Vehicle driving warning methods, devices, electronic equipment and storage media

Abstract

This invention provides a vehicle driving warning method, device, electronic device, and storage medium. The method includes: acquiring a seated image of a vehicle driver; identifying key points in the seated image; determining the current seated center of gravity using key points in the waist and shoulders; comparing the current seated center of gravity with a pre-determined initial seated center of gravity to generate a center of gravity deviation result; detecting whether the vehicle driver's seated posture is abnormal based on the center of gravity deviation result; and, in response to determining that the vehicle driver's seated posture is abnormal, providing a warning to the vehicle driver. This invention intelligently detects key points in the driver's waist and shoulders through seated image analysis to obtain the center of gravity position of the driver's shoulder-waist triangle. By using the seated center of gravity for posture judgment, it can quickly and accurately identify whether the driver's seated posture is abnormal, improve the accuracy of driving posture judgment, and automatically provide driving warnings to ensure the effectiveness of safe driving.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a vehicle driving warning method, device, electronic device, and storage medium. Background Technology

[0002] As an important means of transportation, while improving the comfort of the driving posture, driving safety is becoming increasingly important. Improper driving posture may cause safety risks. For example, if the driver makes large upper body movements while driving, such as turning the head, turning to the side, or bending over, it may easily lead to loss of vehicle control.

[0003] Currently, the judgment of dangerous driving actions by drivers is mainly based on visual detection, such as detecting the driver's head and hands on the steering wheel, or by installing pressure sensors on the seat to detect posture based on pressure changes. However, the existing driving detection methods have a large detection granularity, which leads to inaccurate identification and misjudgment, affecting the accuracy of driving detection and further affecting the effectiveness of safe driving. Summary of the Invention

[0004] In view of this, the present invention aims to propose a vehicle driving warning method, device, electronic device and storage medium to solve the problem that the current detection method is inaccurate and prone to misjudgment, which affects the accuracy of driving detection, and to achieve automatic and accurate vehicle driving warning.

[0005] According to a first aspect of the present invention, a vehicle driving warning method is provided, the method comprising:

[0006] Acquire images of the driver's sitting posture and identify key points in the images; wherein, the key points include key points in the waist and key points in the shoulders;

[0007] The current sitting posture center of gravity is determined using the key points of the waist and shoulders. The current sitting posture center of gravity is compared with the predetermined initial sitting posture center of gravity to generate a center of gravity deviation result.

[0008] Based on the centroid deviation results, detect whether the driver's sitting posture is abnormal;

[0009] In response to determining that the driver of the vehicle is in an abnormal sitting posture, a warning is issued to the driver of the vehicle.

[0010] Optionally, the step of acquiring a seated image of the vehicle driver and identifying key points in the seated image includes: key points in the waist and key points in the shoulders.

[0011] The vehicle camera is used to capture images of a preset area. The captured sample images are then input into a preset neural network model for training to obtain a key point detection model.

[0012] Obtain an image of the vehicle driver's current seating posture;

[0013] A key point detection model is used to identify the sitting posture image and determine the waist key points and shoulder key points of the sitting posture image.

[0014] Optionally, the step of determining the current sitting posture center of gravity using the waist key points and shoulder key points, and comparing the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity to generate a center of gravity deviation result includes:

[0015] Using the aforementioned key points of the waist and the aforementioned key points of the shoulders, a shoulder-waist triangle is generated;

[0016] Calculate the centroid of the shoulder-waist triangle and determine the centroid of the current sitting posture.

[0017] A rectangular coordinate system is constructed using the predetermined initial seated centroid as the origin, and the coordinates of the current seated centroid are determined.

[0018] Calculate the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid, and determine the horizontal and vertical distances of the centroid coordinates as the centroid deviation result.

[0019] Optionally, detecting whether the driver's seating posture is abnormal based on the centroid deviation result includes:

[0020] A first threshold in the horizontal direction and a second threshold in the vertical direction are pre-calibrated based on the actual vehicle.

[0021] If the horizontal distance between the current sitting posture centroid and the predetermined initial sitting posture centroid coordinates is less than or equal to the first threshold, and the vertical distance is less than or equal to the second threshold, then the driver's sitting posture is determined to be normal.

[0022] Otherwise, determine that the driver's sitting posture is abnormal.

[0023] Optionally, the step of providing a warning to the driver in response to determining that the driver's posture is abnormal includes:

[0024] In response to detecting an abnormal sitting posture of the vehicle driver, a warning warning command is generated and issued.

[0025] Based on the warning instruction, the vehicle control system shall use at least one of the following methods—voice reminder or text reminder—to provide a warning to the driver.

[0026] Record the number of warnings issued and the time of each warning.

[0027] Optionally, after recording the number of warning notifications and the time of the warning notifications, the method further includes:

[0028] If the number of warning prompts within a preset period exceeds a preset prompt threshold, a control command is sent to the active pretensioning seat belt.

[0029] The active pretensioning seat belt is controlled to tighten or vibrate, providing a warning to the vehicle driver.

[0030] Optionally, after issuing a warning to the driver in response to determining that the driver's posture is abnormal, the method further includes:

[0031] Monitor the changes in the driver's seating posture within a preset period;

[0032] If the driver's seating posture is detected to be normal within a preset period, the warning prompt to the driver will be stopped.

[0033] According to a second aspect of the present invention, a vehicle driving warning device is provided, the device comprising:

[0034] An image acquisition module is used to acquire images of the driver's sitting posture and identify key points in the images; wherein, the key points include waist key points and shoulder key points.

[0035] The center of gravity deviation module is used to determine the current sitting posture center of gravity using the waist key points and shoulder key points, compare the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generate a center of gravity deviation result.

[0036] The posture detection module is used to detect whether the driver's posture is abnormal based on the centroid deviation result.

[0037] The early warning module is used to provide an early warning to the driver of the vehicle in response to the determination that the driver's sitting posture is abnormal.

[0038] According to another aspect of the present invention, an electronic device is also provided, comprising:

[0039] processor;

[0040] Memory used to store the processor's executable instructions;

[0041] The processor is configured to execute the instructions to implement the vehicle driving warning method described above.

[0042] According to another aspect of the present invention, a readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the vehicle driving warning method as described above.

[0043] The vehicle driving warning method provided in this invention collects images of the driver's sitting posture, identifies key points in the images, determines the current sitting posture centroid using key points in the waist and shoulders, compares the current centroid with a pre-determined initial centroid to generate a centroid deviation result, and detects whether the driver's posture is abnormal based on the centroid deviation result. In response to determining that the driver's posture is abnormal, a warning is issued to the driver. This invention intelligently detects key points in the driver's waist and shoulders through sitting posture images, thereby obtaining the centroid position of the driver's shoulder-waist triangle. Using the sitting posture centroid for posture judgment can quickly and accurately identify whether the driver's sitting posture is abnormal, improving the accuracy of driving posture judgment. When an abnormal sitting posture is detected, i.e., dangerous driving behavior, automatic driving warning and alert actions are taken, reducing the probability of traffic accidents caused by dangerous driving behavior during vehicle operation and ensuring the effectiveness of safe driving.

[0044] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0045] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0046] Figure 1 This is a flowchart illustrating the steps of a vehicle driving warning method provided in an embodiment of the present invention;

[0047] Figure 2 yes Figure 1 A flowchart of step 101 in the vehicle driving warning method provided in this embodiment of the invention;

[0048] Figure 3 yes Figure 1 A flowchart of step 102 in the vehicle driving warning method provided in this embodiment of the invention;

[0049] Figure 4 yes Figure 1A flowchart of step 103 in the vehicle driving warning method provided in this embodiment of the invention;

[0050] Figure 5 This is a flowchart of another vehicle driving warning method provided in an embodiment of the present invention;

[0051] Figure 6 This is a flowchart of the vehicle driving warning method provided in the embodiments of the present invention;

[0052] Figure 7 This is a schematic diagram of the structure of a vehicle driving warning device provided in an embodiment of the present invention;

[0053] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are presented in the various embodiments of the present invention to facilitate a better understanding of this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments. The division of the various embodiments below is for ease of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined with and referenced by each other without contradiction.

[0055] Reference Figure 1 The diagram illustrates a flowchart of the vehicle driving warning method provided in an embodiment of the present invention, the method including:

[0056] Step 101: Acquire images of the driver's sitting posture and identify key points in the images; the key points include waist key points and shoulder key points.

[0057] In this embodiment of the invention, to address the shortcomings of current driving detection methods, such as large detection granularity, inaccurate identification, and susceptibility to misjudgment, which affect the accuracy of driving detection and further impact the effectiveness of safe driving, this embodiment refines the detection granularity by using the driver's sitting center of gravity to judge the driver's posture. This enables rapid and accurate identification of whether the driver's sitting posture is abnormal, improving the accuracy of driving posture judgment. When an abnormal sitting posture is detected, i.e., dangerous driving behavior, automatic driving warning and alert actions are taken, reducing the probability of traffic accidents caused by dangerous driving behavior during vehicle operation and ensuring the effectiveness of safe driving.

[0058] Specifically, in this embodiment, an in-vehicle camera captures images of the driver's seated posture. Through image processing and recognition, key points in the posture images are identified, particularly waist and shoulder key points. These waist and shoulder key points are important reference points for determining the center of gravity of the seated posture. The camera's field of view at least includes the driver's hips and above. It should be noted that the vehicle controller can pre-collect a large number of driver seated posture images. Using a pre-built neural network model, combined with the collected driver seated posture sample images, the model is trained to obtain a human waist and shoulder key point detection model. The driver's seated posture images are then input into the key point detection model to identify key points in the posture images, including waist and shoulder key points.

[0059] It should be noted that in this embodiment, an in-vehicle camera or other image acquisition device can be used to capture the driver's sitting posture image in real time. The in-vehicle camera can be set on the windshield inside the vehicle. This embodiment does not specifically limit the position of the camera and it can be any shooting position that can capture the driver's sitting posture. It will not be described in detail here.

[0060] Step 102: Determine the current sitting posture center of gravity using the waist key point and shoulder key point, compare the position of the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generate the center of gravity deviation result.

[0061] In this embodiment of the invention, the center of mass position of the current sitting posture is calculated based on the identified key points of the waist and shoulders. The center of mass of the current sitting posture is compared with the pre-stored initial center of mass of the sitting posture, and the positional deviation between the two is calculated to generate a center of mass deviation result.

[0062] Specifically, firstly, key points at the waist and shoulders are used to generate a shoulder-waist triangle. The centroid of the shoulder-waist triangle is calculated using a centroid calculation method, and this centroid is determined as the current sitting posture centroid. Then, a Cartesian coordinate system is constructed using the pre-determined initial sitting posture centroid as the origin. The initial sitting posture centroid is usually determined when the driver adjusts to a safe and standard sitting posture. The coordinates of the current sitting posture centroid are determined in the current Cartesian coordinate system, and the coordinate distance between the initial and current sitting posture centroids is calculated. The horizontal and vertical distances of the centroid coordinates are determined as the centroid deviation results.

[0063] In this embodiment, the identified key points of the waist and shoulders are used to generate a shoulder-waist triangle. The shoulder-waist triangle is formed based on the driver's actual sitting posture. The calculation of the centroid involves the average value of the coordinates of the three vertices of the triangle. The centroid coordinates are the average value of the coordinates of the three vertices of the triangle. The centroid of the calculated shoulder-waist triangle is determined as the centroid of the current sitting posture. In the constructed rectangular coordinate system, the coordinates of the current sitting posture centroid are determined, and the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid is calculated. The coordinate distance includes the horizontal distance (X-axis deviation) and the vertical distance (Y-axis deviation). Since the current sitting posture centroid represents the center position of the driver's current sitting posture, the centroid deviation result reflects the difference between the driver's current sitting posture and the safe sitting posture. When the sitting posture centroid moves or deviates, it can sensitively and quickly identify whether the sitting posture is abnormal, further improving the accuracy of sitting posture detection.

[0064] Step 103: Based on the centroid deviation results, check whether the driver's sitting posture is abnormal.

[0065] In this embodiment of the invention, the driver's sitting posture is detected as abnormal based on the horizontal distance (X-axis deviation) and vertical distance (Y-axis deviation) in the centroid deviation results. Specifically, if the horizontal distance between the current sitting posture centroid and the predetermined initial sitting posture centroid coordinates is less than or equal to a first threshold and the vertical distance is less than or equal to a second threshold, the driver's sitting posture is determined to be normal; otherwise, the driver's sitting posture is abnormal. Further details will not be elaborated here.

[0066] Step 104: In response to determining that the driver's sitting posture is abnormal, a warning is issued to the driver.

[0067] In this embodiment of the invention, in response to determining that the driver's posture is abnormal, a warning is issued to the driver. The warning may include a voice reminder, a text reminder, or a seatbelt action warning. Specifically, if an abnormal posture is determined, a warning issuance command is generated, controlling the vehicle system to issue a warning to the driver using at least one of the voice or text reminders. The number of warnings and the timing of the warnings are counted. If the number of warnings within a preset period exceeds a preset threshold, a control command is issued to the active pretensioner seatbelt to control the active pretensioner seatbelt to perform a tightening or vibration action, thus issuing a warning to the driver.

[0068] The vehicle driving warning method provided in this invention collects images of the driver's sitting posture, identifies key points in the images, determines the current sitting posture centroid using key points in the waist and shoulders, compares the current centroid with a pre-determined initial centroid to generate a centroid deviation result, and detects whether the driver's posture is abnormal based on the centroid deviation result. In response to determining that the driver's posture is abnormal, a warning is issued to the driver. This invention intelligently detects key points in the driver's waist and shoulders through sitting posture images, thereby obtaining the centroid position of the driver's shoulder-waist triangle. Using the sitting posture centroid for posture judgment can quickly and accurately identify whether the driver's sitting posture is abnormal, improving the accuracy of driving posture judgment. When an abnormal sitting posture is detected, i.e., dangerous driving behavior, automatic driving warning and alert actions are taken, reducing the probability of traffic accidents caused by dangerous driving behavior during vehicle operation and ensuring the effectiveness of safe driving.

[0069] Furthermore, refer to Figure 2 , showed Figure 1 A flowchart of step 101 in a vehicle driving warning method is provided. This method is basically the same as the vehicle driving warning method provided in the first embodiment of the present invention. Step 101 may include:

[0070] Step 201: Use a vehicle camera to capture images of a preset area, and input the captured sample images into a preset neural network model for training to obtain a key point detection model.

[0071] Step 202: Obtain an image of the current seating posture of the vehicle driver.

[0072] Step 203: Use a key point detection model to identify the sitting posture image and determine the waist key points and shoulder key points of the sitting posture image.

[0073] It should be noted that in this embodiment of the invention, a vehicle camera is used to acquire images of a preset area, typically the driver's seat area, to capture images of the driver's posture. The acquired sample images are input into a preset neural network model. The sample images include images of different drivers in different postures. To improve the accuracy of keypoint detection, the acquired sample images are pre-trained into the preset neural network model to obtain a keypoint detection model. That is, the preset neural network model is trained on the sample images to learn how to identify and locate key points in posture images, especially waist and shoulder key points. The trained neural network model can accurately identify key points in posture images, thus forming a keypoint detection model.

[0074] Specifically, the system acquires real-time images of the driver's current sitting posture for real-time monitoring. A trained key point detection model is used to identify the current sitting posture image. The key point detection model automatically locates and identifies the waist and shoulder key points in the current sitting posture image. Based on the identification results, the specific locations of the waist and shoulder key points in the current sitting posture image are determined.

[0075] This embodiment utilizes a pre-trained key point detection model to identify key points in the driver's sitting posture image in real time, especially the key points in the waist and shoulders. The accuracy of the key points in the waist and shoulders is crucial for determining whether the driver's sitting posture is abnormal. By automatically monitoring and identifying key points in sitting posture through the key point detection model, the accuracy and efficiency of sitting posture monitoring are improved.

[0076] Furthermore, refer to Figure 3 , showed Figure 1 A flowchart of step 102 in a vehicle driving warning method is provided. This method is basically the same as the vehicle driving warning method provided in the first embodiment of the present invention. Step 102 may include:

[0077] Step 301: Use the waist key points and shoulder key points to generate the shoulder-waist triangle.

[0078] Step 302: Calculate the centroid of the shoulder-waist triangle and determine the centroid of the current sitting posture.

[0079] Step 303: Using the predetermined initial seated center of mass as the origin, construct a rectangular coordinate system and determine the coordinates of the current seated center of mass.

[0080] Step 304: Calculate the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid, and determine the horizontal and vertical distances of the centroid coordinates as the centroid deviation result.

[0081] It should be noted that, in this embodiment of the invention, the identified key points of the waist and shoulders are used to generate a shoulder-waist triangle. A centroid calculation method is employed to calculate the centroid of the shoulder-waist triangle, and this calculated centroid is determined as the current sitting posture centroid, reflecting the center position of the driver's current sitting posture. A pre-determined initial sitting posture centroid is used as the origin of a rectangular coordinate system. Within this constructed rectangular coordinate system, the coordinates of the current sitting posture centroid are determined. The coordinates of the current sitting posture centroid are relative to the initial sitting posture centroid (the origin), facilitating the calculation of the coordinate distance between the initial and current sitting posture centroids. This distance includes both horizontal and vertical distances. The calculated horizontal and vertical distances are determined as the centroid deviation result, reflecting the difference between the driver's current sitting posture and a safe sitting posture.

[0082] Specifically, based on key points at the waist and two shoulders, three connecting lines are determined: shoulder-to-shoulder line, shoulder-to-waist line, and shoulder-to-waist line, forming the driver's shoulder-waist triangle. The centroid of the shoulder-waist triangle is then determined, and an xy coordinate system is constructed with the driver's initial normal sitting posture centroid position as the relative origin. When there is a certain vehicle speed (e.g., vehicle speed ≥ 5 km / h), the driver's sitting posture image is acquired in real time. After the key point detection model is trained in the early stage, the current sitting posture centroid position of the driver is quickly obtained. The current sitting posture centroid position is compared with the initial centroid position to determine the deviation of the driver's sitting posture and whether it is abnormal.

[0083] This invention accurately calculates the driver's current center of gravity and compares it with the initial center of gravity to determine whether the posture is abnormal. Based on the center of gravity deviation warning method, it can sensitively and finely detect changes and abnormalities in posture, effectively monitor and remind the driver to adjust their posture, and provide timely posture warnings to the driver, reducing fatigue and driving errors caused by poor posture, and improving driving safety. The calculation and application of the center of gravity deviation results further enhance the vehicle driving posture detection and warning effect.

[0084] Furthermore, refer to Figure 4 , showed Figure 1 A flowchart of step 103 in a vehicle driving warning method is provided. This method is basically the same as the vehicle driving warning method provided in the first embodiment of the present invention. Step 103 may include:

[0085] Step 401: Pre-calibrate a first threshold in the horizontal direction and a second threshold in the vertical direction based on the actual vehicle.

[0086] It should be noted that the first threshold in the horizontal direction and the second threshold in the vertical direction can be determined through prior neural network training and real vehicle calibration. The first threshold is the allowable error value of the sitting posture center of gravity in the horizontal direction, and the second threshold is the allowable error value of the sitting posture center of gravity in the vertical direction. The settings are based on the actual driver characteristics and the vehicle driver's seat, and will not be elaborated on here.

[0087] Step 402: If the horizontal distance between the current sitting posture center of mass and the predetermined initial sitting posture center of mass coordinates is less than or equal to the first threshold, and the vertical distance is less than or equal to the second threshold, then the driver's sitting posture is determined to be normal.

[0088] Step 403, otherwise, determine that the driver's sitting posture is abnormal.

[0089] In this embodiment of the invention, if the driver's posture is excessively offset horizontally, it may affect the driver's observation of the surrounding environment and increase driving risk. If the driver's posture is excessively offset vertically, it may cause the driver's line of sight to deviate from the road ahead, also increasing driving risk. If the horizontal distance between the current posture centroid and the predetermined initial posture centroid coordinates is less than or equal to a first threshold, and the vertical distance is less than or equal to a second threshold, then the driver's posture is determined to be normal. Otherwise, if the horizontal distance between the current posture centroid and the predetermined initial posture centroid coordinates is greater than the first threshold, then the driver's posture is determined to be abnormal, with obvious left-right offset, such as: the driver's upper body making obvious leaning movements towards the passenger seat or door, or turning their head backwards, etc.; if the vertical distance between the current posture centroid and the predetermined initial posture centroid coordinates is greater than the second threshold, then the driver's posture is determined to be abnormal, with obvious up-down offset, such as: the driver getting up, leaning back, or bending over.

[0090] This invention accurately determines whether a driver's posture is abnormal based on the horizontal and vertical distance between the current center of gravity and the initial center of gravity. This helps to detect poor posture in a timely manner and prompts the driver to adjust their posture through early warning prompts, thereby improving driving safety.

[0091] Specifically, the step of responding to determining that the driver's sitting posture is abnormal and providing a warning to the driver includes:

[0092] First, in response to the detection of abnormal driving posture of the vehicle driver, an early warning is generated and a command is issued.

[0093] Secondly, based on the warning instructions issued, the vehicle control system will use at least one of the following methods—voice prompts or text prompts—to provide warning reminders to the vehicle driver.

[0094] Secondly, record the number of warnings issued and the time of each warning.

[0095] It should be noted that in the above steps, in response to the detection of abnormal driving posture, a warning issuance command is generated and sent to the vehicle display screen or voice device. The vehicle system is then controlled to issue a warning to the driver using at least one of voice or text reminders. To further monitor abnormal driving posture, the number of warnings and the timing of the warnings are recorded, which facilitates the statistical analysis of driver posture warnings within the monitoring period and enables timely and effective driving warnings.

[0096] For example, when an abnormal driving posture is detected, a pop-up message can be displayed on the vehicle's large screen or LCD instrument panel saying "There is a dangerous driving action. Please drive safely." Alternatively, a voice prompt or instrument panel alarm can be used to alert the driver. Each alert is spaced at intervals (e.g., 1 minute). If the driver has returned to a normal driving posture before the next alert, the alert will be discontinued until an abnormal driving posture is detected again.

[0097] This embodiment provides timely warnings when abnormal driver posture is detected, and reminds the driver to adjust their posture via voice or text. It also records the number and time of warnings for subsequent analysis and evaluation, effectively monitoring and reminding the driver to adjust their posture, thereby improving driving safety and comfort.

[0098] Specifically, after recording the number of warning notifications and the time of each notification, the steps also include:

[0099] If the number of warning prompts within a preset period exceeds a preset prompt threshold, a control command is sent to the active pretensioning seat belt.

[0100] The active pretensioning seat belt is controlled to tighten or vibrate, providing a warning to the vehicle driver.

[0101] In this embodiment of the invention, the active pretensioning seat belt, also known as a motor-controlled pretensioning seat belt, relies on a servo motor to complete the pretensioning of the seat belt in a timely manner, and can perform tightening or vibration actions even in non-collision scenarios. Specifically, the number of warning prompts is monitored within a preset period, and the number of warning prompts is compared with a preset prompt threshold. The preset period can be a pre-set time period, such as every ten minutes, every hour, or every trip. The preset prompt threshold is a set number of prompts. The preset period and preset prompt threshold are set according to actual driving needs, and the specific values ​​are not limited here.

[0102] In this embodiment, if the number of warning prompts within a preset period exceeds a preset prompt threshold, a control command is sent to the active pretensioner seat belt. The control command is used to trigger the seat belt to issue a warning prompt, thereby controlling the active pretensioner seat belt to perform actions. Specifically, the active pretensioner seat belt is controlled to perform tightening or vibration actions. The tightening action can be achieved by quickly tightening the seat belt through the pretensioning device of the seat belt to remind the driver to pay attention to the sitting posture. The vibration action can be achieved by the vibration device integrated on the seat belt emitting vibration to physically remind the driver.

[0103] It should be noted that if the number of warning prompts within a preset period exceeds a preset threshold, a control command is sent to the active pretensioner seat belt to control it to tighten or vibrate, thus providing a warning to the driver. For example, if the driver is prompted to maintain an abnormal posture for more than a preset threshold number of times, such as more than three times within three minutes, and the driver is still in an abnormal posture, the motor-controlled pretensioner seat belt is controlled to tighten or vibrate to serve as a warning. The tightening force of the seat belt should not be too great, and the duration should not be too long; it is recommended to be less than or equal to 30 seconds.

[0104] This embodiment can provide a stronger warning through the active pretensioning seat belt when an abnormal driver posture is detected and too many warning prompts are issued. This warning method acts directly on the driver through physical feedback (such as tightening or vibration), which can more effectively attract the driver's attention, prompt them to adjust their posture, and improve driving safety.

[0105] Reference Figure 5 The diagram illustrates a flowchart of another vehicle driving warning method provided by an embodiment of the present invention. This method is basically the same as the vehicle driving warning method provided by the first embodiment of the present invention, except that the method may further include:

[0106] Step 101: Acquire images of the driver's sitting posture and identify key points in the images; the key points include waist key points and shoulder key points.

[0107] Step 102: Determine the current sitting posture center of gravity using the waist key point and shoulder key point, compare the position of the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generate the center of gravity deviation result.

[0108] Step 103: Based on the centroid deviation results, check whether the driver's sitting posture is abnormal.

[0109] Step 104: In response to determining that the driver's sitting posture is abnormal, a warning is issued to the driver.

[0110] Steps 101 to 104 described above are the same as those described previously and will not be repeated here.

[0111] Step 105: Monitor the changes in the driver's seating posture within a preset period.

[0112] Step 106: If the driver's sitting posture is detected to be normal within the preset period, then stop issuing warnings to the driver.

[0113] In this embodiment of the invention, the driver's posture changes are monitored in real time within a preset period. The preset period can be a fixed time period, such as every minute, every hour, or every trip. Similarly, the driver's posture can be determined by analyzing the deviation between the current posture center of gravity and the initial posture center of gravity. If the driver's posture is updated to normal within the preset period, that is, the deviation between the current posture center of gravity and the initial posture center of gravity is within a preset threshold range, it is determined that the driver's posture has been adjusted to a normal state. In response to the restoration of normal posture, the warning prompts to the driver are stopped, and the driver is no longer reminded by voice, text, or seat belt tightening.

[0114] Compared with the prior art, the embodiments of the present invention, while achieving the beneficial effects of the first embodiment, address the issue that frequent warning prompts may cause driver fatigue. Stopping the warning prompts helps maintain the driver's concentration. The system can promptly stop the warning prompts when the driver's posture returns to normal, avoiding unnecessary interference and driver fatigue. The dynamically adjusted warning mechanism helps improve driver comfort and driving experience, and also reflects the intelligence of vehicle driving warnings.

[0115] To facilitate a thorough understanding of the above-described vehicle driving warning method by those skilled in the art, refer to Figure 6 , Figure 6 This is a flowchart of a vehicle driving warning method provided in an embodiment of the present invention, illustrating the overall process of the vehicle driving warning provided in this embodiment of the present invention, specifically including:

[0116] S501, Determine the initial seating center of gravity of the vehicle driver during normal driving;

[0117] In this embodiment, the initial centroid of the driver's normal driving posture is determined by using an image of an ideal or safe sitting posture. Specifically, the identified key points of the waist and shoulders are used to generate a shoulder-waist triangle. The centroid of the shoulder-waist triangle is calculated using a centroid calculation method, and the calculated centroid of the shoulder-waist triangle is determined as the initial centroid of the driver's normal driving posture.

[0118] S502, the vehicle camera captures real-time images of the driver's sitting posture;

[0119] S503 detects key points on the shoulders, key points on the waist, and the center of gravity of the current sitting posture;

[0120] Specifically, the system will acquire real-time images of the driver's current sitting posture and input them into a trained key point detection model to identify the current sitting posture image. It will automatically locate and identify the waist and shoulder key points in the current sitting posture image. Based on the identification results, it will determine the specific locations of the waist and shoulder key points in the current sitting posture image and further determine the current sitting posture centroid.

[0121] S504, determine if the driver's seating posture is abnormal;

[0122] S505, if the horizontal distance between the current sitting posture center of gravity and the initial sitting posture center of gravity is less than or equal to the first threshold, and the vertical distance is less than or equal to the second threshold, then the driver's sitting posture is normal.

[0123] S506, driver's seating posture is abnormal;

[0124] It should be noted that if the horizontal distance between the current seat center of gravity and the predetermined initial seat center of gravity is greater than the first threshold, the driver's seat posture is judged to be abnormal, with obvious left and right deviation; if the vertical distance between the current seat center of gravity and the predetermined initial seat center of gravity is greater than the second threshold, the driver's seat posture is judged to be abnormal, with obvious up and down deviation.

[0125] S507, voice and text prompts;

[0126] S508, seatbelt warning reminder;

[0127] S509, Stop issuing warnings.

[0128] It should be noted that in the above steps, if an abnormal sitting posture of the vehicle driver is detected, the vehicle control system will use at least one of the voice or text reminders to warn the driver, and count the number of warning reminders and the time of the warning reminders. If the number of warning reminders within a preset period is greater than a preset reminder threshold, the active pretensioning seat belt will be controlled to tighten or vibrate to warn the driver, and the warning reminders will stop after the driver returns to a safe sitting posture.

[0129] This invention intelligently detects key points in the driver's waist and shoulders using seated posture images to obtain the centroid position of the driver's shoulder-waist triangle. By using the centroid of the seated posture for posture judgment, it can quickly and accurately identify whether the driver's seated posture is abnormal, improving the accuracy of driving posture judgment. When an abnormal seated posture is detected, i.e., when there is a dangerous driving action, the system automatically issues driving warnings and alerts, reducing the probability of traffic accidents caused by dangerous driving actions during vehicle operation and ensuring the effectiveness of safe driving.

[0130] Reference Figure 7 The diagram shows a structural schematic of a vehicle driving warning device according to an embodiment of the present invention. The device includes:

[0131] The image acquisition module 601 is used to acquire images of the driver's sitting posture and identify key points in the images; wherein the key points include waist key points and shoulder key points.

[0132] The center of gravity deviation module 602 is used to determine the current sitting posture center of gravity using the waist key points and shoulder key points, compare the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generate a center of gravity deviation result.

[0133] The posture detection module 603 is used to detect whether the driver's posture is abnormal based on the centroid deviation result.

[0134] The warning module 604 is used to provide a warning to the driver of the vehicle in response to determining that the driver's sitting posture is abnormal.

[0135] Furthermore, the image acquisition module 601 includes:

[0136] The training submodule is used to acquire images of a preset area using a vehicle camera, and input the acquired sample images into a preset neural network model for training to obtain a key point detection model.

[0137] The acquisition submodule is used to acquire an image of the vehicle driver's current seating posture.

[0138] The recognition submodule is used to recognize the sitting posture image using a key point detection model, and to determine the waist key points and shoulder key points of the sitting posture image.

[0139] Furthermore, the centroid deviation module 602 includes:

[0140] The first generation submodule is used to generate a shoulder-waist triangle using the waist key points and the shoulder key points;

[0141] The first calculation submodule is used to calculate the centroid of the shoulder-waist triangle and determine the centroid of the shoulder-waist triangle as the centroid of the current sitting posture.

[0142] The first determining submodule is used to construct a rectangular coordinate system by taking the pre-determined initial sitting posture centroid as the origin of the coordinate system, and to determine the coordinates of the current sitting posture centroid.

[0143] The second calculation submodule is used to calculate the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid, and to determine the horizontal and vertical distances of the centroid coordinates as the centroid deviation result.

[0144] Furthermore, the sitting posture detection module 603 includes:

[0145] The calibration submodule is used to pre-calibrate a first threshold in the horizontal direction and a second threshold in the vertical direction based on the actual vehicle.

[0146] The second determining submodule is used to determine that the driver's sitting posture is normal if the horizontal distance between the current sitting posture centroid and the pre-determined initial sitting posture centroid coordinates is less than or equal to a first threshold and the vertical distance is less than or equal to a second threshold.

[0147] The third determination submodule is used to determine if the driver's sitting posture is abnormal.

[0148] Furthermore, the early warning module 604 includes:

[0149] The second generation submodule is used to generate an early warning issuance command in response to the detection of abnormal sitting posture of the vehicle driver;

[0150] The first early warning submodule is used to control the vehicle system to provide early warning prompts to the driver by using at least one of voice or text reminders, based on the early warning instruction issued.

[0151] The recording submodule is used to record the number of warning prompts and the time of the warning prompts.

[0152] Furthermore, the early warning module 604 also includes:

[0153] The sending submodule is used to send control commands to the active pretensioning seat belt if the number of warning prompts within a preset period is greater than a preset prompt threshold.

[0154] The second warning submodule is used to control the active pretensioning seat belt to perform tightening or vibration actions to provide warning prompts to the vehicle driver.

[0155] Furthermore, the device also includes:

[0156] A posture monitoring module is used to monitor changes in the driver's posture within a preset period.

[0157] The stop warning module is used to stop issuing warning prompts to the driver of the vehicle if the driver's sitting posture is updated to normal within a preset period.

[0158] The vehicle driving warning device provided in this invention collects images of the driver's sitting posture, identifies key points in the images, and uses key points on the waist and shoulders to determine the current center of gravity. The current center of gravity is compared with a pre-determined initial center of gravity to generate a center of gravity deviation result. Based on this deviation result, the device detects whether the driver's posture is abnormal. In response to determining an abnormal posture, the device provides a warning to the driver. This invention intelligently detects key points on the driver's waist and shoulders in the sitting posture image to obtain the center of gravity of the driver's shoulder-waist triangle. Using the center of gravity for posture judgment allows for quick and accurate identification of abnormal driver posture, improving the accuracy of posture judgment. When an abnormal posture is detected, indicating dangerous driving behavior, the device automatically issues warnings and alerts, reducing the probability of traffic accidents caused by dangerous driving actions and ensuring the effectiveness of safe driving.

[0159] Reference Figure 8 The present invention also provides an electronic device, such as... Figure 8 As shown, it includes a processor 701, a communication interface 702, a memory 703, and a communication bus 704, wherein the processor 701, the communication interface 702, and the memory 703 communicate with each other through the communication bus 704.

[0160] Memory 703 is used to store computer programs;

[0161] When processor 701 executes a program stored in memory 703, it performs the following steps:

[0162] Acquire images of the driver's sitting posture and identify key points in the images; wherein, the key points include key points in the waist and key points in the shoulders;

[0163] The current sitting posture center of gravity is determined using the key points of the waist and shoulders. The current sitting posture center of gravity is compared with the predetermined initial sitting posture center of gravity to generate a center of gravity deviation result.

[0164] Based on the centroid deviation results, detect whether the driver's sitting posture is abnormal;

[0165] In response to determining that the driver of the vehicle is in an abnormal sitting posture, a warning is issued to the driver of the vehicle.

[0166] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0167] The communication interface is used for communication between the aforementioned terminal and other devices.

[0168] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0169] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0170] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform any of the vehicle driving warning methods described in the above embodiments.

[0171] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0172] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0173] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0174] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A vehicle driving warning method, characterized in that, The method includes: Acquire images of the driver's sitting posture and identify key points in the images; wherein, the key points include key points in the waist and key points in the shoulders; The current sitting posture center of gravity is determined using the key points of the waist and shoulders. The current sitting posture center of gravity is compared with the predetermined initial sitting posture center of gravity to generate a center of gravity deviation result. The initial sitting posture center of gravity is the initial sitting posture center of gravity of the vehicle driver under normal driving conditions, determined using an image of an ideal or safe sitting posture. Based on the centroid deviation results, detect whether the driver's sitting posture is abnormal; In response to determining that the driver's sitting posture is abnormal, a warning is issued to the driver. The process involves determining the current sitting posture center of gravity using the key points of the waist and shoulders, comparing the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generating a center of gravity deviation result, including: Using the aforementioned key points of the waist and the aforementioned key points of the shoulder, a shoulder-waist triangle is generated; Calculate the centroid of the shoulder-waist triangle and determine the centroid of the current sitting posture. A rectangular coordinate system is constructed using the predetermined initial seated centroid as the origin, and the coordinates of the current seated centroid are determined. Calculate the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid, and determine the horizontal and vertical distances of the centroid coordinates as the centroid deviation result; The step of detecting whether the driver's posture is abnormal based on the centroid deviation result includes: A first threshold in the horizontal direction and a second threshold in the vertical direction are pre-calibrated based on the actual vehicle. If the horizontal distance between the current sitting posture centroid and the predetermined initial sitting posture centroid coordinates is less than or equal to the first threshold, and the vertical distance is less than or equal to the second threshold, then the driver's sitting posture is determined to be normal. Otherwise, determine that the driver's sitting posture is abnormal; The step of generating a shoulder-waist triangle using the key points of the waist and the key points of the shoulder includes: Based on the aforementioned key points of the waist and two key points of the shoulders, three connecting lines are determined: the shoulder-to-shoulder connection line and the shoulder-to-waist connection line, thus generating a shoulder-waist triangle.

2. The method according to claim 1, characterized in that, The process involves acquiring images of the driver's seated posture and identifying key points within those images; these key points include lumbar key points and shoulder key points, including: The vehicle camera is used to capture images of a preset area. The captured sample images are then input into a preset neural network model for training to obtain a key point detection model. Obtain an image of the vehicle driver's current seating posture; A key point detection model is used to identify the sitting posture image and determine the waist key points and shoulder key points of the sitting posture image.

3. The method according to claim 1, characterized in that, The step of responding to determining that the driver's posture is abnormal and providing a warning to the driver includes: In response to detecting an abnormal sitting posture of the vehicle driver, a warning warning command is generated and issued. Based on the warning instruction, the vehicle control system shall use at least one of the following methods—voice reminder or text reminder—to provide a warning to the driver. Record the number of warnings issued and the time of each warning.

4. The method according to claim 3, characterized in that, After recording the number of warning notifications and the time of each warning notification, the method further includes: If the number of warning prompts within a preset period exceeds a preset prompt threshold, a control command is sent to the active pretensioning seat belt; The active pretensioning seat belt is controlled to tighten or vibrate, providing a warning to the vehicle driver.

5. The method according to claim 1, characterized in that, After determining that the driver's posture is abnormal and issuing a warning to the driver, the method further includes: Monitor the changes in the driver's seating posture within a preset period; If the driver's seating posture is detected to be normal within a preset period, the warning prompts to the driver will be stopped.

6. A vehicle driving warning device, characterized in that, The device includes: An image acquisition module is used to acquire images of the driver's sitting posture and identify key points in the images; wherein, the key points include waist key points and shoulder key points. The center of gravity deviation module is used to determine the current sitting posture center of gravity using the waist key points and shoulder key points, compare the current sitting posture center of gravity with the pre-determined initial sitting posture center of gravity, and generate a center of gravity deviation result; wherein, the initial sitting posture center of gravity is the initial sitting posture center of gravity of the vehicle driver under normal driving conditions determined using an ideal sitting posture or safe sitting posture image. The posture detection module is used to detect whether the driver's posture is abnormal based on the centroid deviation result. The early warning module is used to provide an early warning to the driver of the vehicle in response to determining that the driver's sitting posture is abnormal. The centroid deviation module includes: The first generation submodule is used to generate a shoulder-waist triangle using the waist key points and the shoulder key points; The first calculation submodule is used to calculate the centroid of the shoulder-waist triangle and determine the centroid of the shoulder-waist triangle as the centroid of the current sitting posture. The first determining submodule is used to construct a rectangular coordinate system by taking the pre-determined initial sitting posture centroid as the origin of the coordinate system, and to determine the coordinates of the current sitting posture centroid. The second calculation submodule is used to calculate the coordinate distance between the initial sitting posture centroid and the current sitting posture centroid, and to determine the horizontal and vertical distances of the centroid coordinates as the centroid deviation result. The sitting posture detection module includes: The calibration submodule is used to pre-calibrate a first threshold in the horizontal direction and a second threshold in the vertical direction based on the actual vehicle. The second determining submodule is used to determine that the driver's sitting posture is normal if the horizontal distance between the current sitting posture centroid and the pre-determined initial sitting posture centroid coordinates is less than or equal to a first threshold and the vertical distance is less than or equal to a second threshold. The third determination submodule is used to determine if the driver's sitting posture is abnormal. The step of generating a shoulder-waist triangle using the key points of the waist and the key points of the shoulder includes: Based on the aforementioned key points of the waist and two key points of the shoulders, three connecting lines are determined: the shoulder-to-shoulder connection line and the shoulder-to-waist connection line, thus generating a shoulder-waist triangle.

7. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to execute the instructions to implement the vehicle driving warning method as described in any one of claims 1 to 5.

8. A readable storage medium, characterized in that, A computer program is stored on the readable storage medium, which, when executed by a processor, implements the vehicle driving warning method as described in any one of claims 1 to 5.

Images