Data acquisition synchronism verification method and system based on high-frequency and high-definition camera

Abstract

The invention relates to the technical field of camera testing, in particular to a data acquisition synchronization verification method and system based on a high-frequency and high-definition camera, and the method comprises the steps: selecting a certain acquisition synchronization verification strategy from an acquisition synchronization verification strategy set pre-stored in a database; constructing a corresponding verification environment based on the selected certain collection synchronization verification strategy, performing real-time collection through each high-frequency high-definition camera on the vehicle in the verification environment, and outputting camera data corresponding to each high-frequency high-definition camera; analyzing the camera data corresponding to each high-frequency high-definition camera, and outputting picture data in a preset format corresponding to the analyzed camera data; and comparing and judging the picture data corresponding to each high-frequency high-definition camera based on the selected collection synchronism verification strategy according to the analyzed picture data in the preset format, and outputting synchronism verification result data among the high-frequency high-definition cameras.

Description

Technical Field

[0001] The present invention relates to the field of camera testing technology, and in particular to a method and system for verifying the synchronization of data acquisition based on a high-frequency high-definition camera. Background Art

[0002] In areas such as intelligent transportation and autonomous driving, high-frequency, high-definition cameras serve as key sensing devices. The synchronization of their collected data is crucial for the system's accurate perception of the external environment and precise decision-making. For example, in an autonomous driving scenario, multiple high-frequency, high-definition cameras deployed around the vehicle must collect real-time information about the road, pedestrians, and other vehicles. If the data collected by each camera is out of sync, the vehicle control system may make incorrect decisions due to erroneous environmental perception, leading to serious safety accidents.

[0003] Currently, the verification of the synchronization of high-frequency, high-definition camera data acquisition is done by manually setting up specific scenarios for manual testing. This method is not only inefficient, but also has limited test scenarios and cannot cover the complex working conditions in actual applications, making it difficult to fully verify the synchronization of camera data acquisition.

[0004] Based on this, there is an urgent need for a synchronization verification method and system based on high-frequency, high-definition camera data acquisition, which can realize the automation and intelligence of the synchronization verification of high-frequency, high-definition camera data acquisition, and greatly improve the accuracy and efficiency of synchronization verification. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a method and system for synchronicity verification based on high-frequency, high-definition camera data acquisition, which can realize the automation and intelligence of synchronicity verification of high-frequency, high-definition camera data acquisition, and greatly improve the accuracy and efficiency of synchronicity verification.

[0006] In order to achieve the above object, a method for verifying the synchronization of data acquisition based on a high-frequency high-definition camera is provided, comprising the following steps: S1. Select a certain acquisition synchronization verification strategy from the acquisition synchronization verification strategy set pre-stored in the database; S2. Based on the selected acquisition synchronization verification strategy, a corresponding verification environment is constructed. In the verification environment, real-time acquisition is performed through each high-frequency high-definition camera on the vehicle, and the camera data corresponding to each high-frequency high-definition camera is output; S3, parsing the camera data corresponding to each high-frequency high-definition camera, and outputting the corresponding image data in a preset format after parsing; S4. Based on the parsed image data in the preset format and the selected acquisition synchronization verification strategy, the image data corresponding to each high-frequency high-definition camera is compared and judged, and the synchronization verification result data between each high-frequency high-definition camera is output.

[0007] Technical Principles and Results of This Solution: In this solution, a database pre-stores multiple acquisition synchronization verification strategies. These strategies are typically designed based on principles such as timestamp alignment, feature point matching, and event triggering. Time Synchronization Mechanism: When building the verification environment, it is necessary to ensure that all high-frequency, high-definition cameras on the vehicle are operating under a unified time reference.

[0008] In the verification environment, each camera captures the scene at a set high-frequency acquisition rate (e.g., 100 frames per second) and simultaneously outputs camera data containing raw image information and metadata such as timestamps and device identifiers. A parsing algorithm processes the raw camera data, extracting pixel information and converting it into image data in a pre-set format (e.g., JPEG, PNG, or other standard image formats). This process ensures that the parsed image data retains key synchronization verification elements, such as accurate timestamp information.

[0009] According to the parsed image data in the preset format, combined with the selected acquisition synchronization verification strategy, the image data corresponding to each high-frequency high-definition camera is compared and judged, and the synchronization verification result data between each high-frequency high-definition camera is output.

[0010] In this solution, by setting and selecting the acquisition synchronization verification strategy, the corresponding test environment is built and the corresponding test is accurately executed. This can realize the automation and intelligence of the synchronization verification of high-frequency and high-definition camera data acquisition, greatly improving the accuracy and efficiency of synchronization verification.

[0011] A multi-dimensional verification approach combining timestamp comparison and scenario feature matching avoids the limitations of single-dimensional verification. Multiple pre-defined verification strategies can be flexibly selected based on different application scenarios. The entire process, from data collection to synchronization result output, can be completed in a short period of time, meeting the needs of real-time monitoring and feedback, greatly improving the accuracy and speed of synchronization verification.

[0012] Further, the acquisition synchronicity verification strategy set includes a first laboratory verification strategy, a second laboratory verification strategy, and a third path-test verification strategy; The first laboratory validation strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with the marquee, and camera video image data is collected and stored; Analyze the camera video image data corresponding to each high-frequency high-definition camera and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to different high-frequency high-definition cameras at the same trigger timestamp; According to a preset first synchronization real-time calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the real-time and synchronization between the different high-frequency high-definition cameras; The first synchronization real-time calculation formula is: Real-time = the receiving timestamp of the image data - the triggering timestamp of the image data; Synchronicity = the difference in the number of lights on in images with the same trigger timestamp from different high-frequency HD cameras.

[0013] Beneficial Effects: This strategy employs a dual-dimensional quantitative evaluation method, combining time and space, to precisely control camera synchronization performance. By calculating the difference between the image data reception timestamp and the trigger timestamp to measure real-time performance, it can detect even millisecond-level delays during data transmission.

[0014] The ticker, serving as a unified hardware trigger source, ensures that all cameras begin capturing data at the same physical moment, eliminating time deviations caused by software trigger delays and providing a solid and reliable foundation for verification. The captured camera video image data is parsed into JPEG format. This standardized data format not only preserves keyframe information but also significantly reduces the algorithmic complexity of subsequent image processing and feature extraction. This enables efficient cross-device and cross-platform data comparison and analysis, making verification results more reliable and universal.

[0015] The number of lights on the marquee serves as an intuitive visual anchor point and has higher accuracy and stability than feature matching in complex scenes.

[0016] Furthermore, the second laboratory verification strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with a display with an updated timestamp, and camera video image data is collected; Analyze the camera video image data corresponding to each high-frequency high-definition camera and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to different high-frequency high-definition cameras at the same trigger timestamp; According to a preset second synchronization calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the synchronization between the different high-frequency high-definition cameras; The second synchronization calculation formula is: Synchronicity = the difference in timestamps of the first row of image data corresponding to different high-frequency HD cameras.

[0017] Beneficial Effect: By encoding the physical time directly into the image data, the system uses a display with updated timestamps. This approach avoids the delays associated with traditional software timestamp transmission (such as network transmission and system call overhead), bringing the time base closer to the actual physical moment. High-precision timestamps are presented directly in the image via the display, completely eliminating the delays associated with software timestamp transmission.

[0018] The second synchronization calculation formula directly compares the timestamp difference of the first row of image data, achieving pixel-level time synchronization verification. Compared to the difference in the number of lights on in the first strategy, the timestamp difference more intuitively reflects the absolute time difference between cameras, making it particularly suitable for scenarios requiring precise time alignment.

[0019] As a distinct visual feature, the display timestamp is more resistant to interference than a ticker. Even in the presence of vehicle vibration or slight camera shake, the display timestamp typically occupies a fixed position in the image (such as the first row), making it easy to locate and extract. In contrast, the ticker's position may shift in dynamic scenes, making feature matching more difficult.

[0020] Furthermore, the third road test verification strategy is: Install a high-frequency HD camera in front of the vehicle, on the front left, and on the front right. Test whether each camera can normally send and display the captured image. The vehicle is started to drive on a real road. At this time, each high-frequency high-definition camera begins to collect corresponding camera video image data during the vehicle's driving process and stores it in the database; Parse the camera video image data stored in the database and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, the image data of different high-frequency high-definition cameras at the same timestamp is found, and the target objects of different high-frequency high-definition cameras at the same timestamp are compared to see whether they are consistent. If so, the data collection of each high-frequency high-definition camera is synchronized. Otherwise, the data collection of each high-frequency high-definition camera is not synchronized.

[0021] Beneficial Effects: Testing in a real-world road environment, where vehicles face practical challenges such as complex road conditions, variable lighting, and dynamic targets, better demonstrates the synchronization performance of high-frequency, high-definition cameras in real-world applications compared to laboratory environments. For example, dynamic targets such as fast-moving vehicles and frequently changing traffic lights effectively verify the consistency of the camera's capture of the scene at the same moment, ensuring that the verification results are directly applicable to real-world scenarios. This avoids the disconnect between ideal laboratory verification results and actual results, making the verification results highly valuable for practical applications.

[0022] By comparing the consistency of objects in different camera images captured at the same timestamp, we can intuitively and accurately determine whether camera data collection is synchronized. During vehicle operation, objects such as pedestrians and other vehicles on the road are constantly changing. If camera data collection is out of sync, the position and status of the same object in different camera images will inevitably differ.

[0023] Cameras are installed in the front, left, and right of the vehicle, simulating a multi-view perception system in real-world applications, collecting road information from multiple angles. This strategy not only verifies the synchronization of data collection from a single camera, but also emphasizes the synchronization of multiple cameras working together.

[0024] This strategy requires no complex specialized equipment or environment setup; it simply requires installing cameras at designated locations on the vehicle and conducting real-world road tests. The process is simple and intuitive. Automakers, autonomous driving R&D teams, and others can quickly apply this strategy to product testing, conducting large-scale road tests before mass production. This allows them to promptly identify and resolve camera synchronization issues, saving verification costs and time, improving product development efficiency, and facilitating large-scale adoption within the industry, accelerating the implementation of related technologies.

[0025] The present invention also provides a system for verifying the synchronization of data acquisition based on a high-frequency high-definition camera, which uses the above-mentioned method for verifying the synchronization of data acquisition based on a high-frequency high-definition camera. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flow chart of a method for verifying synchronization of data acquisition based on a high-frequency high-definition camera in Example 1 of the present invention.

[0027] Figure 2 This is a diagram showing the acquisition principle of a high-frequency high-definition camera in Example 1 of the present invention.

[0028] Figure 3 This is the display diagram corresponding to the analyzed ticker in the first embodiment of the present invention.

[0029] Figure 4 This is a display diagram of a display with an updated timestamp in the first embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following is further described in detail through specific implementation methods: Example 1 A method for verifying the synchronization of data acquisition based on high-frequency and high-definition cameras is basically as follows: Figure 1 As shown, the following steps are included: S1. Select a certain acquisition synchronization verification strategy from the acquisition synchronization verification strategy set pre-stored in the database; S2. Based on the selected acquisition synchronization verification strategy, a corresponding verification environment is constructed. In the verification environment, real-time acquisition is performed through each high-frequency high-definition camera on the vehicle, and the camera data corresponding to each high-frequency high-definition camera is output; S3, parsing the camera data corresponding to each high-frequency high-definition camera, and outputting the corresponding image data in a preset format after parsing; S4. Based on the parsed image data in the preset format and the selected acquisition synchronization verification strategy, the image data corresponding to each high-frequency high-definition camera is compared and judged, and the synchronization verification result data between each high-frequency high-definition camera is output.

[0031] The acquisition synchronicity verification strategy set includes a first laboratory verification strategy, a second laboratory verification strategy, and a third path-test verification strategy; The first laboratory validation strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with the marquee, and camera video image data is collected and stored; The camera video image data corresponding to each high-frequency high-definition camera is parsed and the picture data in JPEG format is output; in this embodiment, the stored H264 / H265 data of each channel is encoded and decoded into a JPEG format picture.

[0032] According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to the same trigger timestamp of different high-frequency high-definition cameras; the information of each frame of each channel includes the trigger timestamp, the receiving timestamp, and the number of lights on the ticker in the picture. In this embodiment, the parsed ticker picture is as follows Figure 3 As shown in the figure, we can see the number of lights captured by each camera. The difference in the number of lights in the pictures with the same trigger timestamp in different channels is the synchronization difference of the cameras in different channels.

[0033] The control principle of the LED light strip marquee is to control the change of the current signal to light up and extinguish the LED lamp beads in sequence, forming a circular flow visual effect.

[0034] According to a preset first synchronization real-time calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the real-time and synchronization between the different high-frequency high-definition cameras; The first synchronization real-time calculation formula is: Real-time = the receiving timestamp of the image data - the triggering timestamp of the image data; Synchronicity = the difference in the number of lights on in images with the same trigger timestamp from different high-frequency HD cameras.

[0035] The second laboratory verification strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with a display with an updated timestamp, and camera video image data is collected; Analyze the camera video image data corresponding to each high-frequency high-definition camera and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to different high-frequency high-definition cameras at the same trigger timestamp; According to a preset second synchronization calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the synchronization between the different high-frequency high-definition cameras; The second synchronicity calculation formula is: Synchronicity = the difference between the first line timestamps in the image data corresponding to different high-frequency high-definition cameras. In this embodiment, the positions corresponding to the first line timestamps in the image data corresponding to different high-frequency high-definition cameras are as follows: Figure 4 shown.

[0036] The third road test verification strategy is: Install a high-frequency HD camera in front of the vehicle, on the front left, and on the front right. Test whether each camera can normally send and display the captured image. The vehicle is started to drive on a real road. At this time, each high-frequency high-definition camera begins to collect corresponding camera video image data during the vehicle's driving process and stores it in the database; Parse the camera video image data stored in the database and output image data in JPEG format; Based on the image data corresponding to each high-frequency HD camera, the image data of different high-frequency HD cameras at the same timestamp is found and compared to see if the target objects at the same timestamp are consistent. If so, the data collection of the high-frequency HD cameras is synchronized. Otherwise, the data collection of the high-frequency HD cameras is not synchronized. The targets are other vehicles, pedestrians, intersections, and lane change intersections. If an inconsistency is detected, a re-verification is performed, which requires checking whether the industrial computer timing and camera triggering are successful, and whether the frame rates of each camera are consistent; only then can the subsequent synchronization verification be carried out.

[0037] In this embodiment, if Figure 2As shown, the high-frequency, high-definition camera is an on-board high-frequency, high-definition camera based on a PCIE data acquisition system. Through interaction between the upper and lower computer subsystems, it performs real-time and synchronous acquisition of on-board camera data, simultaneously storing and uploading data to a NAS server, enabling real-time data updates and sharing. This improves the performance and safety of intelligent driving vehicles, provides high-quality data to support autonomous navigation and decision-making, and facilitates data injection, simulation testing, and other aspects of autonomous driving. Specifically, the acquisition system includes a lower computer subsystem and a host computer subsystem. The lower computer subsystem includes a data acquisition host, a PCIE video acquisition card connected to the data acquisition host, a PTP timing server, an inertial navigation system, and a network-attached storage device. The data acquisition host includes a video encoding processing module, a system status module, and a data acquisition software module. The video encoding processing module is used to receive camera video stream data and simultaneously perform a first encoding encapsulation and a second encoding. The first encoding encapsulation forms first data and sends it to the network-attached storage device, while the second encoding forms second data and sends it to the host computer subsystem. The system status module is used to collect system status data and send it to the host computer subsystem and the network-attached storage device. The storage encoding format of the camera video stream is H264 / H265. The camera video stream data is camera video image data.

[0038] To ensure accurate time on the data acquisition host, the timing server and acquisition host are connected to the same switch. This allows the timing server and inertial navigation system to synchronize time with the industrial computer and camera. The camera output is a timestamped video stream. The lower-level computer data acquisition software module timestamps each frame of camera data received. Each frame includes the timestamp of the camera triggering the image output and the timestamp of the reception moment. The network-attached storage device then receives and stores the first data, system status data, and annotation information associated with the timestamp, and uploads them to the server in real time.

[0039] This embodiment also discloses a system for verifying the synchronization of data acquisition using a high-frequency, high-definition camera, which uses the above-mentioned method for verifying the synchronization of data acquisition using a high-frequency, high-definition camera.

[0040] The above is only an embodiment of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is excessively described here. Ordinary technicians in the relevant field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for verifying the synchronization of data acquisition based on a high-frequency high-definition camera, characterized by: The following steps are involved: S1. Select a certain acquisition synchronization verification strategy from the acquisition synchronization verification strategy set pre-stored in the database; S2. Based on the selected acquisition synchronization verification strategy, a corresponding verification environment is constructed. In the verification environment, real-time acquisition is performed through each high-frequency high-definition camera on the vehicle, and the camera data corresponding to each high-frequency high-definition camera is output; S3, parsing the camera data corresponding to each high-frequency high-definition camera, and outputting the corresponding image data in a preset format after parsing; S4. Based on the parsed image data in the preset format and the selected acquisition synchronization verification strategy, the image data corresponding to each high-frequency high-definition camera is compared and judged, and the synchronization verification result data between each high-frequency high-definition camera is output.

2. The method for verifying synchronization of high-frequency and high-definition camera data acquisition according to claim 1, characterized in that: The acquisition synchronicity verification strategy set includes a first laboratory verification strategy, a second laboratory verification strategy, and a third path-test verification strategy; The first laboratory validation strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with the marquee, and camera video image data is collected and stored; Analyze the camera video image data corresponding to each high-frequency high-definition camera and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to different high-frequency high-definition cameras at the same trigger timestamp; According to a preset first synchronization real-time calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the real-time and synchronization between the different high-frequency high-definition cameras; The first synchronization real-time calculation formula is: Real-time = the receiving timestamp of the image data - the triggering timestamp of the image data; Synchronicity = the difference in the number of lights on in images with the same trigger timestamp from different high-frequency HD cameras.

3. The method for verifying synchronization of high-frequency and high-definition camera data acquisition according to claim 2, characterized in that: The second laboratory verification strategy is: In a verification environment, each high-frequency, high-definition camera on the vehicle is aligned with a display with an updated timestamp, and camera video image data is collected; Analyze the camera video image data corresponding to each high-frequency high-definition camera and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, find the image data corresponding to different high-frequency high-definition cameras at the same trigger timestamp; According to a preset second synchronization calculation formula, the image data of different high-frequency high-definition cameras at the same trigger timestamp are compared and calculated to determine the synchronization between the different high-frequency high-definition cameras; The second synchronicity calculation formula is: Synchronicity = the difference in timestamps of the first row of image data corresponding to different high-frequency HD cameras.

4. The method for verifying synchronization of high-frequency and high-definition camera data acquisition according to claim 3, characterized in that: The third road test verification strategy is: Install a high-frequency HD camera in front of the vehicle, on the front left, and on the front right. Test whether each camera can normally send and display the captured image. The vehicle is started to drive on a real road. At this time, each high-frequency high-definition camera begins to collect corresponding camera video image data during the vehicle's driving process and stores it in the database; Parse the camera video image data stored in the database and output image data in JPEG format; According to the image data corresponding to each high-frequency high-definition camera, the image data of different high-frequency high-definition cameras at the same timestamp is found, and the target objects of different high-frequency high-definition cameras at the same timestamp are compared to see whether they are consistent. If so, the data collection of each high-frequency high-definition camera is synchronized. Otherwise, the data collection of each high-frequency high-definition camera is not synchronized.

5. A system for verifying the synchronization of data acquisition based on a high-frequency, high-definition camera, characterized by: A method for verifying the synchronization of data acquisition based on a high-frequency high-definition camera using any one of claims 1 to 4 above.

Images