Synchronous judgment method and device for multi-camera photographing and computer readable storage medium

Abstract

The invention discloses a synchronous judgment method and device for multi-camera photographing and a computer readable storage medium. The synchronous judgment method comprises the steps that image frames collected by N cameras respectively are acquired; the photographing trigger signals received by the N cameras are sent at the same moment; determining a target line of each image frame; the target row is a pixel row acquired when the state of an external light source is changed; and determining whether the N image frames are synchronized or not based on the target line of each image frame. According to the scheme, whether the images shot by the multiple cameras are synchronous or not can be accurately judged.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method and apparatus for simultaneous judgment of multi-camera photography, and a computer-readable storage medium. Background Technology

[0002] A multi-camera system consists of multiple cameras located at different positions, each capturing images within its corresponding field of view and processing them accordingly. Multi-camera systems are widely used in scenarios such as autonomous driving, industrial automation inspection, virtual reality and augmented reality, and motion capture for sports events.

[0003] From the perspective of data fusion and 3D reconstruction, multi-camera systems typically need to capture image information of target objects from different perspectives, and then fuse this multi-view data to construct a complete and accurate 3D model or scene. Taking the field of autonomous driving as an example, multiple cameras deployed around a vehicle need to simultaneously capture environmental information such as roads, vehicles, and pedestrians. Only when the cameras are highly synchronized in time can the acquired images be ensured to correspond strictly in the time dimension.

[0004] Currently, to ensure high time synchronization among cameras, the same trigger signal is sent to different cameras to initiate the capture. However, considering the varying processing delays of different cameras, the images captured by different cameras may not be synchronized. Summary of the Invention

[0005] The purpose of this invention is at least to provide a method and apparatus for determining the synchronization of multi-camera photography, and a computer-readable storage medium, which can accurately determine whether images captured by multiple cameras are synchronized.

[0006] In a first aspect, the present invention provides a method for determining the synchronization of multi-camera photography, comprising: acquiring image frames collected by N cameras respectively; the N cameras receiving the photo-taking trigger signal being sent at the same time; determining the target row of each image frame; the target row being the pixel row collected when the state of an external light source changes; and determining whether the N image frames are synchronized based on the target row of each image frame.

[0007] By acquiring image frames from N cameras, extracting the corresponding target row from each image frame, and then determining whether the N image frames are synchronized based on the target row of each image frame, it is possible to accurately determine whether image frames acquired by different cameras are synchronized.

[0008] Optionally, the target row includes at least one of the following: the first pixel row acquired when the external light source changes from an off state to an on state, and the first pixel row acquired when the external light source changes from an on state to an off state.

[0009] Optionally, the target row of the i-th image frame in N image frames is determined by the following steps: obtaining the feature value of each pixel row in the i-th image frame; the feature value of any pixel row includes any of the following: the average brightness of the corresponding pixel row, the average color temperature of the corresponding pixel row; obtaining the difference between the feature value of each pixel row and the feature value of the next adjacent pixel row; determining the pixel row with the largest feature value among the two pixel rows with the largest difference in feature values, as the target row; and / or, determining the pixel row with the smallest feature value among the two pixel rows with the smallest difference in feature values, as the target row.

[0010] Optionally, different image frames correspond to the same resolution; determining whether N image frames are synchronized based on the target row of each image frame includes: obtaining the row number of the target row of each of the N image frames; determining that the N image frames are synchronized if the difference between the row numbers of the target rows of any two image frames is not greater than a preset first difference value; and determining that the N image frames are not synchronized if the difference between the row numbers of the target rows of at least two image frames is greater than the first difference value.

[0011] Optionally, different image frames correspond to the same resolution; determining whether N image frames are synchronized based on the target row of each image frame includes: determining the position of the extreme point of the feature value corresponding to the i-th target row based on the i-th target row, the row above the i-th target row, and the row below the i-th target row; the i-th target row is the target row of the i-th image frame; if the difference between the positions of the extreme points corresponding to any two image frames is not greater than a preset second difference, the N image frames are determined to be synchronized; if the difference between the positions of the extreme points corresponding to at least two image frames is greater than the second difference, the N image frames are determined to be out of sync.

[0012] Optionally, the resolution of the first image frame among the N image frames is different from the resolution of the second image frame; determining whether the N image frames are synchronized based on the target row corresponding to each image frame includes: determining a first time point based on the target row of the first image frame, the first time point being associated with the vertical resolution corresponding to the first image frame; determining a second time point based on the target row of the second image frame, the first time point being associated with the vertical resolution corresponding to the second image frame; and determining that the first image frame and the second image frame are synchronized in response to the time difference between the first time point and the second time point being less than a preset synchronization determination threshold.

[0013] For image frames of different resolutions, the target rows corresponding to each resolution are converted into corresponding time points. By comparing the time points of image frames of different resolutions, it is determined whether the image frames of different resolutions are synchronized.

[0014] Optionally, all N cameras are rolling shutter cameras, and all N cameras have the same frame rate.

[0015] Optionally, the N cameras receive the same photo-taking trigger signal; or, the N cameras receive the same photo-taking trigger signal at the same time.

[0016] Secondly, the present invention also provides a synchronization judgment device for multi-camera photography, comprising: an acquisition unit for acquiring image frames collected by N cameras respectively; wherein the photo-taking trigger signals received by the N cameras are sent at the same time; N is a positive integer and N≥2; a determination unit for determining the target row of each image frame; wherein the target row is the pixel row collected when the state of the external light source changes; and a judgment unit for determining whether the N image frames are synchronized based on the target row of each image frame.

[0017] Thirdly, the present invention also provides a computer-readable storage medium, which is a non-volatile storage medium or a non-transient storage medium, on which a computer program is stored, wherein the computer program, when run by a processor, executes the steps of any of the above-described methods for determining the synchronization of multi-camera photography.

[0018] Fourthly, the present invention also provides another synchronization judgment device for multi-camera photography, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of any of the above-described synchronization judgment methods for multi-camera photography when running the computer program. Attached Figure Description

[0019] Figure 1 This is a flowchart of a method for synchronously determining multi-camera photography in an embodiment of the present invention; Figure 2 This is a schematic diagram of the exposure process of a rolling shutter camera; Figure 3 This is a schematic diagram showing the correspondence between an image frame captured by a rolling shutter camera and the average brightness of a pixel row; Figure 4 This is a schematic diagram of the structure of a multi-camera photography synchronization judgment device according to an embodiment of the present invention. Detailed Implementation

[0020] Currently, to ensure high synchronization among cameras, a single control device sends the same trigger signal to different cameras to initiate image capture. However, considering that the data transmission latency between different cameras and the control device may vary, and that different cameras may have different processing latencies, the images captured by different cameras may not be synchronized.

[0021] In this embodiment of the invention, image frames acquired by N cameras are obtained, and a corresponding target row is extracted from each image frame. Then, based on the target row of each image frame, it is determined whether the N image frames are synchronized. Therefore, it is possible to accurately determine whether image frames acquired by different cameras are synchronized.

[0022] To make the above-mentioned objectives, features and beneficial effects of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] This invention provides a method for determining the synchronization of multi-camera photography, referring to... Figure 1 The following will provide a detailed explanation through specific steps.

[0024] In practical implementation, the following method for synchronously determining multi-camera image capture can be executed by a control device (or controller, control unit) with data processing capabilities. This control device can establish communication connections with multiple cameras. The control device can be a central processing unit (CPU), microprocessor (MCU), graphics processing unit (GPU), etc.

[0025] Step 101: Obtain image frames captured by N cameras respectively.

[0026] In practice, the control device can send a photo-taking trigger signal to N cameras. Upon receiving the trigger signal, each of the N cameras performs a photo-taking action to acquire its corresponding image frame. This results in N image frames, with each frame corresponding one-to-one with one of the N cameras. N is a positive integer and N≥2.

[0027] In practice, the photo-taking trigger signals sent to N cameras are sent simultaneously. Accordingly, if we ignore the transmission delay of each of the N cameras receiving the photo-taking trigger signal (this delay is related to factors such as the length of the wiring harness transmitting the photo-taking trigger signal and whether it is affected by external environmental interference), then the time when the N cameras receive the photo-taking trigger signal is also the same.

[0028] In some embodiments, the photo-taking trigger signal received by N cameras can be the same, that is, the same device can send the photo-taking trigger signal to each of the N cameras.

[0029] For example, the control device in this embodiment of the invention sends a photo-taking trigger signal to N cameras, thereby enabling the simultaneous sending of photo-taking control signals to N cameras.

[0030] In other embodiments, the photo-taking trigger signals received by the N cameras may be sent by different devices, with each device sending its corresponding photo-taking trigger signal to the N cameras at the same time.

[0031] In this embodiment of the invention, all N cameras can be rolling shutter cameras, and all N cameras have the same frame rate.

[0032] For example, N is 4, all 4 cameras are rolling shutter cameras, and the frame rate of all 4 cameras is 30Hz (that is, 30 frames per second).

[0033] In practical applications, it is known that the exposure process of a rolling shutter camera is line-by-line. For an image frame captured by a rolling shutter camera, the exposure time of each row of pixels has a fixed delay relative to the exposure time of the previous row of pixels. This delay is related to the hardware parameters of the image sensor.

[0034] In some embodiments, for a rolling shutter camera with a resolution of 1920×1080 and a frame rate of 30Hz, the time delay between the exposure times of adjacent row pixels is typically less than 30 microseconds (μs).

[0035] Reference Figure 2 A schematic diagram of the exposure process of a rolling shutter camera is given. Figure 2 In the graph, the horizontal axis represents time, and the vertical axis represents pixels. Figure 2 It can be seen that the time delay between the exposure time of the first row of pixels and the exposure time of the second row of pixels is t. row Correspondingly, the time delay between the exposure time of the second row of pixels and the exposure time of the third row of pixels is also t. row And so on.

[0036] Step 102: Determine the target row corresponding to each image frame.

[0037] In this embodiment of the invention, after acquiring N image frames, the control device can detect each image frame to determine whether the target row exists in each image frame. If the target row is not detected in any of the image frames, it is determined that there is no need to perform subsequent synchronization judgment on the N image frames acquired in step 101, that is, the subsequent step 103 is not executed; if the target row is detected in every image frame, the following step 103 is executed.

[0038] In this embodiment of the invention, for any image frame, the target row can be: the pixel row acquired when the state of the external light source changes. The change in the state of the external light source can refer to the external light source switching from an off state to an on state, or it can refer to the external light source switching from an on state to an off state.

[0039] It should be noted that for N image frames, the change in the state of the corresponding external light source is the same when acquiring the target row. That is, when determining the target row, the target row in the N image frames either refers to the pixel row acquired when the external light source switches from the off state to the on state, or it refers to the pixel row acquired when the external light source switches from the on state to the off state.

[0040] In some embodiments, the external light source can be a flash. For example, the external light source is a flash installed on both sides of the road to photograph traffic violations.

[0041] In practical implementation, the external light source can change its state based on an independent external trigger signal. The external trigger signal can be a single pulse signal or a single pulse signal with a longer period. Using this external trigger signal, it is possible to avoid the occurrence of two or more rows of data with high brightness across N image frames.

[0042] In practice, it is known that when an external light source switches from an off state to an on state, that is, when the external light source suddenly turns on (such as a flash), if N cameras are all acquiring images, the feature values ​​of the corresponding pixel rows in the acquired image frame will suddenly change. The feature values ​​of a pixel row can be the average brightness of the pixel row, the average color temperature of the pixel row, etc.

[0043] In other words, by detecting whether there are pixel rows in each image frame whose feature values ​​suddenly change, it can be determined whether there is a target row in the image frame.

[0044] like Figure 3 As shown, a schematic diagram is given of the correspondence between an image frame captured by a rolling shutter camera and the average brightness of a pixel row. Figure 3 In the middle, the left side is image frame 31, and curve 32 represents the average brightness value corresponding to each pixel row in image frame 31.

[0045] Figure 3 In the image frame 31, there is a region with a relatively high average brightness (that is, the region between the two dashed lines). This region is where the pixel rows are located when the external light source is lit. This region may include multiple pixel rows.

[0046] In this embodiment of the invention, the first pixel row collected when the external light source changes from an off state to an on state can be used as the target row, and / or the first pixel row collected when the external light source changes from an on state to an off state can be used as the target row.

[0047] In other words, the first row of pixels in an image frame that transitions from dark to light can be used as the target row. And / or, the first row of pixels in an image frame that transitions from light to dark can be used as the target row.

[0048] In practice, the target row in an image frame can be obtained by acquiring the average brightness of each pixel row in the image frame.

[0049] Specifically, the average brightness value corresponding to each pixel row in the image frame is obtained sequentially. The difference between the average brightness values ​​corresponding to two adjacent pixel rows is calculated. The pixel row with the largest average brightness value among the two pixel rows with the largest difference in average brightness values ​​is taken as the target row; and / or, the pixel row with the smallest average brightness value among the two pixel rows with the smallest difference in average brightness values ​​is taken as the target row.

[0050] For example, in a certain image frame, the average brightness of the 100th pixel row is L100, and the average brightness of the 101st pixel row is L101. The difference between L101 and L100 is greater than the difference in average brightness of any other adjacent pixel row. In this case, it can be assumed that the external light source was off when pixel row L100 was captured, and the external power supply was on when pixel row L101 was captured. Therefore, pixel row L101 is taken as the target row.

[0051] In practice, the target row in the image frame can be obtained by acquiring the average color temperature of each pixel row in the image frame.

[0052] Specifically, the average color temperature of each pixel row in the image frame is obtained sequentially. The difference between the average color temperatures of two adjacent pixel rows is calculated. The pixel row with the largest average color temperature among the two pixel rows with the largest difference in average color temperature is taken as the target row. And / or, the pixel row with the smallest average color temperature among the two pixel rows with the smallest difference in average color temperature is taken as the target row.

[0053] In practical applications, it is known that when an external light source is illuminated, it actually affects the color temperature of the corresponding pixel row in an image frame. Therefore, the target row can also be determined by observing the change in the average color temperature of the pixel row.

[0054] It is understood that other algorithms can also be used to determine the target row of an image frame. The specific algorithm for determining the target row of an image frame is not limited to the scheme provided in the above embodiments.

[0055] Step 103: Based on the target row of each image frame, determine whether N image frames are synchronized.

[0056] In this embodiment of the invention, after obtaining the target row of each image frame, it can be determined whether N image frames are synchronized. If the N image frames are synchronized, it can be determined that the N cameras are synchronized. Conversely, if the N image frames are not synchronized, it can be determined that the N cameras are not synchronized.

[0057] In this embodiment of the invention, the resolutions of the N image frames can all be the same. Alternatively, the resolutions of the N image frames can not all be exactly the same. For example, if N is 4, the resolution of each of the four image frames is 1920×1080 pixels.

[0058] In practice, when N image frames have the same resolution, the row number of the target row of each image frame can be obtained, and the difference between the row numbers of the target rows of any two image frames can be calculated. If the difference between the row numbers of the target rows of any two image frames is not greater than a preset first difference value, then the N image frames are considered synchronized; if the difference between the row numbers of the target rows of at least two image frames is greater than the first difference value, then the N image frames are considered desynchronized.

[0059] As mentioned above, for a rolling shutter camera with a resolution of 1920×1080 and a frame rate of 30Hz, the time delay between the exposure times of adjacent rows of pixels is typically less than 30 microseconds. In other words, when the difference in row numbers of the target rows of two image frames is no greater than 1, it means the time delay between these two image frames does not exceed 30 microseconds; when the difference in row numbers of the target rows of two image frames is no greater than 2, it means the time delay between these two image frames does not exceed 60 microseconds, and so on.

[0060] In practice, the first difference can be set according to the specific application scenario. Specifically, if the synchronization requirements of image frames are high in some application scenarios, the first difference can be set to a smaller value, such as 0 or 1; if the synchronization requirements of image frames are low in some application scenarios, the first difference can be set to a larger value, such as 2, 3 or a larger value.

[0061] For example, if the delay between any two image frames captured by a rolling shutter camera is set to be within 30 microseconds, the two image frames can be determined to be synchronized. If the delay between the exposure times of adjacent rows of pixels of the rolling shutter camera is 20 microseconds, then the first difference value is set to 1.

[0062] For example, if the delay between any two image frames captured by a rolling shutter camera is set to be within 50 microseconds, the two image frames can be determined to be synchronized. If the delay between the exposure time of adjacent rows of pixels of the rolling shutter camera is 20 microseconds, then the first difference value is set to 2.

[0063] In specific implementation, the row number of the target row obtained in the above embodiments is an integer, that is, the value of the target row represents the time when the external light source (such as a flash) is turned on or off, and the first difference mentioned above is the time delay t. row It is an integer multiple of the specified value. When determining whether different image frames are synchronized, it is also possible to more precisely determine the time when the external light source is turned on or off.

[0064] In this embodiment of the invention, when different image frames correspond to the same resolution, the i-th target row, the row above the i-th target row, and the row below the i-th target row can be obtained to determine the location of the extreme point of the feature value corresponding to the i-th target row. The i-th target row is the target row of the i-th image frame. The location of the extreme point represents the time when the external light source is turned on or off. Compared to representing the time when the external light source is turned on or off by the row number of the target row, the location of the feature point can more accurately represent the time when the external light source is turned on or off.

[0065] In other words, for the i-th image frame, we obtain the feature values ​​corresponding to its target row (let's say row j), the row above the target row (i.e., row j-1), and the row below the target row (i.e., row j+1), and then determine the location of the extreme point of the feature value in row j. By analogy, we can obtain the location of the extreme point of the feature value for each image frame.

[0066] If the difference between the positions of the extreme points corresponding to any two image frames is not greater than a preset second difference, then N image frames can be determined to be synchronized; otherwise, if the difference between the positions of the extreme points corresponding to at least two image frames is greater than a preset second difference, then N image frames can be determined to be out of sync.

[0067] Taking the average brightness value as an example, for the i-th image frame, obtain the average brightness value of the j-th row, the (j+1)-th row, and the (j-1)-th row. Perform a second-order fitting on the average brightness values ​​of the j-th row, the (j+1)-th row, and the (j-1)-th row to obtain the position with the maximum average brightness value, which is taken as the location of the extreme point. The extreme point is located in the j-th row.

[0068] Understandably, the location of the extreme point with the largest average brightness in the j-th row can also be determined through other methods.

[0069] In practice, the location of the final determined extreme point is represented by the row number with fractional precision.

[0070] For example, the target line is the 190th line of the image frame, and the location of the extreme point with the largest average brightness is the 190.8th line of the image frame.

[0071] In specific implementations, the aforementioned second difference can be preset. In some embodiments, the aforementioned second difference is associated with a specific application scenario. The larger the set second difference, the greater the probability that different images are determined to be synchronized; the smaller the set second difference, the smaller the probability that different image frames are determined to be synchronized.

[0072] As described in the above embodiments, if the synchronization requirement is high, a smaller second difference can be selected; if the synchronization requirement is low, a larger difference can be selected. The specific method for determining the second difference can be referred to the first difference, and will not be elaborated here.

[0073] In this embodiment of the invention, there may be cases where the resolutions of the N image frames are not entirely the same. For example, N may be 4, where two image frames have a resolution of 1920×1080 pixels and the other two image frames have a resolution of 1280×720 pixels.

[0074] For image frames with different resolutions, the difference between the row numbers of their corresponding target rows may be significant. For example, for an image frame with the first resolution, the corresponding target row is row 188; while for an image frame with the second resolution, the corresponding target row is row 192.

[0075] For a first image frame and a second image frame with different resolutions, to determine whether they are synchronized, a first time point can be determined based on the target row of the first image frame, which is associated with the vertical resolution of the first image frame; a second time point can be determined based on the target row of the second image frame, which is associated with the vertical resolution of the second image frame. The time difference between the first and second time points is obtained. If the time difference between the first and second time points is less than a preset synchronization determination threshold, the first and second image frames are determined to be synchronized; otherwise, if the time difference between the first and second time points is greater than the preset synchronization determination threshold, the first and second image frames are determined to be out of sync.

[0076] The synchronization threshold mentioned above is the time threshold used to determine whether two image frames are synchronized. This synchronization threshold can be preset. For example, the synchronization threshold can be preset to 50μs. Or, the synchronization threshold can be preset to 30μs.

[0077] In some embodiments, the first time point can be: The second time point can be: ;in, , These are the vertical resolution (high) of the two cameras, respectively. , These are the row numbers where the same brightness transition (either from dark to light or from light to dark) occurs in two rolling shutter cameras.

[0078] In the above embodiments of the present invention, when determining whether two image frames are synchronized, the idle time before or after the image frame (T) is ignored. blanking In other words, in the above embodiments, the idle time is set to 0 by default.

[0079] In practical implementation, for scenarios where the idle time is not zero, it is assumed that the image frames are synchronized at the frame header, and the idle time is set at the frame tail. For a given image frame, the time interval T from the frame header to the detection of a change in brightness (i.e., the external light source changing from off to on) can be calculated: , among them Here, H is the row number of the target row in the image frame, H is the vertical resolution, and f is the frame rate.

[0080] In this scenario, the synchronization status of two image frames can be determined based on the difference between their corresponding time intervals T. Specifically, when the time difference between the time interval T1 of the first image frame and the time interval T2 of the second image frame is less than the aforementioned synchronization threshold, the first and second image frames are considered synchronized; otherwise, they are considered out of sync.

[0081] In summary, the synchronization determination method provided in this embodiment of the invention acquires image frames from N cameras, obtains the corresponding target row from each image frame, and then determines whether the N image frames are synchronized based on the target row of each image frame. Therefore, it can accurately determine whether image frames acquired by different cameras are synchronized.

[0082] Reference Figure 4 The present invention provides a synchronization determination device 40 for multi-camera photography, comprising: an acquisition unit 401, a determination unit 402, and a determination unit 403, wherein: The acquisition unit 401 is used to acquire image frames captured by N cameras respectively; the photo-taking trigger signals received by the N cameras are sent at the same time; N is a positive integer and N≥2; The determining unit 402 is used to determine the target row of each image frame; the target row is the pixel row acquired when the state of the external light source changes. The judgment unit 403 is used to determine whether N image frames are synchronized based on the target row of each image frame.

[0083] In specific implementation, the specific execution process of the above-mentioned acquisition unit 401, determination unit 402 and judgment unit 403 can be referred to the above steps 101 to 103, which will not be repeated here.

[0084] In specific implementation, the modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both.

[0085] For example, for various devices and products applied to or integrated into a chip, each module / unit can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits; for various devices and products applied to or integrated into a chip module, each module / unit can be implemented using hardware methods such as circuits, and different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The components can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, each of its components / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0086] This invention also provides a computer-readable storage medium, which is a non-volatile or non-transient storage medium, storing a computer program thereon. When the computer program is run by a processor, it executes the steps of the multi-camera photography synchronization judgment method provided in any of the above embodiments.

[0087] This invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the multi-camera photography synchronization judgment method provided in any of the above embodiments.

[0088] This invention also provides another synchronization determination device for multi-camera photography, including a memory and a processor. The memory stores a computer program that can run on the processor. When the processor runs the computer program, it executes the steps of the synchronization determination method for multi-camera photography provided in any of the above embodiments.

[0089] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include ROM, RAM, disk, or optical disk, etc.

[0090] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A method for determining synchronization during multi-camera photography, characterized in that, include: Acquire image frames captured by N cameras; The photo-taking trigger signals received by the N cameras are sent at the same time; N is a positive integer and N≥2; Determine the target row for each image frame; the target row is the row of pixels acquired when the state of the external light source changes. Based on the target row of each image frame, determine whether N image frames are synchronized.

2. The synchronization determination method as described in claim 1, characterized in that, The target row includes at least one of the following: The first row of pixels captured when the external light source changes from an off state to an on state.

3. The synchronization determination method as described in claim 2, characterized in that, The target row of the i-th image frame out of N image frames is determined using the following steps: Obtain the feature value of each pixel row in the i-th image frame; the feature value of any pixel row includes any of the following: the average brightness of the corresponding pixel row, the average color temperature of the corresponding pixel row; Obtain the difference between the feature value of each pixel row and the feature value of the next adjacent pixel row; The pixel row with the largest feature value among the two pixel rows with the largest difference in feature values ​​is determined as the target row; and / or, the pixel row with the smallest feature value among the two pixel rows with the smallest difference in feature values ​​is determined as the target row.

4. The synchronization determination method as described in claim 3, characterized in that, Different image frames correspond to the same resolution; determining whether N image frames are synchronized based on the target row of each image frame includes: Obtain the row number of the target row for each of the N image frames; If the difference between the row numbers of the target rows of any two image frames is not greater than a preset first difference value, then the N image frames are synchronized. If the difference between the row numbers of the target rows corresponding to at least two image frames is greater than the first difference, it is determined that the N image frames are out of sync.

5. The synchronization determination method as described in claim 3, characterized in that, Different image frames correspond to the same resolution; determining whether N image frames are synchronized based on the target row of each image frame includes: Based on the i-th target row, the row preceding the i-th target row, and the row following the i-th target row, determine the location of the extreme point of the feature value corresponding to the i-th target row; the i-th target row is the target row of the i-th image frame; If the difference between the positions of the extreme points corresponding to any two image frames is not greater than a preset second difference, the N image frames are synchronized. If the difference between the positions of the extreme points corresponding to at least two image frames is greater than the second difference, it is determined that the N image frames are out of sync.

6. The synchronization determination method as described in claim 3, characterized in that, The resolution of the first image frame in the N image frames is different from the resolution of the second image frame; determining whether the N image frames are synchronized based on the target row of each image frame includes: A first time point is determined based on the target row of the first image frame, and the first time point is associated with the vertical resolution corresponding to the first image frame. A second time point is determined based on the target row of the second image frame, and the first time point is associated with the vertical resolution corresponding to the second image frame. In response to the time difference between the first time point and the second time point being less than a preset synchronization determination threshold, the first image frame and the second image frame are determined to be synchronized.

7. The synchronization determination method according to any one of claims 1 to 6, characterized in that, All N cameras are rolling shutter cameras, and all N cameras have the same frame rate.

8. The synchronization determination method as described in claim 7, characterized in that, The N cameras receive the same photo-taking trigger signal; or, the N cameras receive the same photo-taking trigger signal at the same time.

9. A device for synchronous judgment of multi-camera photography, characterized in that, include: The acquisition unit is used to acquire image frames captured by N cameras respectively; The photo-taking trigger signals received by the N cameras are sent at the same time; N is a positive integer and N≥2; A determining unit is used to determine the target row for each image frame; the target row is the pixel row acquired when the state of the external light source changes. The judgment unit is used to determine whether N image frames are synchronized based on the target row of each image frame.

10. A computer-readable storage medium, wherein the computer-readable storage medium is a non-volatile storage medium or a non-transient storage medium, and a computer program is stored thereon, characterized in that, When the computer program is run by the processor, it performs the steps of the multi-camera photography synchronization determination method according to any one of claims 1 to 8.

11. A device for simultaneous judgment of multi-camera photography, comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor runs the computer program, it performs the steps of the multi-camera photography synchronization determination method according to any one of claims 1 to 8.

Images