Data acquisition method and device, electronic equipment and storage medium
By performing object detection and scene matching on autonomous driving video frame sequences and combining them with vehicle signal data, the accuracy and stability of autonomous driving data collection are improved, solving the problem of low collection efficiency in existing technologies.
Patent Information
- Application Number
- CN202510847742.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-24
- Publication Date
- 2025-09-26
AI Technical Summary
In the existing technology, autonomous driving data collection in special scenarios has the problems of low data collection quality, low accuracy and stability, resulting in low collection efficiency.
By acquiring the current frame image in the target video frame sequence, object detection is performed, object categories are determined, the current scene is matched with the preset scene, vehicle status and driving behavior data are collected, and double matching is performed in combination with vehicle signal data to ensure the accuracy of scene recognition and precise data collection.
It improves the accuracy and stability of data collection, improves collection efficiency, and ensures the data quality of preset scenarios.
Smart Images
Figure CN120708133A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of autonomous driving data collection, and in particular to a data collection method, device, electronic device, and storage medium. Background Art
[0002] With the continuous development of autonomous driving technology, the demand for autonomous driving data is also growing. How to achieve fast and accurate collection of autonomous driving data has also become the focus of improving the efficiency of related model training and optimizing algorithms.
[0003] In related technologies, manual recognition is often used to trigger the collection of driving data in special scenarios (highway scenarios, parking scenarios, extreme weather scenarios, etc.). However, in related technologies, manual recognition is prone to omissions and inaccuracies, and the data volume is large and the collection is unstable, resulting in low data collection quality, low data collection accuracy and stability, and thus low collection efficiency. Summary of the Invention
[0004] The present disclosure provides a data collection method, device, electronic device, and storage medium to at least address the problems of low data collection quality, low accuracy and stability, and low collection efficiency in special scenarios in related technologies. The technical solutions of the present disclosure are as follows: According to a first aspect of an embodiment of the present disclosure, there is provided a data collection method, comprising: Get the current frame image in the target video frame sequence; Performing object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; determining a current scene corresponding to the current frame image based on the object category corresponding to each of the at least one object; Matching the current scene with at least one preset scene to obtain a first matching result; In a case where the first matching result indicates that the current scene is a target scene, current vehicle state data and current driving behavior data are collected, and the target scene is any one of the at least one preset scene.
[0005] In an optional embodiment, the method further includes: Obtaining current vehicle signal data and preset conditions corresponding to at least one preset scenario; obtaining a second matching result based on the current vehicle signal data and the at least one preset condition; When the first matching result indicates that the current scene is the target scene, collecting current vehicle state data and current driving behavior data includes: When the first matching result indicates that the current scene is the target scene, and the second matching result indicates that the current scene is the target scene, current vehicle state data and current driving behavior data are collected.
[0006] In an optional embodiment, obtaining a second matching result based on the current vehicle signal data and the at least one preset condition includes: determining a current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition; Based on the current condition, a second matching result is obtained.
[0007] In an optional embodiment, the vehicle signal data includes at least one signal data, the at least one preset condition includes at least one preset signal threshold corresponding to each of the at least one preset scenarios, and determining the current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition includes: Obtaining a current signal threshold reached by the at least one signal data based on at least one preset signal threshold corresponding to each of the at least one signal data and the at least one preset scenario; Based on the current signal threshold, a current condition corresponding to the current vehicle signal data is obtained.
[0008] In an optional embodiment, performing object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image includes: Inputting the current frame image into a preset detection model to obtain a confidence score corresponding to at least one object in the current frame image; Based on the preset confidence threshold and the at least one confidence level, an object category corresponding to each of the at least one object is obtained.
[0009] In an optional embodiment, the method further includes: In response to a data collection completion instruction, obtaining current vehicle state data and current driving behavior data; The current vehicle status data and the current driving behavior data are labeled and stored.
[0010] In an optional embodiment, the method is applied to a preset terminal including a first mainboard and a second mainboard, the method is performed by the first mainboard and the second mainboard, and the method further includes: The first mainboard obtains performance parameters corresponding to the second mainboard; The first mainboard adjusts the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range, and based on the adjusted execution task volume, repeats the cyclic iterative steps of the first mainboard obtaining the performance parameters corresponding to the second mainboard, and the first mainboard adjusting the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range until the performance parameters are in the preset parameter range.
[0011] According to a second aspect of an embodiment of the present disclosure, there is provided a data acquisition device, comprising: A current frame image acquisition module is used to acquire the current frame image in the target video frame sequence; an object detection module, configured to perform object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; a current scene determining module, configured to determine a current scene corresponding to the current frame image based on the object category corresponding to each of the at least one object; a scene matching module, configured to match the current scene with at least one preset scene to obtain a first matching result; A data acquisition module is used to collect current vehicle status data and current driving behavior data when the first matching result indicates that the current scene is a target scene, and the target scene is any preset scene among the at least one preset scene.
[0012] In an optional embodiment, the device further comprises: A preset condition acquisition module, used to obtain the preset conditions corresponding to the current vehicle signal data and at least one preset scenario; a second matching result determination module, configured to obtain a second matching result based on the current vehicle signal data and the at least one preset condition; The data acquisition module includes: A data collection unit is configured to collect current vehicle state data and current driving behavior data when the first matching result indicates that the current scene is the target scene and the second matching result indicates that the current scene is the target scene.
[0013] In an optional embodiment, the second matching result determination module includes: a current condition determining unit, configured to determine a current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition; The second matching result determining unit is configured to obtain a second matching result based on the current condition.
[0014] In an optional embodiment, the vehicle signal data includes at least one signal data, the at least one preset condition includes at least one preset signal threshold corresponding to each of the at least one preset scenarios, and the current condition determination unit includes: a current signal threshold determination subunit, configured to obtain a current signal threshold reached by the at least one signal data based on the at least one signal data and the at least one preset signal threshold corresponding to each of the at least one preset scenario; The current condition determination subunit is configured to obtain the current condition corresponding to the current vehicle signal data based on the current signal threshold.
[0015] In an optional embodiment, the object detection module includes: a current confidence determination unit, configured to input the current frame image into a preset detection model to obtain a confidence corresponding to at least one object in the current frame image; The confidence filtering unit is configured to obtain the object category corresponding to at least one object based on a preset confidence threshold and the at least one confidence.
[0016] In an optional embodiment, the device further comprises: a data acquisition module, configured to acquire current vehicle status data and current driving behavior data in response to a data acquisition completion instruction; The data labeling and storage module is used to label and store the current vehicle status data and the current driving behavior data.
[0017] In an optional embodiment, the method is applied to a preset terminal including a first mainboard and a second mainboard, the method is performed by the first mainboard and the second mainboard, and the apparatus further includes: A performance parameter acquisition module, configured for the first motherboard to acquire performance parameters corresponding to the second motherboard; The loop iteration step repeats the module, wherein the first mainboard adjusts the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range, and repeats the loop iteration step of the first mainboard obtaining the performance parameters corresponding to the second mainboard based on the adjusted execution task volume, and the first mainboard adjusts the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range, until the performance parameters are in the preset parameter range.
[0018] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement a method as described in any one of the first aspects above.
[0019] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any method described in the first aspect of the embodiment of the present disclosure.
[0020] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects: Acquire a current frame image in a target video frame sequence; perform object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; determine a current scene corresponding to the current frame image based on the object category corresponding to at least one object; match the current scene with at least one preset scene to obtain a first matching result; when the first matching result indicates that the current scene is any one of the at least one preset scene, collect current vehicle status data and current driving behavior data, which can ensure the data collection quality of the preset scene, improve the accuracy and stability of data collection, and thus improve the collection efficiency.
[0021] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.
[0023] Figure 1 is a schematic diagram showing an application environment according to an exemplary embodiment; Figure 2 is a flow chart showing a data collection method according to an exemplary embodiment; Figure 3 is a block diagram of a data acquisition device according to an exemplary embodiment; Figure 4 is a block diagram of an electronic device for data collection according to an exemplary embodiment; Figure 5 It is a block diagram of an electronic device for data collection according to an exemplary embodiment. DETAILED DESCRIPTION
[0024] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
[0025] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish between similar and different contents, and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.
[0026] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.
[0027] See also Figure 1 , Figure 1 is a schematic diagram showing an application environment according to an exemplary embodiment. Figure 1 As shown, the application environment may include a terminal 100 and a server 200 .
[0028] In an optional embodiment, the terminal 100 can be used to provide data collection services. Specifically, the terminal 100 may include, but is not limited to, electronic devices such as smartphones, desktop computers, tablet computers, laptop computers, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, and smart wearable devices. It may also be software running on these electronic devices, such as applications. Optionally, the operating system running on the electronic device may include, but is not limited to, Android, iOS, Linux, Windows, etc.
[0029] In an optional embodiment, the server 200 may provide background services for the terminal 100 and perform data collection and processing. Optionally, the server 200 may match the current scene corresponding to the current frame image with at least one preset scene to obtain a first matching result. Accordingly, when the first matching result indicates that the current scene is any of the at least one preset scene, the current vehicle status data and the current driving behavior data may be collected. Specifically, the server 200 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.
[0030] In addition, it should be noted that Figure 1 What is shown is only one application environment provided by the present disclosure. In actual applications, other application environments may also be included. For example, data collection and processing may also be implemented in a terminal.
[0031] In the embodiments of this specification, the terminal 100 and the server 200 may be directly or indirectly connected via wired or wireless communication, which is not limited in this disclosure.
[0032] Figure 2 is a flow chart showing a data collection method according to an exemplary embodiment. The data collection method is used in a terminal or a server, and can also be executed interactively by the terminal and the server. Figure 2 As shown, the following steps are included.
[0033] In step S201 , a current frame image in a target video frame sequence is obtained.
[0034] In a specific embodiment, the target video frame sequence may be a video frame sequence that needs to be scene matched, and the current frame image may be an image frame corresponding to the target video frame sequence at the current moment.
[0035] In step S203 , object detection is performed on the current frame image to obtain an object category corresponding to at least one object in the current frame image.
[0036] In a specific embodiment, at least one object may be at least one target in the current frame image. Specifically, taking the target video frame sequence as the front-view camera video data as an example, the current frame image may be the front-view camera image frame at the current moment, and at least one object may be at least one target in the front-view camera image frame at the current moment. Optionally, the at least one target may include road signs, pedestrians, vehicles, etc.
[0037] In an optional embodiment, performing object detection on the current frame image to obtain the object category corresponding to at least one object in the current frame image may include: Inputting the current frame image into a preset detection model to obtain the confidence level corresponding to at least one object in the current frame image; Based on a preset confidence threshold and at least one confidence level, an object category corresponding to each of the at least one object is obtained.
[0038] In a specific embodiment, a preset detection model can be used to perform confidence detection. Specifically, taking the current frame image as the front-view camera image frame at the current moment as an example, the confidence corresponding to at least one object can be (ID=001, "car", 0.8), (ID=002, "road marking", 0.8), (ID=003, "speed limit sign", 0.9), (ID=004, "pedestrian", 0.3), etc.
[0039] In a specific embodiment, a preset confidence threshold can be used to determine the object category corresponding to at least one object. Specifically, taking the preset confidence threshold of 0.75 as an example, when the confidence corresponding to at least one object is (ID=001, "car", 0.8), (ID=002, "road marking", 0.8), (ID=003, "speed limit sign", 0.9), (ID=004, "pedestrian", 0.3), based on the preset confidence threshold, at least one confidence is filtered to obtain the object category corresponding to at least one object (ID=001, "car", 0.8), (ID=002, "road marking", 0.8), (ID=003, "speed limit sign", 0.9).
[0040] In the above embodiment, by presetting the confidence threshold, filtering the confidence corresponding to at least one object, and obtaining the object category corresponding to at least one object, the reliability of the output object category can be guaranteed, invalid data processing can be reduced, and data collection efficiency can be improved.
[0041] In step S207 , a current scene corresponding to the current frame image is determined based on the object category corresponding to each of the at least one objects.
[0042] In a specific embodiment, the current scene is the scene corresponding to the current frame image. Taking at least one object corresponding to the object category of (ID=001, "car", 0.8), (ID=002, "road marking", 0.8), (ID=003, "speed limit sign", 0.9), (ID=004, "tunnel entrance", 0.9) as an example, it can be determined that the current scene corresponding to the current frame image is a highway scene.
[0043] In step S209, the current scene is matched with at least one preset scene to obtain a first matching result.
[0044] In a specific embodiment, the at least one preset scene may include at least one scene that requires manual identification. Specifically, the at least one preset scene may include a highway scene, a parking scene, an extreme weather scene, etc.
[0045] In a specific embodiment, the first matching result may be used to indicate whether the current scene is a target scene, and the target scene may be any preset scene among at least one preset scene.
[0046] In step S2011, when the first matching result indicates that the current scene is the target scene, current vehicle state data and current driving behavior data are collected.
[0047] In a specific embodiment, taking at least one preset scene as a parking scene, a highway scene, and an extreme weather scene as an example, when the first matching result indicates that the current scene is a target scene, the target scene may be a highway scene.
[0048] In a specific embodiment, the current vehicle status data may be the vehicle operating parameters corresponding to the current moment. Specifically, the current vehicle status data may include vehicle speed, acceleration, tire pressure, etc. The current driving behavior data may be the behavior data generated by the driving vehicle corresponding to the current moment. Specifically, the current driving behavior data may include the number of sudden accelerations, the number of sudden brakes, the number of speeding, etc.
[0049] In an optional embodiment, the above method may further include: Obtaining current vehicle signal data and preset conditions corresponding to at least one preset scenario; obtaining a second matching result based on the current vehicle signal data and at least one preset condition; When the first matching result indicates that the current scene is the target scene, collecting the current vehicle state data and the current driving behavior data includes: When the first matching result indicates that the current scene is the target scene, and the second matching result indicates that the current scene is the target scene, current vehicle state data and current driving behavior data are collected.
[0050] In a specific embodiment, the current vehicle signal data may be the vehicle signal data corresponding to the current moment. Specifically, the current vehicle signal data may be the wiper status, steering wheel angle, headlight operating status, etc. The preset conditions corresponding to at least one preset scene may include the preset trigger conditions corresponding to any preset scene in at least one preset scene.
[0051] In a specific embodiment, the current vehicle signal data is vehicle speed = 100 km / h, accelerator pedal depth change rate = 8% / s, steering wheel angle fluctuation <5° / 10 seconds, and the preset conditions corresponding to the highway scene are vehicle speed ≥ 80 km / h, accelerator pedal depth change rate <10% / s, steering wheel angle fluctuation <5° / 10 seconds, the preset conditions corresponding to the parking scene are vehicle speed ≤ 5 km / h, single gear shift interval <10s, steering wheel angle fluctuation > 180° / 10 seconds, and the preset conditions corresponding to the heavy rain scene are wiper running status (interval time <2 seconds). The second matching result indicates whether the current scene is the target scene. Accordingly, the target scene is the highway scene.
[0052] In a specific embodiment, taking at least one preset scene as a parking scene, a highway scene, and an extreme weather scene, and the target scene as a highway scene as an example, when the first matching result indicates that the current scene is a highway scene, and the second matching result indicates whether the current scene is a highway scene, the current vehicle status data and the current driving behavior data are collected.
[0053] In the above embodiment, a second matching result is obtained based on the preset conditions corresponding to the current vehicle signal data and at least one preset scene. When the first matching result indicates that the current scene is the target scene, and the second matching result indicates that the current scene is the target scene, the current vehicle status data and the current driving behavior data are collected. The confidence of scene recognition can be improved through double matching, and the risks brought by single matching can be reduced, thereby improving the effectiveness of the collected data.
[0054] In an optional embodiment, obtaining the second matching result based on the current vehicle signal data and at least one preset condition may include: determining a current condition triggered by the current vehicle signal data based on the current vehicle signal data and at least one preset condition; Based on the current conditions, a second matching result is obtained.
[0055] In a specific embodiment, the current condition may be a preset condition triggered by the current vehicle signal data. Specifically, if the current vehicle signal data is vehicle speed = 3 km / h, single gear shift interval <10s, steering wheel angle fluctuation >200° / 10 seconds, and the preset conditions corresponding to the highway scenario are vehicle speed ≥80 km / h, accelerator pedal depth change rate <10% / s, steering wheel angle fluctuation <5° / 10 seconds, the preset conditions corresponding to the parking scenario are vehicle speed ≤5 km / h, single gear shift interval <10s, steering wheel angle fluctuation >180° / 10 seconds, and the preset conditions corresponding to the heavy rain scenario are wiper running status (interval time <2 seconds), it can be determined that the current condition triggered by the current vehicle signal data is the preset condition corresponding to the parking scenario. Accordingly, the second matching result indicates that the current scenario is the parking scenario.
[0056] In the above embodiment, the second matching result is determined by the preset conditions triggered by the current vehicle signal data, which can realize on-demand collection, accurately filter out scenes that do not need to be collected, improve matching reliability, and reduce invalid data.
[0057] In an optional embodiment, the vehicle signal data may include at least one signal data, and the at least one preset condition may include at least one preset signal threshold corresponding to at least one preset scenario. The above-mentioned determination of the current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition may include: Obtaining a current signal threshold reached by the at least one signal data based on at least one preset signal threshold corresponding to each of the at least one signal data and the at least one preset scenario; Based on the current signal threshold, the current condition corresponding to the current vehicle signal data is obtained.
[0058] In a specific embodiment, taking at least one signal data as vehicle speed, accelerator pedal depth change rate, steering wheel angle fluctuation, single gear shift interval, wiper operating status, etc. as an example, at least one preset signal threshold corresponding to the highway scene may be vehicle speed ≥80km / h, accelerator pedal depth change rate <10% / s, steering wheel angle fluctuation <5° / 10 seconds; at least one preset signal threshold corresponding to the parking scene may be vehicle speed ≤5km / h, single gear shift interval <10s, steering wheel angle fluctuation >180° / 10 seconds; at least one preset signal threshold corresponding to the heavy rain scene may be wiper operating status (interval time <2 seconds). When the wiper operating status in at least one signal data is intermittent time <2 seconds, the current signal threshold reached by at least one signal data is the wiper operating status (intermittent time <2 seconds), and the current condition corresponding to the current vehicle signal data is the preset condition corresponding to the heavy rain scene.
[0059] In the above embodiment, based on at least one preset signal threshold corresponding to at least one preset scenario, the current condition triggered by at least one signal data in the current vehicle signal data is judged, and the triggering condition can be accurately judged through the mapping relationship between the threshold and the condition.
[0060] In an optional embodiment, the above method may further include: In response to a data collection completion instruction, obtaining current vehicle state data and current driving behavior data; The current vehicle status data and current driving behavior data are labeled and stored.
[0061] In a specific embodiment, the data collection completion instruction can be used to trigger the acquisition of current vehicle status data and current driving behavior data. Accordingly, the current vehicle status data and current driving behavior data corresponding to each moment in a preset time period can be manually or semi-automatically labeled and stored.
[0062] In the above embodiment, after data collection is completed, the current vehicle status data and the current driving behavior data are labeled and stored, which can improve data availability and utilization efficiency.
[0063] In an optional embodiment, the above method is applied to a preset terminal including a first mainboard and a second mainboard. The method is performed by the first mainboard and the second mainboard and may include: The first mainboard obtains performance parameters corresponding to the second mainboard; The first main board adjusts the execution task amounts corresponding to the first main board and the second main board respectively based on the performance parameters and the preset parameter range, and based on the adjusted execution task amounts, repeats the cyclic iterative steps of the first main board obtaining the performance parameters corresponding to the second main board, and adjusting the execution task amounts corresponding to the first main board and the second main board respectively based on the performance parameters and the preset parameter range until the performance parameters are in the preset parameter range.
[0064] In a specific embodiment, the first mainboard may be a control mainboard, and the second mainboard may be an execution mainboard. The first mainboard may be used to control the amount of execution tasks of the first mainboard and the second mainboard. The performance parameters may include CPU (Central Processing Unit) usage, memory occupancy, etc.
[0065] In a specific embodiment, the preset parameter interval can be a reference interval of the performance parameters corresponding to the second mainboard. Specifically, taking the preset parameter interval as CPU usage of (50%, 80%) as an example, specifically, when the CPU usage corresponding to the second mainboard is lower than 50%, the first mainboard can assign data collection tasks to the second mainboard. Optionally, when the CPU usage corresponding to the second mainboard is higher than 80%, the first mainboard suspends assigning data collection tasks to the second mainboard and shares the data collection tasks being executed by the second mainboard. Accordingly, based on the adjusted execution task volume, the first mainboard repeats the steps of obtaining the CPU usage corresponding to the second mainboard, and adjusting the execution task volumes corresponding to the first mainboard and the second mainboard based on the CPU usage corresponding to the second mainboard and the preset parameter interval CPU usage of (50%, 80%), until the CPU usage corresponding to the second mainboard is at (50%, 80%).
[0066] In the above embodiment, by setting up double-sided mainboard synchronous collection, the problem of mainboard heating and freezing caused by single-sided mainboard collection is avoided, and the execution task volume of the second mainboard is controlled in real time by the first mainboard, dynamic load balancing is achieved, the risk of mainboard heating is effectively reduced, the stability and reliability of the mainboard operation are improved, and the efficient data collection is ensured.
[0067] It can be seen from the technical solutions provided in the above embodiments of this specification that this specification obtains the current frame image in the target video frame sequence; performs object detection on the current frame image to obtain the object category corresponding to at least one object in the current frame image; determines the current scene corresponding to the current frame image based on the object category corresponding to at least one object; matches the current scene with at least one preset scene to obtain a first matching result; when the first matching result indicates that the current scene is any one of at least one preset scene, collects the current vehicle status data and the current driving behavior data, which can ensure the data collection quality of the preset scene, improve the accuracy and stability of data collection, and thus improve the collection efficiency.
[0068] Figure 3 FIG. 1 is a block diagram of a data acquisition device according to an exemplary embodiment. Figure 3 , the device comprises: The current frame image acquisition module 310 is used to acquire the current frame image in the target video frame sequence; The object detection module 330 is configured to perform object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; a current scene determination module 350 for determining a current scene corresponding to a current frame image based on an object category corresponding to at least one object; A scene matching module 370 is configured to match a current scene with at least one preset scene to obtain a first matching result; The data collection module 390 is used to collect current vehicle state data and current driving behavior data when the first matching result indicates that the current scene is a target scene, and the target scene is any preset scene of at least one preset scene.
[0069] In an optional embodiment, the above device further includes: A preset condition acquisition module, used to obtain the preset conditions corresponding to the current vehicle signal data and at least one preset scenario; a second matching result determination module, configured to obtain a second matching result based on the current vehicle signal data and at least one preset condition; The data acquisition module 390 includes: The data collection unit is used to collect current vehicle state data and current driving behavior data when the first matching result indicates that the current scene is the target scene and the second matching result indicates that the current scene is the target scene.
[0070] In an optional embodiment, the second matching result determination module includes: a current condition determination unit, configured to determine a current condition triggered by the current vehicle signal data based on the current vehicle signal data and at least one preset condition; The second matching result determining unit is configured to obtain a second matching result based on the current condition.
[0071] In an optional embodiment, the vehicle signal data includes at least one signal data, the at least one preset condition includes at least one preset signal threshold corresponding to each of at least one preset scenario, and the current condition determination unit includes: a current signal threshold determination subunit, configured to obtain a current signal threshold reached by at least one signal data based on at least one signal data and at least one preset signal threshold corresponding to each of the at least one preset scenario; The current condition determination subunit is used to obtain the current condition corresponding to the current vehicle signal data based on the current signal threshold.
[0072] In an optional embodiment, the object detection module 330 includes: a current confidence determination unit, configured to input the current frame image into a preset detection model to obtain a confidence level corresponding to at least one object in the current frame image; The confidence filtering unit is configured to obtain an object category corresponding to at least one object based on a preset confidence threshold and at least one confidence.
[0073] In an optional embodiment, the above device further includes: a data acquisition module, configured to acquire current vehicle status data and current driving behavior data in response to a data acquisition completion instruction; The data labeling and storage module is used to label and store the current vehicle status data and current driving behavior data.
[0074] In an optional embodiment, the method is applied to a preset terminal including a first mainboard and a second mainboard, the method is performed by the first mainboard and the second mainboard, and the apparatus further includes: A performance parameter acquisition module, used by the first mainboard to acquire the performance parameters corresponding to the second mainboard; The loop iteration step repeats the module, the first main board adjusts the execution task volume corresponding to the first main board and the second main board based on the performance parameters and the preset parameter range, and based on the adjusted execution task volume, repeats the loop iteration step of the first main board obtaining the performance parameters corresponding to the second main board, and the first main board adjusts the execution task volume corresponding to the first main board and the second main board based on the performance parameters and the preset parameter range, until the performance parameters are in the preset parameter range.
[0075] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0076] Figure 4 This is a block diagram of an electronic device for data collection according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 4 As shown. The electronic device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a data acquisition method is implemented. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the electronic device, or an external keyboard, touchpad or mouse, etc. Figure 5 This is a block diagram of an electronic device for data collection according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5As shown. The electronic device includes a processor, a memory, and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a data acquisition method is implemented. Those skilled in the art will understand that Figure 4 or Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present disclosure, and does not constitute a limitation on the electronic device to which the scheme of the present disclosure is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, an electronic device is further provided, including: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the data collection method in the embodiment of the present disclosure.
[0077] In an exemplary embodiment, a computer-readable storage medium is further provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the data collection method in the embodiment of the present disclosure.
[0078] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.
[0079] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A data collection method, characterized in that: include: Get the current frame image in the target video frame sequence; Performing object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; determining a current scene corresponding to the current frame image based on the object category corresponding to each of the at least one object; Matching the current scene with at least one preset scene to obtain a first matching result; In a case where the first matching result indicates that the current scene is a target scene, current vehicle state data and current driving behavior data are collected, and the target scene is any one of the at least one preset scene.
2. A data collection method according to claim 1, characterized in that: The method further comprises: Obtaining current vehicle signal data and preset conditions corresponding to at least one preset scenario; obtaining a second matching result based on the current vehicle signal data and the at least one preset condition; When the first matching result indicates that the current scene is the target scene, collecting current vehicle state data and current driving behavior data includes: When the first matching result indicates that the current scene is the target scene, and the second matching result indicates that the current scene is the target scene, current vehicle state data and current driving behavior data are collected.
3. The data collection method according to claim 2, characterized in that: Obtaining a second matching result based on the current vehicle signal data and the at least one preset condition includes: determining a current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition; Based on the current condition, a second matching result is obtained.
4. The data collection method according to claim 3, characterized in that: The vehicle signal data includes at least one signal data, the at least one preset condition includes at least one preset signal threshold corresponding to each of the at least one preset scenarios, and determining the current condition triggered by the current vehicle signal data based on the current vehicle signal data and the at least one preset condition includes: Obtaining a current signal threshold reached by the at least one signal data based on at least one preset signal threshold corresponding to each of the at least one signal data and the at least one preset scenario; Based on the current signal threshold, a current condition corresponding to the current vehicle signal data is obtained.
5. The data collection method according to claim 1, wherein: The performing object detection on the current frame image to obtain the object category corresponding to at least one object in the current frame image includes: Inputting the current frame image into a preset detection model to obtain a confidence score corresponding to at least one object in the current frame image; Based on a preset confidence threshold and the at least one confidence level, an object category corresponding to each of the at least one object is obtained.
6. The data collection method according to any one of claims 1 to 5, characterized in that: The method further comprises: In response to a data collection completion instruction, obtaining current vehicle state data and current driving behavior data; The current vehicle status data and the current driving behavior data are labeled and stored.
7. The data collection method according to claim 1, characterized in that: The method is applied to a preset terminal including a first mainboard and a second mainboard, the method being executed by the first mainboard and the second mainboard, and further comprising: The first mainboard obtains performance parameters corresponding to the second mainboard; The first mainboard adjusts the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range, and based on the adjusted execution task volume, repeats the cyclic iterative steps of the first mainboard obtaining the performance parameters corresponding to the second mainboard, and the first mainboard adjusting the execution task volume corresponding to the first mainboard and the second mainboard respectively based on the performance parameters and the preset parameter range until the performance parameters are in the preset parameter range.
8. A data acquisition device, characterized in that: include: A current frame image acquisition module is used to acquire the current frame image in the target video frame sequence; an object detection module, configured to perform object detection on the current frame image to obtain an object category corresponding to at least one object in the current frame image; a current scene determining module, configured to determine a current scene corresponding to the current frame image based on the object category corresponding to each of the at least one object; a scene matching module, configured to match the current scene with at least one preset scene to obtain a first matching result; A data acquisition module is used to collect current vehicle status data and current driving behavior data when the first matching result indicates that the current scene is a target scene, and the target scene is any preset scene among the at least one preset scene.
9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the data collection method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the data collection method according to any one of claims 1 to 7.