Data processing method and apparatus, image processing device, and vehicle

Abstract

The embodiment of the application discloses a data processing method and device, an image processing device and a vehicle, the method comprises the following steps: acquiring a video stream collected based on an image collecting device installed on the vehicle during use of the vehicle, analyzing frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determining detection parameters, the detection parameters comprising an optical flow amplitude, an optical flow direction and a picture still time, and determining a falling detection result according to the detection parameters. Thus, by analyzing the optical flow change during use of the vehicle, the falling detection result can be determined by using visual information under a global perspective, various complex scenes can be coped with, and the accuracy and reliability of the falling detection result are improved.

Description

Technical Field

[0001] This invention relates to the field of two-wheeled vehicle technology, specifically to a data processing method and apparatus, an image processing device, and a vehicle. Background Technology

[0002] Falls are a major cause of accidents while riding two-wheeled vehicles, especially late at night or in deserted areas. In cases of severe falls, the lack of assistance can pose a serious risk of injury. Therefore, fall detection based on user behavior can help reduce this risk.

[0003] Existing fall detection methods often rely on the vehicle's heartbeat or location information reported by the user's cycling software to determine whether a user has fallen. However, this detection method has poor accuracy and reliability, leading to a high number of false positives and poor interpretability. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a data processing method and apparatus, an image processing device and a vehicle to improve the accuracy and reliability of fall detection results.

[0005] In a first aspect, embodiments of the present invention aim to provide a data processing method, the method comprising:

[0006] Acquire video streams captured during vehicle use, the video streams being acquired based on image acquisition devices installed on the vehicle;

[0007] The frame images of adjacent frames in the video stream are analyzed based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction and image stillness time.

[0008] The fall detection result is determined based on the detection parameters.

[0009] Furthermore, determining the fall detection result based on the detection parameters includes:

[0010] The detection parameters are input into a pre-trained fall detection model for processing to determine the fall detection result. The fall detection model is trained based on historical detection parameters and corresponding fall states.

[0011] Furthermore, determining the fall detection result based on the detection parameters includes:

[0012] Determine whether the detection parameters meet the preset judgment conditions;

[0013] In response to the detection parameters meeting the preset judgment conditions, the corresponding fall detection result is determined to be a fall.

[0014] Furthermore, determining the fall detection result based on the detection parameters includes:

[0015] The fall detection result is determined based on the detection parameters and sensor parameters, wherein the sensor parameters are determined based on the accelerometer or gyroscope on the vehicle.

[0016] Furthermore, the frame images of adjacent frames in the video stream are analyzed based on the preset optical flow calculation method to determine the detection parameters, including:

[0017] The frame images in the video stream are preprocessed, including image grayscale conversion and smoothing.

[0018] The preprocessed adjacent frame images are analyzed based on a preset optical flow calculation method to determine the detection parameters.

[0019] Furthermore, the optical flow calculation method employs either the Lucas-Kanade method or the Farneback method.

[0020] Furthermore, the method also includes:

[0021] In response to the fall detection result indicating a fall, a reporting data is sent to initiate assistance to the fallen person based on the reported data. The reported data includes location information, image data, vehicle information, and order information.

[0022] Secondly, embodiments of the present invention aim to provide a data processing apparatus, the apparatus comprising:

[0023] The acquisition module is used to acquire video streams collected during vehicle use, the video streams being acquired based on image acquisition devices installed on the vehicle;

[0024] The analysis module is used to analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determine the detection parameters, including optical flow amplitude, optical flow direction and image stillness time; and determine the fall detection result based on the detection parameters.

[0025] Thirdly, embodiments of the present invention aim to provide an image processing device, the image processing device comprising:

[0026] The image acquisition module is configured to acquire video streams during vehicle use;

[0027] The processing module is configured to acquire the video stream; analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction and frame stillness time; and determine the fall detection result based on the detection parameters.

[0028] Furthermore, the image processing device also includes:

[0029] The communication module is configured to send reporting data in response to the fall detection result indicating a fall, so as to initiate assistance to the fallen person based on the reporting data, the reporting data including location information, image data, vehicle information and order information.

[0030] Fourthly, embodiments of the present invention aim to provide a vehicle, the vehicle comprising:

[0031] Vehicle body;

[0032] An image acquisition unit, disposed on the vehicle body, is configured to acquire video streams during vehicle use;

[0033] The processing unit, disposed on the vehicle body, is configured to acquire the video stream; analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, the detection parameters including optical flow amplitude, optical flow direction and frame stillness time; and determine the fall detection result based on the detection parameters.

[0034] Furthermore, the image acquisition unit and the processing unit can be integrated or separated.

[0035] Fifthly, embodiments of the present invention aim to provide a data processing system, the system comprising:

[0036] The vehicle is configured to acquire a video stream during vehicle use, the video stream being acquired based on an image acquisition device installed on the vehicle; analyze adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, the detection parameters including optical flow amplitude, optical flow direction, and frame stillness time; determine a fall detection result based on the detection parameters; and in response to the fall detection result being a fall, send reporting data, the reporting data including location information, image data, vehicle information, and order information;

[0037] The server is configured to receive the reported data and send a prompt message to the relevant terminal based on the reported data, so as to prompt the corresponding personnel to initiate rescue for the fallen person.

[0038] Sixthly, embodiments of the present invention aim to provide a computer program product, the computer program product including a computer program / instruction, which, when executed by a processor, implements the method as described in any of the preceding claims.

[0039] In a seventh aspect, embodiments of the present invention aim to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in any of the preceding claims.

[0040] The technical solution of this invention acquires a video stream captured by an image acquisition device installed on the vehicle during vehicle use, analyzes the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determines detection parameters including optical flow amplitude, optical flow direction, and image stillness time, and determines the fall detection result based on the detection parameters. It can determine the fall detection result by analyzing the changes in optical flow during vehicle use and using visual information from a global perspective, and can cope with various complex scenarios, thereby improving the accuracy and reliability of the fall detection result. Attached Figure Description

[0041] The above and other objects, features and advantages of the present invention will become clearer from the following description of embodiments of the invention with reference to the accompanying drawings, in which:

[0042] Figure 1 This is a flowchart of the data processing method according to an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of an image processing device according to an embodiment of the present invention;

[0044] Figure 3 This is a flowchart of the image processing device according to an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of a vehicle according to an embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the data processing system according to an embodiment of the present invention;

[0047] Figure 6 This is a schematic diagram of a data processing apparatus according to an embodiment of the present invention. Detailed Implementation

[0048] The present application is described below based on embodiments, but it is not limited to these embodiments. In the detailed description of the present application below, certain specific details are described in detail. Those skilled in the art can fully understand the present application without these details. To avoid obscuring the substance of the present application, well-known methods, processes, flows, elements, and circuits are not described in detail.

[0049] Furthermore, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0050] Unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0051] For ease of explanation, spatially related terms such as “inside,” “outside,” “below,” “below,” “lower,” “above,” “upper,” etc., are used herein to describe the relationship between one element or feature illustrated in the figure and another. It will be understood that spatially related terms may be intended to encompass different orientations of the device in use or operation besides those depicted in the figure. For example, if the device in the figure is flipped, an element described as “below” or “below” another element or feature would then be positioned “above” that other element or feature. Thus, the exemplified term “below” can encompass both above and below orientations. The device may be oriented in other ways (rotated 90 degrees or in other orientations), and the spatially related descriptive terms used herein should be interpreted accordingly.

[0052] Unless the context explicitly requires it, words such as "including" or "contains" throughout the application should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".

[0053] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0054] The solutions described in this specification and embodiments, if involving information acquisition, will collect data under legal and compliant conditions, ensuring the legality of the data source, and will take appropriate technical and management measures to ensure data security. If involving personal information processing, processing will be carried out under legal grounds (e.g., obtaining the consent of the personal information subject, or being necessary for contract performance), and will only be conducted within the prescribed or agreed scope. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0055] It should also be noted that the vehicle-side or user-related data involved in this embodiment are determined with the user's authorization, in order to provide convenience to users while protecting their privacy and thereby improving their user experience. Furthermore, this application acquires, collects, stores, uses, processes, transmits, provides, and presents data in accordance with relevant laws and regulations, without infringing on the privacy of others, and without violating public order and good morals.

[0056] Existing two-wheeled vehicle fall detection methods primarily analyze vehicle position data, sensor data such as onboard gyroscopes, and dwell time, and determine whether a user has fallen by establishing a fall feature strategy. However, this method is limited by the sensors' reliance on localized data and environmental blind spots, resulting in a relatively high false positive rate and an inability to handle complex scenarios, such as users riding on severely bumpy roads or during prolonged periods of normal riding. Therefore, this invention aims to provide a data processing method to improve the accuracy and reliability of fall detection results.

[0057] Figure 1 This is a flowchart of a data processing method according to an embodiment of the present invention. Figure 1 As shown, the fall detection result is determined using the following method in this embodiment.

[0058] In step S110, a video stream is acquired during vehicle use, the video stream being acquired based on an image acquisition device installed on the vehicle.

[0059] Optionally, the image acquisition device in this embodiment can be installed in the vehicle's basket, handlebars, or other convenient locations on the vehicle for capturing video streams of the riding process. The image acquisition device automatically activates its image acquisition function when the vehicle is first used and continuously acquires video streams during vehicle use. The acquired video stream can be a forward-facing image of the vehicle (i.e., the direction the vehicle is traveling) or a video image of the rider, and the video stream includes multiple consecutive frames. The vehicle's start-of-use time can be represented by the start time of the ride order or the vehicle unlocking time.

[0060] Furthermore, the image acquisition device in this embodiment can be a height- and / or angle-adjustable image acquisition device. Each time the vehicle receives a user's request to use the vehicle, the image acquisition device automatically calibrates its settings (such as setting the height or shooting angle). For example, the settings can be adjusted to default standard values ​​(e.g., 90cm from the ground, shooting angle aligned with the front wheel's orientation). Alternatively, the settings can be adjusted based on the rider's height and other attributes to ensure that the image acquisition device can capture video streams with valid information during different riders' vehicle use, thereby improving the accuracy and reliability of fall detection results determined based on the video stream.

[0061] In step S120, the frame images of adjacent frames in the video stream are analyzed based on a preset optical flow calculation method to determine the detection parameters. The detection parameters include optical flow amplitude, optical flow direction, and frame stillness time.

[0062] Because the image acquisition device moves rapidly and violently during a fall, resulting in significant changes in optical flow in the captured image, the fall detection result can be determined based on the changes in optical flow (i.e., the movement of pixel features in the image) in the video stream acquired by the image acquisition device. Simultaneously, this embodiment analyzes the frame image set of adjacent frames in the video stream using a preset optical flow calculation method, calculates the movement of pixel features in adjacent frame images to determine detection parameters, and uses these detection parameters to reflect the motion state of the person or object in the image.

[0063] Among the detection parameters, optical flow amplitude represents the distance a pixel feature moves in the image, which can be expressed as the average or standard deviation of the distance. Optical flow direction represents the direction of movement of the pixel feature in the image. Frame stillness time represents the time during which a pixel feature remains stationary in the image.

[0064] Optionally, to improve the accuracy of the detection parameters, in this embodiment, before performing optical flow analysis on the video stream, the frame images in the video stream are preprocessed, and then the frame images of adjacent preprocessed frames are analyzed based on a preset optical flow calculation method to determine the detection parameters.

[0065] Furthermore, the preprocessing in this embodiment may include image grayscale conversion, smoothing, and other image preprocessing operations. Image grayscale conversion is used to convert each frame of the image into a grayscale image. Smoothing can employ Gaussian blur or other smoothing techniques to reduce the impact of noise in the frame image on optical flow analysis. Therefore, by preprocessing each frame of the image in this embodiment, the pixel features in the image can be made more apparent, allowing the image optical flow analysis to fully and effectively utilize the pixel features, thereby improving the accuracy of the detection parameter determination and further enhancing the overall accuracy of the fall detection results.

[0066] Optionally, the preset optical flow calculation method in this embodiment can be either the Lucas-Kanade method or the Farneback method. The Lucas-Kanade (LK) method is an efficient sparse optical flow estimation algorithm that calculates the motion vectors of feature points in the image by assuming pixel motion consistency within a local window. The Farneback method is a dense optical flow estimation algorithm based on polynomial expansion that calculates the motion vectors of all pixels in the image by modeling brightness changes within a local neighborhood.

[0067] Specifically, when determining detection parameters based on the Lucas-Kanade method, sparse feature points (such as corner points) in the frame image are detected. Within a local window, spatial and temporal gradients are used to construct the optical flow equation, and weighted least squares is used to solve for the displacement vector. The optical flow amplitude is the magnitude of the displacement vector, and the optical flow direction is the angle of the displacement vector. In contrast, when determining detection parameters based on the Farneback method, dense optical flow calculations are performed on global pixels in the frame image. Gaussian pyramid hierarchical optimization is used, and a quadratic polynomial motion model is used to fit the neighborhood gradient distribution. The gradient distribution within the neighborhood is used to estimate the displacement field vector of each pixel, and the corresponding optical flow amplitude and direction are determined based on the magnitude and angle of this vector.

[0068] Therefore, in this embodiment, optical flow analysis of frame images in the video stream is performed using the above-mentioned optical flow calculation method to determine detection parameters. This fully and effectively utilizes pixel features, thereby improving the accuracy of the determination of detection parameters and enhancing the accuracy and reliability of subsequent fall detection results based on the detection parameters.

[0069] In step S130, the fall detection result is determined based on the detection parameters.

[0070] Optionally, in this embodiment, the fall detection result can be determined solely based on the detection parameters obtained from optical flow analysis; alternatively, the fall detection result can be determined simultaneously based on the detection parameters obtained from optical flow analysis and sensor parameters collected by other types of sensors on the vehicle, such as accelerometers and gyroscopes. This allows for the use of multimodal information for fall judgment, further improving the accuracy of fall detection.

[0071] Furthermore, in one optional implementation of determining the fall detection result based on the detection parameters, this embodiment determines the fall detection result by processing the detection parameters input into a pre-trained fall detection model. The fall detection model is trained and determined based on historical detection parameters and corresponding fall states.

[0072] Optionally, the fall detection model in this embodiment can employ CNN+LSTM or other neural network models with classification capabilities to determine the fall detection result corresponding to the detection parameters through classification. However, it should be understood that the fall detection model type given here is merely an example and does not limit the specific type of fall detection model.

[0073] In another optional implementation, this embodiment can also determine the fall detection result based on detection parameters and preset judgment conditions. Specifically, by judging whether the detection parameters meet the preset judgment conditions, and in response to the detection parameters meeting the preset judgment conditions, the corresponding fall detection result is determined to be a fall. Optionally, the preset judgment conditions in this embodiment can be at least one of the following: optical flow amplitude is greater than a preset amplitude value, optical flow direction is a preset direction, and image stillness time exceeds a preset duration. For example, the corresponding fall detection result is determined to be a fall when the optical flow amplitude is greater than the preset amplitude value, the optical flow direction is a preset direction, or the image stillness time exceeds the preset duration; or, the fall detection result is determined to be a fall when the optical flow amplitude in the preset direction is greater than the preset amplitude value.

[0074] Furthermore, when determining the fall detection result based on the detection parameters obtained from optical flow analysis and the sensor parameters collected by the sensors on the vehicle, in one optional implementation, this embodiment can first use the above method to determine the first fall detection result based on the detection parameters, then use the sensor parameters to verify the detection parameters, and when both the first fall detection result and the sensor parameters indicate that the fall detection result is a fall, the final fall detection result is determined to be a fall.

[0075] In another optional implementation, this embodiment can pre-set confidence weights and confidence thresholds. When determining the fall detection result, the first fall detection result is determined using the above method based on the detection parameters, and the second fall detection result is determined using the sensor parameters. Then, the confidence weights corresponding to the first and second fall detection results, representing a fall, are summed to determine the result confidence. Finally, this result confidence is compared with the confidence threshold. When the result confidence reaches the confidence threshold, the corresponding fall detection result is determined to be a fall. Optionally, the confidence weight of the first fall detection result determined based on the detection parameters is greater than the confidence weight of the second fall detection parameter determined based on the sensor parameters. Therefore, this embodiment, by using the above method to determine the fall detection result using detection parameters and sensor parameters, can utilize multimodal information for fall judgment, which is beneficial to further improve the accuracy of fall detection.

[0076] Furthermore, to reduce the risk of injury to users after a fall, the data processing method in this embodiment also includes: in response to a fall detection result, sending reporting data to initiate assistance to the fallen person based on the reported data. The reported data may include fall detection results, location information, image data, vehicle information, order information, and other information related to vehicle use. Location information refers to the vehicle's location, and image data may be a corresponding video stream or a sequence of frames within a certain time period, including the moment the cyclist falls, from the video stream. Vehicle information includes vehicle identification and other vehicle-related attribute information, such as the vehicle's usage period. Order information includes order start time, order user identification, and other information related to the ride-hailing order.

[0077] Specifically, after detecting a user's fall, the vehicle in this embodiment will send a report to the operation platform server so that the operation platform can take necessary rescue actions based on the report data, including making outbound phone calls, contacting emergency medical services, contacting emergency contacts, and on-site rescue by city operation personnel, thereby initiating assistance to the fallen person and reducing the risk of injury to the user after the fall.

[0078] Optionally, after receiving the reported data, the operating platform in this embodiment can determine potential rescuers near the fallen person based on the location information in the reported data. For example, other cyclists or operators within a certain distance (e.g., 100m) from the location information can be identified as potential rescuers, and prompts can be sent to these potential rescuers to prompt them to initiate rescue. Further, after receiving feedback from potential rescuers within the current location range, the platform follows up on the rescue process through interaction with these candidates. If no feedback is received from potential rescuers within the current location range, the current location range is expanded, and prompts are sent to potential rescuers within the expanded range until feedback is received from them.

[0079] Alternatively, in this embodiment, after receiving the reported data or when there are no candidate rescuers within a preset location range or no feedback information from any candidate rescuers, the operating platform can send the fall detection results, location information, and image data from the reported data to the emergency contact of the fallen person, so that the emergency contact can be informed of the safety status of the fallen user in a timely manner and take appropriate action.

[0080] Optionally, the operating platform in this embodiment can determine the danger level of the fall based on the reported data after receiving it. When the danger level reaches a certain level, it can contact emergency services by telephone to arrange timely assistance for cyclists with more severe injuries and prevent the risk of further injury from escalating.

[0081] The technical solution of this invention acquires a video stream captured by an image acquisition device installed on the vehicle during vehicle use. Based on a preset optical flow calculation method, it analyzes adjacent frames in the video stream to determine detection parameters, including optical flow amplitude, optical flow direction, and frame stillness time. The solution then determines a fall detection result based on these parameters. By analyzing changes in optical flow during vehicle use and utilizing visual information from a global perspective, the fall detection result can be determined, addressing various complex scenarios and improving the accuracy and reliability of the fall detection. Furthermore, by sending a report to the operation platform when a fall is detected, the system can promptly notify the platform to intervene after a fall, thereby reducing the risk of injury to the rider and ensuring rider safety.

[0082] Figure 2 This is a schematic diagram of an image processing device according to an embodiment of the present invention. Figure 2 As shown, the image processing device in this embodiment includes an image acquisition module 21 and a processing module 22. The image acquisition module 21 is installed on the vehicle and configured to acquire video streams during vehicle use. The processing module 22 is configured to acquire the video stream, analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determine detection parameters, and determine the fall detection result based on the detection parameters. The detection parameters include optical flow amplitude, optical flow direction, and frame stillness time.

[0083] Furthermore, the image processing device in this embodiment also includes a communication module 23. The communication module 23 is configured to send reporting data in response to a fall detection result indicating a fall, so as to initiate assistance to the fallen person based on the reported data. The reported data includes location information, image data, vehicle information, and order information.

[0084] Optionally, to facilitate the operation platform to quickly initiate rescue for the fallen person, the communication module 23 in this embodiment is specifically configured to communicate with the operation platform and send reporting data through the communication connection with the operation platform server to notify relevant personnel of the operation platform to provide timely rescue for the fallen person and reduce the risk of injury to the fallen person.

[0085] Alternatively, since the central control unit in the vehicle has the function of communicating with external terminals or an operating platform, in this embodiment, the communication function between the vehicle's central control unit and the operating platform server can be used to send the reported data to the operating platform. In this case, the communication module 23 is specifically configured to communicate with the vehicle's central control unit and send the reported data through the communication connection with the central control unit. After receiving the reported data, the vehicle's central control unit sends the reported data to the operating platform, causing the operating platform to initiate rescue to the fallen person.

[0086] It should be noted that the method for determining the fall detection result in this embodiment has been described in detail above, and will not be repeated here.

[0087] Figure 3 This is a flowchart illustrating the operation of an image processing device according to an embodiment of the present invention. Figure 3 As shown, the image processing device in this embodiment can communicate directly with the operation platform server. During vehicle use, the image acquisition module 21 in the image processing device acquires the video stream during vehicle use and transmits the video stream to the processing module 22. The processing module 22 analyzes the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determines the detection parameters, and then inputs the detection parameters into a pre-trained fall detection model for classification processing to determine the corresponding fall detection result. When the fall detection result is a fall, the processing module 22 generates reporting data based on the location information and vehicle information of the vehicle, the image data acquired by the acquisition module, and the order information of the vehicle use order, and transmits the reporting data to the communication module 23, so that the reporting data can be sent to the operation platform through the communication module 23. After receiving the reporting data sent by the image processing device, the operation platform takes necessary rescue actions based on the reporting data, including telephone outbound calls, emergency contact, emergency contact person contact, and on-site rescue by city operation personnel, thereby initiating assistance to the fallen person and reducing the risk of injury to the user after a fall.

[0088] Figure 4 This is a schematic diagram of a vehicle according to an embodiment of the present invention. Figure 4 As shown, the vehicle in this embodiment includes a vehicle body 41, an image acquisition unit 42, and a processing unit 43. Both the image acquisition unit 42 and the processing unit 43 are mounted on the vehicle body. The image acquisition unit 42 is configured to acquire a video stream during vehicle use. The processing unit 43, mounted on the vehicle body, is configured to acquire the video stream; analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction, and frame stillness time; and determine the fall detection result based on the detection parameters.

[0089] Optionally, the image acquisition unit 42 and the processing unit 43 in this embodiment can be configured as an integrated unit or as separate units. Further, in this embodiment, the image acquisition unit 42 and the processing unit 43 can be integrated into the aforementioned image processing device, and this image processing device can be installed on the basket or handlebars of the vehicle body 41. This facilitates the image acquisition unit in acquiring video streams during vehicle use, thereby determining the fall detection results of the rider by performing optical flow analysis on the video stream, thus improving the accuracy and reliability of the fall detection results.

[0090] Alternatively, when the image acquisition unit 42 and the processing unit 43 can be set up separately, the image acquisition unit 42 can be a regular camera, and the processing unit 43 can be the vehicle's central control unit. By combining the regular camera with the vehicle's central control unit to determine the fall detection results, the accuracy and reliability of the fall detection results can be improved while reducing the deployment of additional equipment, which is beneficial to reducing the cost of fall detection.

[0091] Figure 5 This is a schematic diagram of a data processing system according to an embodiment of the present invention. Figure 5 As shown, the data processing system in this embodiment includes a vehicle 51 and a server 52. The vehicle 51 and server 52 establish a communication connection via a network 53. The vehicle 51 is configured to acquire video streams collected during vehicle use, the video streams being acquired using an image acquisition device installed on the vehicle; analyze adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction, and frame stillness time; determine a fall detection result based on the detection parameters; and, in response to a fall detection result, send reporting data, including location information, image data, vehicle information, and order information. The server 52 is configured to receive the reported data and, based on the reported data, send prompt information to relevant terminals to prompt corresponding personnel to initiate assistance to the fallen person.

[0092] It should be noted that the methods for determining the fall detection results and providing assistance to the fallen person based on the fall detection results in this embodiment have been described in the foregoing content and will not be repeated here.

[0093] Figure 6 This is a schematic diagram of a data processing apparatus according to an embodiment of the present invention. Figure 6 As shown, the data processing device in this embodiment includes an acquisition module 61 and an analysis module 62. The acquisition module 61 is used to acquire a video stream collected during vehicle use, the video stream being acquired using an image acquisition device installed on the vehicle. The analysis module 62 is used to analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determine detection parameters, including optical flow amplitude, optical flow direction, and frame stillness time; and determine the fall detection result based on the detection parameters.

[0094] Optionally, in this embodiment, the optical flow calculation method adopts the Lucas-Kanade method or the Farneback method. When determining the fall detection result based on the detection parameters, the analysis module 62 is also used to input the detection parameters into a pre-trained fall detection model for processing to determine the fall detection result. The fall detection model is trained and determined based on historical detection parameters and corresponding fall states; or it determines whether the detection parameters meet preset judgment conditions; in response to the detection parameters meeting the preset judgment conditions, the corresponding fall detection result is determined to be a fall.

[0095] Furthermore, when analyzing the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine the detection parameters, the analysis module 62 in this embodiment is also used to preprocess the frame images in the video stream, including image grayscale conversion and smoothing; and to analyze the preprocessed frame images of adjacent frames based on the preset optical flow calculation method to determine the detection parameters.

[0096] Alternatively, the analysis module 62 in this embodiment is also used to determine the fall detection result based on the detection parameters and sensor parameters, wherein the sensor parameters are determined based on the accelerometer or gyroscope on the vehicle.

[0097] Furthermore, in this embodiment, the data processing device is also used to send reporting data in response to a fall detection result, so as to initiate assistance to the fallen person based on the reporting data. The reporting data includes location information, image data, vehicle information, and order information.

[0098] The technical solution of this invention acquires a video stream collected by an image acquisition device installed on the vehicle during vehicle use using the aforementioned data processing device. It then analyzes adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction, and frame stillness time. Finally, it determines a fall detection result based on these detection parameters. This method analyzes changes in optical flow during vehicle use, utilizes visual information from a global perspective to determine the fall detection result, and addresses various complex scenarios, improving the accuracy and reliability of fall detection results.

[0099] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus (devices), or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0100] This application is described with reference to flowchart illustrations of methods, apparatus (devices), and computer program products according to embodiments of this application. It should be understood that each step in the flowchart can be implemented by computer program instructions. These computer program instructions may be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the steps. Figure 1 The function specified in one or more processes.

[0101] These computer program instructions may also be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, produce instructions for implementing processes. Figure 1 A device for a function specified in one or more processes.

[0102] Another embodiment of the present invention relates to a non-volatile storage medium for storing a computer-readable program for use by a computer to execute some or all of the above-described method embodiments.

[0103] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program specifying the relevant hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0104] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A data processing method, characterized in that, The method includes: Acquire video streams captured during vehicle use, the video streams being acquired based on image acquisition devices installed on the vehicle; The frame images of adjacent frames in the video stream are analyzed based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction and image stillness time. The fall detection result is determined based on the detection parameters.

2. The method according to claim 1, characterized in that, Determining the fall detection result based on the detection parameters includes: The detection parameters are input into a pre-trained fall detection model for processing to determine the fall detection result. The fall detection model is trained based on historical detection parameters and corresponding fall states.

3. The method according to claim 1, characterized in that, Determining the fall detection result based on the detection parameters includes: Determine whether the detection parameters meet the preset judgment conditions; In response to the detection parameters meeting the preset judgment conditions, the corresponding fall detection result is determined to be a fall.

4. The method according to claim 1, characterized in that, Determining the fall detection result based on the detection parameters includes: The fall detection result is determined based on the detection parameters and sensor parameters, wherein the sensor parameters are determined based on the accelerometer or gyroscope on the vehicle.

5. The method according to claim 1, characterized in that, The preset optical flow calculation method analyzes the frame images of adjacent frames in the video stream to determine the detection parameters, including: The frame images in the video stream are preprocessed, including image grayscale conversion and smoothing. The preprocessed adjacent frame images are analyzed based on a preset optical flow calculation method to determine the detection parameters.

6. The method according to claim 1, characterized in that, The optical flow calculation method adopts either the Lucas-Kanade method or the Farneback method.

7. The method according to claim 1, characterized in that, The method further includes: In response to the fall detection result indicating a fall, a reporting data is sent to initiate assistance to the fallen person based on the reported data. The reported data includes location information, image data, vehicle information, and order information.

8. A data processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire video streams collected during vehicle use, the video streams being acquired based on image acquisition devices installed on the vehicle; The analysis module is used to analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method, determine the detection parameters, including optical flow amplitude, optical flow direction and image stillness time; and determine the fall detection result based on the detection parameters.

9. An image processing device, characterized in that, The image processing device includes: The image acquisition module, installed on the vehicle, is configured to acquire video streams during vehicle use; The processing module is configured to acquire the video stream; analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, including optical flow amplitude, optical flow direction and frame stillness time; and determine the fall detection result based on the detection parameters.

10. The image processing apparatus according to claim 9, characterized in that, The image processing device further includes: The communication module is configured to send reporting data in response to the fall detection result indicating a fall, so as to initiate assistance to the fallen person based on the reporting data, the reporting data including location information, image data, vehicle information and order information.

11. A vehicle, characterized in that, The vehicles include: Vehicle body; An image acquisition unit, disposed on the vehicle body, is configured to acquire video streams during vehicle use; The processing unit, disposed on the vehicle body, is configured to acquire the video stream; analyze the frame images of adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, the detection parameters including optical flow amplitude, optical flow direction and frame stillness time; and determine the fall detection result based on the detection parameters.

12. The vehicle according to claim 11, characterized in that, The image acquisition unit can be integrated with or separated from the processing unit.

13. A data processing system, characterized in that, The system includes: The vehicle is configured to acquire a video stream during vehicle use, the video stream being acquired based on an image acquisition device installed on the vehicle; analyze adjacent frames in the video stream based on a preset optical flow calculation method to determine detection parameters, the detection parameters including optical flow amplitude, optical flow direction, and frame stillness time; determine a fall detection result based on the detection parameters; and in response to the fall detection result being a fall, send reporting data, the reporting data including location information, image data, vehicle information, and order information; The server is configured to receive the reported data and send a prompt message to the relevant terminal based on the reported data, so as to prompt the corresponding personnel to initiate rescue for the fallen person.

14. A computer program product, characterized in that, The computer program product includes a computer program / instruction that, when executed by a processor, implements the method of any one of claims 1-7.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1-7.

Images