的脉冲宽度为T0。那么，八个抽头23中，每个抽头23的采样时长(也即抽头23处于开启状态以获取电信号的时长)均为T0。
如图6所示，八个子周期T1组成一个帧周期T2。在每个子周期T1内，开启顺序相邻的两个抽头23中，前一个抽头23的关闭时刻为下一个抽头23的开启时刻。例如，在图6中的第一个子周期T1中，抽头1至抽头8依次序开启。其中，抽头1的开启时刻与光发射 模组10开始发射光脉冲信号的时刻相同，抽头1的关闭时刻与光发射模组10停止发射光脉冲信号的时刻相同。抽头2的开启时刻与抽头1的关闭时刻相同。抽头3的开启时刻与抽头2的关闭时刻相同。抽头4的开启时刻与抽头3的关闭时刻相同。以此类推，抽头8的开启时刻与抽头7的关闭时刻相同。再例如，在图6中的第二子周期T1中，抽头2和抽头8依次开启。抽头2的开启时刻与光发射模组10停止发射光脉冲信号的时刻相同。抽头3的开启时刻与抽头2的关闭时刻相同。抽头4的开启时刻与抽头3的关闭时刻相同。以此类推，抽头8的开启时刻与抽头7的关闭时刻相同。也即是说，第N个子周期T1内的第n个抽头23的开启时刻满足以下规律：该第n个抽头23的开启时刻相对于该第N个子周期T1的光脉冲信号的发射时刻存在有(n-1)T0的延迟。
如图6所示，第一个子周期T1内，开启的抽头23的数量为8个，开启的8个抽头23分别为抽头1至抽头8。第二个子周期T1内，开启的抽头23的数量为7个，开启的7个抽头23分别为抽头2至抽头8。第三个子周期T1内，开启的抽头23的数量为6个，开启的6个抽头23分别为抽头3至抽头8。第四个子周期内T1内，开启的抽头23的数量为5个，开启的5个抽头23分别为抽头4至抽头8。第五个子周期T1内，开启的抽头23的数量为4个，开启的4个抽头23分别为抽头5至抽头8。第六个子周期T1内，开启的抽头23的数量为3个，开启的3个抽头23分别为抽头6至抽头8。第七个子周期T1内，开启的抽头23的数量为2个，开启的2个抽头23分别为抽头7及抽头8。第八个子周期T1内，开启的抽头23的数量为1个，开启的1个抽头23为抽头8。也即是说，在每个帧周期T2内，开启时刻相对于该子周期T1的光脉冲信号的发射时刻存在的延迟时间越短的抽头23，在该帧周期T1内的开启频率越低。开启时刻相对于该子周期T1的光脉冲信号的发射时刻存在的延迟时间越长的抽头23，在该帧周期T2内的开启频率越高。
如图6所示，抽头1至抽头8依次序开启，越早开启的抽头23可以用于测量距离深度相机100越近的被摄物体，越晚开启的抽头23可以用于测量距离深度相机100越远的被摄物体。而由于随着飞行路程的增加，光线的能量衰减也会更多，导致测量越近距离的被摄物体的抽头23可以接收到的光线的能量越高，信号容易饱和，测量越远距离的被摄物体的抽头23可以接收到的光线的能量越低，获取的电信号的信噪比较差。
本申请实施方式的深度相机100通过这种深度自适应的抽头23的开启频率分配操作，可以最大程度地避免用于测量近距离的被摄物体的抽头23的饱和，同时可以保证用于测量远距离的被摄物体能够最大化地采集到光线，从而获得信噪比较高的电信号。基于信噪比较高的电信号进行深度信息的计算，有利于提升深度信息的测量精度。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于根据每个抽头23在一个帧周期T2内获得的所有电信号计算每个抽头23在一个帧周期内获得的总电信号， 并根据多个总电信号计算被摄场景的深度信息。
具体地，如图6所示，抽头1在一个帧周期T2内获得的电信号包括Q11(在第一个子周期T1内获得)，则抽头1的总电信号为Q1＝Q11。抽头2在一个帧周期T2内获得的电信号包括Q21(在第一个子周期T1内获得)和Q22(在第二个子周期T1内获得)，则抽头2的总电信号为Q2＝Q21+Q22。抽头3在一个帧周期T2内获得的电信号包括Q31(在第一个子周期T1内获得)、Q32(在第二个子周期T1内获得)及Q33(在第三个子周期T1内获得)，则抽头3的总电信号为Q3＝Q31+Q32+Q33。以此类推，抽头8在一个帧周期内获得的电信号包括Q81(在第一个子周期T1内获得)、Q82(在第二个子周期T1内获得)、Q83(在第三个子周期T1内获得)、Q84(在第四个子周期T1内获得)、Q85(在第五个子周期T1内获得)、Q86(在第六个子周期T1内获得)、Q87(在第七个子周期T1内获得)、Q88(在第八个子周期T1内获得)，则处理电路30计算到的抽头8的总电信号为Q8＝Q81+Q82+Q83+Q84+Q85+Q86+Q87+Q88。处理电路30获得多个抽头23的总信号后，即可根据多个总信号来计算被摄场景的深度信息。由于用于测量远距离物体的抽头23开启的频率较高，使得用于测量远距离物体的抽头23可以最大化地采集到光线，从而可以获得更多的电信号。
请参阅图1、图2和图6，在某些实施方式中，处理电路30可以用于从多个总信号中确定预定数量个目标信号，并根据预定数量个目标信号计算深度信息。其中，预定数量个目标信号由开启顺序相邻的预定数量个抽头23获得。在一个例子中，预定数量的取值为三，三个目标信号由开启顺序相邻的三个抽头23获得。在处理电路30用于从多个总信号中确定三个目标信号时，处理电路30主要用于对多个总信号进行归一化处理以获得多个归一化信号，并根据多个抽头23的开启顺序计算出任意两个相邻的归一化信号的和值，再从多个和值中选取最大的和值，并将组成该和值的两个归一化信号确定为目标信号，最后根据多个抽头23的开启顺序确定多个归一化信号中的与两个目标信号中的任意一个相邻的归一化信号为目标信号。
具体地，处理电路30首先对多个总信号进行归一化处理以获得多个归一化信号。例如，抽头1的归一化信号Q1’＝Q1/1。抽头2的归一化信号Q2’＝Q2/2。抽头3的归一化信号Q3’＝Q3/3。以此类推，抽头8的归一化信号Q8’＝Q8/8。也即抽头n的归一化信号Qn’＝Qn/n。任意一个抽头23的归一化信号的信噪比都大于该抽头23在任意一个子周期T1内获得的电信号的信噪比。
随后，处理电路30从Q1’、Q2’、Q3’、Q4’、Q5’、Q6’、Q7’、Q8’这八个归一化信号中选取三个归一化信号作为目标信号。处理电路30可以计算任意两个相邻的归一化信号的和值(归一化信号的幅值之和)，其中，相邻的归一化信号表示这两个归一化信号由开启顺 序相邻的两个抽头23获得。例如，处理电路30计算Q1’和Q2’的和值S12，计算Q2’和Q3’的和值S23，计算Q3’和Q4’的和值S34，计算Q4’和Q5’的和值S45，计算Q5’和Q6’的和值S56，计算Q6’和Q7’的和值S67，计算Q7’和Q8’的和值S78。最后，处理电路30从S12、S23、S34、S45、S56、S67、S78这6个和值中选出最大的和值。假设最大的和值为S34，则说明抽头3获取的归一化信号Q3’及抽头4获取的归一化信号Q4’均是由包含光脉冲信号的光线生成的，归一化信号Q3’及归一化信号Q4’均为目标信号。随后，处理电路30选取与这两个目标信号中的任意一个相邻的归一化信号作为目标信号，与归一化信号Q3’相邻的归一化信号为Q2’，与归一化信号Q4’相邻的归一化信号为Q5’，处理电路30可选取归一化信号Q2’或归一化信号Q5’来作为目标信号。假设处理电路30选择归一化信号Q2’来作为目标信号，则深度信息
Figure PCTCN2021087277-appb-000001
其中，括号中的数字2表示的是抽头3的开启时刻相对于光脉冲信号的发射时刻存在的延迟时间对应的系数。假设处理电路30选择归一化信号Q5’来作为目标信号，则深度信息
Figure PCTCN2021087277-appb-000002
其中，括号中的数字2表示的是抽头3的开启时刻相对于光脉冲信号的发射时刻存在的延迟时间对应的系数。
当光接收模组20中的每个像素21都具有八个抽头23时，则深度相机100的最远可以测量到的距离
Figure PCTCN2021087277-appb-000003
测距范围为[0，3c·T0]。与仅具有三个抽头的像素相比(三个抽头的像素的测距范围为
Figure PCTCN2021087277-appb-000004
)，本申请实施方式的深度相机100中设置了更多数量的抽头23，使得深度相机100的测距范围得到了提升。并且，本申请实施方式的深度相机100将多个抽头23获得的电信号进行相减来消除固有噪声和环境光信号的干扰，不需要额外引入环境光信号的采样窗口。
在某些实施方式中，当预定数量为三时，处理电路30在获得多个归一化信号后，也可以从多个归一化信号中选出幅值最大的一个归一化信号，例如幅值最大的归一化信号为Q(n+1)’，则处理电路30首先判断与归一化信号Q(n+1)’相邻的两个归一化信号Qn’和归一化信号Q(n+2)’之间的大小。当Qn’-Q(n+2)’＞0时，处理电路30确定归一化信号Q(n+2)’由环境光信号生成，而归一化信号Qn’及归一化信号Q(n+1)’由光脉冲信号及环境光信号共同生成，则处理电路30可以根据以下公式计算深度信息D，
Figure PCTCN2021087277-appb-000005
其中，Xn表示和延迟时间相关的线性系数，
Figure PCTCN2021087277-appb-000006
dn＝Q(n+1)’-Q(n+2)’，表示第n+1个抽头23的电信号与第n+2个抽头23的电信号之间的差值，Zn＝Qn’+Q(n+1)’-2Q(n+2)′，表示反射回的光脉冲信号的幅值，则
Figure PCTCN2021087277-appb-000007
当Q(n)’-Q(n+2)’＜0时，处理电路30确定归一化信号Qn’由环境光信号生成，而归一化信号Q(n+1)’及归一化信号Q(n+2)’由光脉冲信号及环境光信号共同生成，则处理电路30可以根据以下公式计算深度信息D，
Figure PCTCN2021087277-appb-000008
其中，Xn表示和延迟时间相关的线 性系数，
Figure PCTCN2021087277-appb-000009
dn＝Q(n+2)’-Qn’，表示第n+2个抽头23的电信号与第n个抽头23的电信号之间的差值，Zn＝Q(n+1)’+Q(n+2)’-2Qn’，表示反射回的光脉冲信号的幅值，则
Figure PCTCN2021087277-appb-000010
本申请实施方式的深度相机100中设置了更多数量的抽头23，使得深度相机100的测距范围得到了提升。并且，本申请实施方式的深度相机100将多个抽头23获得的电信号进行相减来消除固有噪声和环境光信号的干扰，不需要额外引入环境光信号的采样窗口。
在某些实施方式中，预定数量的取值也可以为四，四个目标信号由开启顺序相邻的四个抽头23获得。在处理电路30用于从多个总信号中确定四个目标信号时，处理电路30主要用于对多个总信号进行归一化处理以获得多个归一化信号，并根据多个抽头23的开启顺序计算出任意两个相邻的归一化信号的和值，再从多个和值中选取最大的和值，并将组成该和值的两个归一化信号确定为目标信号，最后根据多个抽头23的开启顺序确定多个归一化信号中的与两个目标信号相邻的两个归一化信号为目标信号。
具体地，处理电路30首先对多个总信号进行归一化处理以获得多个归一化信号。例如，抽头1的归一化信号Q1’＝Q1/1。抽头2的归一化信号Q2’＝Q2/2。抽头3的归一化信号Q3’＝Q3/3。以此类推，抽头8的归一化信号Q8’＝Q8/8。也即抽头n的归一化信号Qn’＝Qn/n。任意一个抽头23的归一化信号的信噪比都大于该抽头23在任意一个子周期T1内获得的电信号的信噪比。
随后，处理电路30从Q1’、Q2’、Q3’、Q4’、Q5’、Q6’、Q7’、Q8’这八个归一化信号中选取三个归一化信号作为目标信号。处理电路30可以计算任意两个相邻的归一化信号的和值(归一化信号的幅值之和)，其中，相邻的归一化信号表示这两个归一化信号由开启顺序相邻的两个抽头23获得。例如，处理电路30计算Q1’和Q2’的和值S12，计算Q2’和Q3’的和值S23，计算Q3’和Q4’的和值S34，计算Q4’和Q5’的和值S45，计算Q5’和Q6’的和值S56，计算Q6’和Q7’的和值S67，计算Q7’和Q8’的和值S78。最后，处理电路30从S12、S23、S34、S45、S56、S67、S78这6个和值中选出最大的和值。假设最大的和值为S34，则说明抽头3获取的归一化信号Q3’及抽头4获取的归一化信号Q4’均是由包含光脉冲信号的光线生成的，归一化信号Q3’及归一化信号Q4’均为目标信号。随后，处理电路30选取与这两个目标信号相邻的两个归一化信号作为目标信号，与归一化信号Q3’相邻的归一化信号为Q2’，与归一化信号Q4’相邻的归一化信号为Q5’，处理电路30可选取归一化信号Q2’和归一化信号Q5’来作为目标信号，此时，深度信息
Figure PCTCN2021087277-appb-000011
其中，括号中的数字2表示的是抽头3的开启时刻相对于光脉冲信号的发射时刻存在的延迟时间对应的系数。
与选择三个目标信号进行深度信息计算的方式相比，选择四个目标信号进行深度信息 计算的方式中，在消除抽头3的归一化信号Q3’中由环境光信号生成的部分时，减去的是抽头2的归一化信号Q2’，而不是抽头5的归一化信号Q5’。可以理解的是，抽头2与抽头3是开启顺序相邻的两个抽头23，二者的归一化信号中由环境光信号生成的部分的量更为相近，从而可以更加准确地去除抽头3的归一化信号Q3’中的环境光分量，有利于获得更准确的深度信息。
在某些实施方式中，预定数量为四时，处理电路30在获得多个归一化信号后，也可以从多个归一化信号中选出幅值最大的一个归一化信号作为目标信号，并选取与这一归一化信号相邻的另外三个(包括直接相邻的两个及间接相邻的一个，间接相邻的归一化信号与幅值最大的归一化信号之间间隔一个归一化信号)归一化信号以作为目标信号。例如幅值最大的归一化信号为Q(n+1)’，则处理电路30首先判断与归一化信号Q(n+1)’直接相邻的两个归一化信号Qn’和归一化信号Q(n+2)’之间的大小。当Qn’-Q(n+2)’＞0时，处理电路30确定归一化信号Q(n+2)’由环境光信号生成，而归一化信号Qn’及归一化信号Q(n+1)’由光脉冲信号及环境光信号共同生成。随后，处理器30确定归一化信号Qn’、归一化信号Q(n+1)’、归一化信号Q(n+2)’为目标信号，并再从剩余的归一化信号中选取与归一化信号Qn’直接相邻的归一化信号Q(n-1)’。随后，处理电路30可以根据以下公式计算深度信息D，
Figure PCTCN2021087277-appb-000012
其中，Xn表示和延迟时间相关的线性系数，
Figure PCTCN2021087277-appb-000013
dn＝Q(n+1)’-Q(n+2)’，表示第n+1个抽头23的电信号与第n+2个抽头23的电信号之间的差值，Zn＝[Qn’-Q(n-1)′]+[Q(n+1)’-Q(n+2)′]，表示反射回的光脉冲信号的幅值，则
Figure PCTCN2021087277-appb-000014
当Q(n)’-Q(n+2)’＜0时，处理电路30确定归一化信号Qn’由环境光信号生成，而归一化信号Q(n+1)’及归一化信号Q(n+2)’由光脉冲信号及环境光信号共同生成。随后，处理器30确定归一化信号Qn’、归一化信号Q(n+1)’、归一化信号Q(n+2)’为目标信号，并再从剩余的归一化信号中选取与归一化信号Q(n+2)’直接相邻的归一化信号Q(n+3)’。随后，处理电路30可以根据以下公式计算深度信息D，
Figure PCTCN2021087277-appb-000015
其中，Xn表示和延迟时间相关的线性系数，
Figure PCTCN2021087277-appb-000016
dn＝Q(n+2)’-Q(n+3)’，表示第n+2个抽头23的电信号与第n+3个抽头23的电信号之间的差值，Zn＝[Q(n+1)’-Qn′]+[Q(n+2)’-Q(n+3)’]，表示反射回的光脉冲信号的幅值，则
Figure PCTCN2021087277-appb-000017
与选择三个目标信号进行深度信息计算的方式相比，选择四个目标信号进行深度信息计算的方式中，在消除第n个抽头23的归一化信号Qn’中由环境光信号生成的部分时，减去的是第n-1个抽头23的归一化信号Q(n-1)’，而不是第n+2个抽头23的归一化信号Q(n+2)’。可以理解的是，第n-1个抽头23与第n个抽头23是开启顺序相邻的两个抽头23，二者的归一化信号中由环境光信号生成的部分的量更为相近，从而可以更加准确地去除第n个抽 头23的归一化信号Qn’中的环境光分量，有利于获得更准确的深度信息。同样地，在消除第n+2个抽头23的归一化信号Q(n+2)’中由环境光信号生成的部分时，减去的是第n+3个抽头23的归一化信号Q(n+3)’，而不是第n个抽头23的归一化信号Qn’。可以理解的是，第n+2个抽头23与第n+3个抽头23是开启顺序相邻的两个抽头23，二者的归一化信号中由环境光信号生成的部分的量更为相近，从而可以更加准确地去除第n+2个抽头23的归一化信号Q(n+2)’中的环境光分量，有利于获得更准确的深度信息。
在某些实施方式中，预定数量的取值还可以是五个、六个、七个、八个、十个等，在此不作限制。多个目标信号中除了两个同时由环境光信号和光脉冲信号生成的目标信号外，其余的目标信号均是由环境光信号生成，处理电路30可以对该多个由环境光信号生成的目标信号取均值，该均值可以用于去除该两个同时由环境光信号和光脉冲信号生成的目标信号中的环境光分量。
请参阅图2和图7，本申请实施方式还提供一种电子设备1000。电子设备1000包括壳体200及上述任意一个实施方式所述的深度相机100。深度相机100与壳体200结合。在一个例子中，深度相机100可以设置在壳体200的安装有显示屏的一面；在另一个例子中，深度相机100可以设置在壳体200的与安装有显示屏的一面相背的另一面，在此不作限制。电子设备1000可以是手机、平板电脑、笔记本电脑、智能穿戴设备(智能手表、智能手环、智能眼镜、智能头盔等)、虚拟现实设备等，在此不作限制。
本申请实施方式的电子设备1000安装有深度相机100，深度相机100中的像素21设置了多个抽头23，且多个抽头23在帧周期内以不同的开启频率开启。通过这种深度自适应的抽头23的开启频率分配操作，可以最大程度地避免用于测量近距离的被摄物体的抽头23的饱和，同时可以保证用于测量远距离的被摄物体能够最大化地采集到光线，从而获得信噪比较高的电信号。基于信噪比较高的电信号进行深度信息的计算，有利于提升深度信息的测量精度。
请参阅图1、图2及图8，本申请实施方式还提供一种控制方法。本申请实施方式的控制方法可以用于上述任意一个实施方式所述的深度相机100。控制方法包括：
01：光发射模组10在一个子周期内发射一个光脉冲信号，在一个帧周期内发送多个光脉冲信号，帧周期包括多个连续的子周期；
02：多个抽头23获取由光脉冲信号和/或环境光信号生成的多个电信号，任意两个抽头23的开启时间不重叠，且任意两个抽头23在帧周期内的开启频率不同；及
03：根据多个电信号计算被摄场景的深度信息。
请参阅图2，在某些实施方式中，步骤03根据多个电信号计算被摄场景的深度信息，包括：
计算每个抽头23在一个帧周期内获得的总电信号；及
根据多个总电信号计算深度信息。
请参阅图2和图6，在某些实施方式中，根据多个总电信号计算深度信息的步骤，包括：
根据每个抽头23在一个帧周期T2内获得的所有电信号计算每个抽头23在一个帧周期T2内获得的总电信号；及
根据多个总电信号计算深度信息。
请参阅图2和图6，在某些实施方式中，根据多个总电信号计算深度信息的步骤，包括：
从多个总信号中确定预定数量个目标信号，预定数量个目标信号由开启顺序相邻的预定数量个抽头23获得；及
根据预定数量个目标信号计算深度信息。
在某些实施方式中，预定数量为三，从多个总信号中确定预定数量个目标信号的步骤，包括：
对多个所述总信号进行归一化处理以获得多个归一化信号；
从所述多个归一化信号中选出幅值最大的归一化信号作为所述目标信号；及
从所述多个所述归一化信号中选出与所述幅值最大的归一化信号相邻的两个归一化信号作为所述目标信号。
请参阅图2，在某些实施方式中，预定数量为三，从多个总信号中确定预定数量个目标信号的步骤，包括：
对多个总信号进行归一化处理以获得多个归一化信号；
根据多个抽头23的开启顺序计算出任意两个相邻的归一化信号的和值；
从多个和值中选取最大的和值，并将组成该和值的两个归一化信号确定为目标信号；
根据多个抽头23的开启顺序确定多个归一化信号中的与两个目标信号中的任意一个相邻的归一化信号为目标信号。
在本说明书的描述中，参考术语“一个实施方式”、“一些实施方式”、“示意性实施方式”、“示例”、“具体示例”或“一些示例”等的描述意指结合所述实施方式或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施方式或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施方式或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施方式或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
尽管上面已经示出和描述了本申请的实施方式，可以理解的是，上述实施方式是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施方式进行变化、修改、替换和变型。

Claims (22)

  1. 一种深度相机，其特征在于，包括：
    光发射模组，所述光发射模组用于在一个子周期内发送一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所述帧周期包括多个连续的所述子周期；
    光接收模组，所述光接收模组包括像素，所述像素包括多个抽头，所述多个抽头用于获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同；及
    处理电路，所述处理电路用于根据所述多个电信号计算被摄场景的深度信息。
  2. 根据权利要求1所述的深度相机，其特征在于，所述像素包括一个光电转换元件，每个所述抽头包括一个转移栅极，所述光电元件用于接收所述光脉冲信号和/或环境光信号以获得所述电信号，所述转移栅极用于转移所述电信号。
  3. 根据权利要求2所述的深度相机，其特征在于，所述光电转换元件的收光面为多边形，所述多边形的每条边上分布有一个或多个所述抽头。
  4. 根据权利要求2所述的深度相机，其特征在于，所述光电转换元件的收光面为圆形，多个所述抽头环绕所述收光面的中心分布。
  5. 根据权利要求1所述的深度相机，其特征在于，所述多个抽头具有相同的采样时长，且所述采样时长与所述光脉冲信号的脉冲宽度相同；
    在一个所述子周期内，前一个所述抽头的关闭时刻为下一个所述抽头的开启时刻。
  6. 根据权利要求5所述的深度相机，其特征在于，在每个所述帧周期内的第一个子周期中，所述多个抽头依次序开启；
    在每个所述帧周期内，在所述第一个子周期中越先开启的所述抽头在该帧周期内的开启频率越低，在所述第一个子周期中越晚开启的所述抽头在该帧周期内的开启频率越高。
  7. 根据权利要求1所述的深度相机，其特征在于，所述处理电路还用于：
    根据每个所述抽头在一个所述帧周期内获得的所有所述电信号计算每个所述抽头在一个所述帧周期内获得的总电信号；及
    根据多个所述总电信号计算所述深度信息。
  8. 根据权利要求7所述的深度相机，其特征在于，所述处理电路还用于：
    从多个所述总信号中确定预定数量个目标信号，所述预定数量个目标信号由开启顺序相邻的预定数量个抽头获得；及
    根据所述预定数量个目标信号计算所述深度信息。
  9. 根据权利要求8所述的深度相机，其特征在于，所述预定数量为三，所述从多个所述总信号中确定预定数量个目标信号，包括：
    对多个所述总信号进行归一化处理以获得多个归一化信号；
    从所述多个归一化信号中选出幅值最大的归一化信号作为所述目标信号；及
    从所述多个所述归一化信号中选出与所述幅值最大的归一化信号相邻的两个归一化信号作为所述目标信号。
  10. 一种电子设备，其特征在于，包括：
    壳体；及
    深度相机，所述深度相机与所述壳体结合，所述深度相机包括：
    光发射模组，所述光发射模组用于在一个子周期内发送一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所述帧周期包括多个连续的所述子周期；
    光接收模组，所述光接收模组包括像素，所述像素包括多个抽头，所述多个抽头用于获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同；及
    处理电路，所述处理电路用于根据所述多个电信号计算被摄场景的深度信息。
  11. 根据权利要求10所述的电子设备，其特征在于，所述像素包括一个光电转换元件，每个所述抽头包括一个转移栅极，所述光电元件用于接收所述光脉冲信号和/或环境光信号以获得所述电信号，所述转移栅极用于转移所述电信号。
  12. 根据权利要求11所述的电子设备，其特征在于，所述光电转换元件的收光面为多边形，所述多边形的每条边上分布有一个或多个所述抽头。
  13. 根据权利要求11所述的电子设备，其特征在于，所述光电转换元件的收光面为圆形，多个所述抽头环绕所述收光面的中心分布。
  14. 根据权利要求10所述的电子设备，其特征在于，所述多个抽头具有相同的采样时长，且所述采样时长与所述光脉冲信号的脉冲宽度相同；
    在一个所述子周期内，前一个所述抽头的关闭时刻为下一个所述抽头的开启时刻。
  15. 根据权利要求14所述的电子设备，其特征在于，在每个所述帧周期内的第一个子周期中，所述多个抽头依次序开启；
    在每个所述帧周期内，在所述第一个子周期中越先开启的所述抽头在该帧周期内的开启频率越低，在所述第一个子周期中越晚开启的所述抽头在该帧周期内的开启频率越高。
  16. 根据权利要求10所述的电子设备，其特征在于，所述处理电路还用于：
    根据每个所述抽头在一个所述帧周期内获得的所有所述电信号计算每个所述抽头在一个所述帧周期内获得的总电信号；及
    根据多个所述总电信号计算所述深度信息。
  17. 根据权利要求16所述的电子设备，其特征在于，所述处理电路还用于：
    从多个所述总信号中确定预定数量个目标信号，所述预定数量个目标信号由开启顺序相邻的预定数量个抽头获得；及
    根据所述预定数量个目标信号计算所述深度信息。
  18. 根据权利要求17所述的电子设备，其特征在于，所述预定数量为三，所述从多个所述总信号中确定预定数量个目标信号，包括：
    对多个所述总信号进行归一化处理以获得多个归一化信号；
    从所述多个归一化信号中选出幅值最大的归一化信号作为所述目标信号；及
    从所述多个所述归一化信号中选出与所述幅值最大的归一化信号相邻的两个归一化信号作为所述目标信号。
  19. 一种控制方法，用于深度相机，其特征在于，所述深度相机包括光发射模组及光接收模组，所述光接收模组包括像素，所述像素包括多个抽头；所述控制方法包括：
    所述光发射模组在一个子周期内发射一个光脉冲信号，在一个帧周期内发送多个所述光脉冲信号，所述帧周期包括多个连续的所述子周期；
    所述多个抽头获取由所述光脉冲信号和/或环境光信号生成的多个电信号，任意两个所述抽头的开启时间不重叠，且任意两个所述抽头在所述帧周期内的开启频率不同；及
    根据所述多个电信号计算被摄场景的深度信息。
  20. 根据权利要求19所述的控制方法，其特征在于，所述根据所述多个电信号计算被摄场景的深度信息，包括：
    根据每个所述抽头在一个所述帧周期内获得的所有所述电信号计算每个所述抽头在一个所述帧周期内获得的总电信号；及
    根据多个所述总电信号计算所述深度信息。
  21. 根据权利要求20所述的控制方法，其特征在于，所述根据多个所述总电信号计算所述深度信息，包括：
    从多个所述总信号中确定预定数量个目标信号，所述预定数量个目标信号由开启顺序相邻的预定数量个抽头获得；及
    根据所述预定数量个目标信号计算所述深度信息。
  22. 根据权利要求21所述的控制方法，其特征在于，所述预定数量为三，所述从多个所述总信号中确定预定数量个目标信号，包括：
    对多个所述总信号进行归一化处理以获得多个归一化信号；
    从所述多个归一化信号中选出幅值最大的归一化信号作为所述目标信号；及
    从所述多个所述归一化信号中选出与所述幅值最大的归一化信号相邻的两个归一化信号作为所述目标信号。
PCT/CN2021/087277 2020-05-29 2021-04-14 深度相机、电子设备及控制方法 WO2021238477A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010472386.1 2020-05-29
CN202010472386.1A CN111580119B (zh) 2020-05-29 2020-05-29 深度相机、电子设备及控制方法

Publications (1)

Publication Number Publication Date
WO2021238477A1 true WO2021238477A1 (zh) 2021-12-02

Family

ID=72112713

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/087277 WO2021238477A1 (zh) 2020-05-29 2021-04-14 深度相机、电子设备及控制方法

Country Status (2)

Country Link
CN (1) CN111580119B (zh)
WO (1) WO2021238477A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111580119B (zh) * 2020-05-29 2022-09-02 Oppo广东移动通信有限公司 深度相机、电子设备及控制方法
JP2022109077A (ja) * 2021-01-14 2022-07-27 凸版印刷株式会社 距離画像撮像装置、及び距離画像撮像方法
CN115047436A (zh) * 2022-08-16 2022-09-13 武汉极动智能科技有限公司 抗干扰方法、装置、系统及存储介质
CN116320667A (zh) * 2022-09-07 2023-06-23 奥比中光科技集团股份有限公司 一种消除运动伪影的深度相机及方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5767953A (en) * 1993-04-12 1998-06-16 The Regents Of The University Of California Light beam range finder
EP2402784A2 (en) * 2010-06-25 2012-01-04 Samsung Electronics Co., Ltd. Apparatus and method for generating depth image
CN102694998A (zh) * 2011-03-24 2012-09-26 三星电子株式会社 深度传感器、深度信息误差补偿方法及信号处理系统
CN110221274A (zh) * 2019-05-09 2019-09-10 深圳奥比中光科技有限公司 时间飞行深度相机及多频调制解调的距离测量方法
CN110320528A (zh) * 2019-06-14 2019-10-11 深圳奥比中光科技有限公司 时间深度相机及多频调制解调的降低噪声的距离测量方法
CN110361751A (zh) * 2019-06-14 2019-10-22 深圳奥比中光科技有限公司 时间飞行深度相机及单频调制解调的降低噪声的距离测量方法
CN111580119A (zh) * 2020-05-29 2020-08-25 Oppo广东移动通信有限公司 深度相机、电子设备及控制方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110187355B (zh) * 2019-05-21 2023-07-04 奥比中光科技集团股份有限公司 一种距离测量方法及深度相机

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5767953A (en) * 1993-04-12 1998-06-16 The Regents Of The University Of California Light beam range finder
EP2402784A2 (en) * 2010-06-25 2012-01-04 Samsung Electronics Co., Ltd. Apparatus and method for generating depth image
CN102694998A (zh) * 2011-03-24 2012-09-26 三星电子株式会社 深度传感器、深度信息误差补偿方法及信号处理系统
CN110221274A (zh) * 2019-05-09 2019-09-10 深圳奥比中光科技有限公司 时间飞行深度相机及多频调制解调的距离测量方法
CN110320528A (zh) * 2019-06-14 2019-10-11 深圳奥比中光科技有限公司 时间深度相机及多频调制解调的降低噪声的距离测量方法
CN110361751A (zh) * 2019-06-14 2019-10-22 深圳奥比中光科技有限公司 时间飞行深度相机及单频调制解调的降低噪声的距离测量方法
CN111580119A (zh) * 2020-05-29 2020-08-25 Oppo广东移动通信有限公司 深度相机、电子设备及控制方法

Also Published As

Publication number Publication date
CN111580119A (zh) 2020-08-25
CN111580119B (zh) 2022-09-02

Similar Documents

Publication Publication Date Title
WO2021238477A1 (zh) 深度相机、电子设备及控制方法
US9058081B2 (en) Application using a single photon avalanche diode (SPAD)
US9417326B2 (en) Pulsed light optical rangefinder
US9316735B2 (en) Proximity detection apparatus and associated methods having single photon avalanche diodes for determining a quality metric based upon the number of events
CN110312945A (zh) 光子传感器装置
CN105190426A (zh) 飞行时间传感器装仓
EP2458484B1 (en) An improved input device and associated method
US11259002B2 (en) Time-of-flight camera and proximity detector
CN102004254B (zh) 调制光学飞行时间相位估计中的延迟补偿
US10754031B2 (en) Power control method, distance measuring module and electronic device
CN109870704A (zh) Tof相机及其测量方法
CN110221274B (zh) 时间飞行深度相机及多频调制解调的距离测量方法
KR20120105169A (ko) 복수의 거리 픽셀들을 포함하는 3차원 이미지 센서의 구동 방법
CN109725326A (zh) 飞行时间相机
CN110221273B (zh) 时间飞行深度相机及单频调制解调的距离测量方法
CN103731611A (zh) 深度传感器、图像捕获方法和图像处理系统
CN107907885A (zh) 一种基于单光子计数方法的水下目标探测装置
WO2022017366A1 (zh) 一种深度成像方法及深度成像系统
US20130342518A1 (en) Information processing apparatus and method, and photoelectric conversion apparatus
CN111308449A (zh) 飞行时间深度成像系统的内置校准
GB2486164A (en) Using a single photon avalanche diode (SPAD) as a proximity detector
US20120133617A1 (en) Application using a single photon avalanche diode (spad)
US8749765B2 (en) Application using a single photon avalanche diode (SPAD)
WO2022242348A1 (zh) dTOF深度图像的采集方法、装置、电子设备及介质
WO2023008465A1 (ja) 測距装置および測距方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21812428

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21812428

Country of ref document: EP

Kind code of ref document: A1