A method for detecting and processing traffic accidents based on an electric bicycle helmet and an electric bicycle
By installing a camera in front of the motorcycle helmet to detect traffic accident images and record videos in real time, and combining inclination sensors to determine the status of the motorcycle, the problems of manual detection lag and inefficiency are solved, and efficient traffic accident detection and management are achieved.
Patent Information
- Application Number
- CN202310582718.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-23
- Publication Date
- 2025-07-22
- Estimated Expiration
- 2043-05-23
AI Technical Summary
In the prior art, traffic accident detection mainly relies on manual operations, with obvious lag and low detection efficiency problems. Especially for shared electric motorcycles, there is a lack of effective autonomous detection methods.
By installing a camera in front of the motorcyclist helmet, ambient image information is obtained in real time, the traffic accident image is detected using image matching algorithm or deep learning, and entering the video recording mode when an accident is detected, image information and warning instructions are sent to the target object, and the bicycle's own status is judged by combining inclination sensors and vehicle posture detection, and timely processing is carried out.
Realize real-time detection and recording of the external environment of electric motorcycles and its own traffic accidents, improve detection efficiency and timeliness, and can deal with serious accidents in a timely manner and optimize operation and maintenance management.
Smart Images

Figure CN116486615B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of traffic accident detection, and in particular, to a method and an electric bicycle for detecting and processing traffic accidents based on an electric bicycle helmet. Background Art
[0002] With the increasing construction of the highway traffic network and the rising living standards, cars and shared electric bicycles have become important means of transportation for many families / users. With the increase in the number of cars and shared vehicles, traffic accidents often occur. Once a traffic accident occurs, there may be casualties, and it is necessary to detect and rescue in a timely manner.
[0003] In the prior art, the detection of traffic accidents still mainly relies on manual operation, that is, relevant staff view the monitoring video from the monitoring equipment on the traffic section to find the image information of the traffic accident scene, and then report the relevant video or picture information. Limited by factors such as the number of personnel and energy, the manual detection and analysis of traffic accidents has obvious lag. On the one hand, it may delay the best rescue time, and on the other hand, there is an obvious problem of low detection efficiency.
[0004] In addition, shared electric bicycles have the characteristics of large quantity, strong convenience, and high usage frequency. In the prior art, there is no detection method for detecting traffic accidents by using the characteristics of shared electric bicycles themselves.
[0005] In view of this, it is necessary to propose a method and an electric bicycle for detecting and processing traffic accidents based on an electric bicycle helmet to solve or at least alleviate the above defects. Summary of the Invention
[0006] The main object of the present invention is to provide a method and an electric bicycle for detecting and processing traffic accidents based on an electric bicycle helmet, so as to solve the problem that the detection of traffic accidents still mainly relies on manual operation, with obvious lag and low detection efficiency.
[0007] To achieve the above object, the present invention provides a method for detecting and processing traffic accidents based on an electric bicycle helmet, including the steps of:
[0008] S1, obtaining the real-time status information of the electric bicycle, and judging whether the electric bicycle is in an on state according to the real-time status information;
[0009] S2, when the electric bicycle is in an on state, controlling the camera mounted in front of the electric bicycle helmet to enter the capture mode, and obtaining the environmental image information captured by the camera at a first preset time interval;
[0010] S3, judging whether there is a preset traffic accident image in the environmental image information;
[0011] S4. When the preset traffic accident image exists in the environmental image information, control the camera to enter the first video recording mode. After waiting for a first preset duration, send the environmental image information recorded by the camera within the first preset duration and / or a first warning instruction to a first target object.
[0012] Preferably, the step S4 specifically includes the following steps:
[0013] S41. When the preset traffic accident image exists in the environmental image information, determine a traffic accident assessment level corresponding to the preset traffic accident image according to the preset traffic accident image, and judge whether the traffic accident assessment level is greater than a preset severe level; when the traffic accident assessment level is greater than the preset severe level, execute steps S42 - S44; when the traffic accident assessment level is less than or equal to the preset severe level, execute steps S45 - S47;
[0014] S42. Control the camera to enter the first video recording mode, obtain the current running speed of the electric bicycle, and obtain the distance between the electric bicycle and the traffic accident point;
[0015] S43. Determine a theoretically calculated time for the electric bicycle to move to the traffic accident point according to the distance and the current running speed, and use the theoretically calculated time as the first preset duration;
[0016] S44. After waiting for the first preset duration, send the environmental information recorded by the camera within the first preset duration, the traffic accident assessment level, and the current position of the electric bicycle to the first target object;
[0017] S45. Control the camera to enter the first video recording mode, record an environmental information video obtained by the camera within a second preset duration after entering the first video recording mode, and judge whether a preset traffic accident image exists in an environmental information node video corresponding to a first time node at the end of the second preset duration;
[0018] S46. When a preset traffic accident image exists in the environmental information node video corresponding to the first time node at the end of the second preset duration, extend the second preset duration by a third preset duration and use it as the first preset duration;
[0019] S47. After waiting for the first preset duration, send the environmental image information recorded by the camera within the first preset duration, the traffic accident assessment level, and the current position of the electric bicycle to the first target object.
[0020] Preferably, after the step S3, the following steps are further included:
[0021] S51. When there is no preset traffic accident image in the environmental image information, obtain the real-time running speed of the electric bicycle, and determine whether the electric bicycle is in a stationary state according to the real-time running speed;
[0022] S52. When the electric bicycle is in a stationary state, obtain the vehicle body attitude information detected by the vehicle body inclination detection device and the vertical distance between the camera and the ground; wherein, the vehicle body inclination detection device is arranged inside the vehicle body of the electric bicycle;
[0023] S53. Judge whether the electric bicycle is in a toppled state according to the vehicle body attitude information and judge whether the vertical distance is less than a preset threshold;
[0024] S54. When any one of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, control the camera to enter the second video recording mode, obtain the environmental information video obtained by the camera within a fourth preset duration after entering the second video recording mode, and at a second time node when the fourth preset duration ends, judge whether the electric bicycle is in a toppled state and judge whether the vertical distance between the camera and the ground is less than the preset threshold;
[0025] S55. When any one of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, send the environmental image information and / or a second warning instruction recorded by the camera within the fourth preset duration to the selected operation and maintenance personnel;
[0026] S56. When neither of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, maintain the camera in the capture mode.
[0027] Preferably, after the step S54, the following step is further included:
[0028] S541. When both of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold are established, determine that the electric bicycle is in a serious traffic accident level, and send the environmental image information and a third warning instruction recorded by the camera within the fourth preset duration to a second target object.
[0029] Preferably, after the step S3, the following step is further included:
[0030] S301. When there is no preset traffic accident image in the environmental image information, obtain the angular change rate of the helmet detected by the helmet inclination detection device and the speed change rate of the electric bicycle body, and determine whether the angular change rate is greater than a first preset threshold within a first set time, and determine whether the speed change rate of the electric bicycle is greater than a second preset threshold within a second set time;
[0031] S302. When any one of the conditions that the angular change rate is greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle is greater than the second preset threshold within the second set time is satisfied, determine that the electric bicycle is in a serious traffic accident level, and send the environmental image information recorded by the camera within the set time and a third warning instruction to a second target object;
[0032] S303. When neither of the two conditions that the angular change rate is greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle is greater than the second preset threshold within the second set time is satisfied, return to step S301.
[0033] Preferably, after step S3, the following steps are further included:
[0034] S31. When there is no preset traffic accident image in the environmental image information, determine whether there is a target obstacle in the environmental image information;
[0035] S32. When there is the target obstacle in the environmental image information, adjust the first preset time interval to a second preset time interval; wherein, the second preset time interval is less than the first preset time interval.
[0036] Preferably, after step S32, the following steps are further included:
[0037] S33. Starting from the third time node at the end of a fifth preset duration after the target obstacle is first detected, determine whether there is a target obstacle in the corresponding environmental image information;
[0038] S34. When there is no target obstacle in the environmental image information, adjust the second preset time interval to the first time interval;
[0039] S35. When there is a target obstacle in the environmental image information, maintain the camera to continue capturing images at the second preset time interval.
[0040] Preferably, after step S4, the following steps are further included:
[0041] S401. Obtain the historical position data of the electric bicycle within the sixth preset time period before the current moment, and determine the vector motion path of the electric bicycle based on the historical position data set;
[0042] S402. According to the current position of the electric bicycle at the current moment and the vector motion path, determine whether there is a target shared vehicle in the area behind the current electric bicycle and within a preset range from the current position; wherein, the target shared vehicle is communicatively connected to the electric bicycle;
[0043] S403. When there is a target shared vehicle in the area behind the current electric bicycle and within a preset range from the current position, send a fourth warning instruction to the target shared vehicle.
[0044] Preferably, the "selected operation and maintenance personnel" in step S55 is obtained through the following steps:
[0045] S551. Obtain the real-time position of the electric bicycle, and based on the real-time position of the electric bicycle, obtain the operation and maintenance task volume of multiple alternative operation and maintenance personnel within a preset range from the real-time position, and the distance between the alternative operation and maintenance personnel and the real-time position;
[0046] S552. Determine the alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distance is within a preset range and whose operation and maintenance task volume is less than the preset task volume as the selected operation and maintenance personnel;
[0047] S553. When the number of alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distance is within a preset range and whose operation and maintenance task volume is less than the preset task volume is multiple, randomly select one of the alternative operation and maintenance personnel as the selected operation and maintenance personnel.
[0048] The present invention also provides an electric bicycle, including a vehicle body, a helmet, and a control system disposed in the vehicle body; wherein, an inclination sensing device for detecting the vehicle body attitude information is further disposed in the vehicle body, the inclination sensor is connected to the control system, the helmet is provided with a communication module communicatively connected to the control system, a camera for acquiring an image of the riding environment is installed in front of the helmet, the control system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the method for detecting and processing traffic accidents based on an electric bicycle helmet as described above are implemented.
[0049] Compared with the prior art, the present invention has the following beneficial effects:
[0050] The present invention provides a method and an electric bicycle for detecting and processing traffic accidents based on an electric bicycle helmet. By obtaining real-time status information of the electric bicycle and determining whether the electric bicycle is in an on state according to the real-time status information, when the electric bicycle is in the on state, the camera mounted in front of the electric bicycle helmet is controlled to enter the capture mode, and environmental image information captured by the camera is obtained at a first preset time interval. It is determined whether there is a preset traffic accident image in the environmental image information. When there is the preset traffic accident image in the environmental image information, the camera is controlled to enter the first video recording mode. After waiting for a first preset duration, the environmental image information and / or a first warning instruction recorded by the camera within the first preset duration are sent to a first target object. In this way, the traffic accident image information captured by the camera can be instantaneously detected, recorded, and uploaded, with high timeliness and high detection efficiency.
[0051] In addition, when there is the preset traffic accident image in the environmental image information, the traffic accident evaluation level corresponding to the preset traffic accident image is determined according to the preset traffic accident image, and it is judged whether the traffic accident evaluation level is greater than a preset serious level, so as to perform emergency key processing on serious traffic accidents. At the same time, the situation of traffic accidents occurring to the electric bicycle itself is also considered, and relevant processing is carried out when a traffic accident occurs to the electric bicycle itself, so as to carry out rescue in time. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.
[0053] Figure 1 It is a flowchart in an embodiment of the present invention;
[0054] Figure 2 It is a flowchart showing the specific steps included in step S4 in an embodiment of the present invention;
[0055] Figure 3 It is a flowchart showing the steps further included after step S3 in an embodiment of the present invention.
[0056] The implementation, functional features, and advantages of the objectives of the present invention will be further described with reference to the embodiments and the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0058] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0059] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.
[0060] In addition, the descriptions involving "first", "second", etc. in the present invention are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.
[0061] Please refer to the Figures 1 to 3 accompanying drawings. A method for detecting and processing traffic accidents based on an electric bicycle helmet provided by an embodiment of the present invention includes the steps:
[0062] S1. Obtain the real-time status information of the electric bicycle, and determine whether the electric bicycle is in an on state according to the real-time status information; for example, it can be detected by whether the output current of the motor is greater than a preset threshold or by detecting the power-on status of the entire electric bicycle. Specifically, those skilled in the art can set it according to actual needs.
[0063] S2. When the electric bicycle is in an on state, control the camera mounted on the front of the electric bicycle helmet to enter the capture mode, and obtain the environmental image information captured by the camera at a first preset time interval; it can be understood that the camera can have a capture mode and a video recording mode. The camera in this application is set in front of the helmet and is used to capture the environmental image information during the riding process. Among them, the first preset time interval can be set as needed. For example, it can be set to 2s, that is, capture one piece of the environmental image information every 2s.
[0064] S3. Determine whether there is a preset traffic accident image in the environmental image information. It should be noted that there are various ways to determine whether there is a preset traffic accident image in the environmental image information. For example, through an image template matching algorithm, by setting a preset traffic accident image in advance, such as setting obvious image features such as detected bloodstains, vehicle debris, and vehicle fracture creases as the preset traffic accident image, and then using the image matching algorithm to match the matching degree between the current environmental image information and the template. When the matching degree is greater than the preset value, it can be considered that the current environmental image information has the preset traffic accident image. In another preferred example, deep learning can be used to detect whether there is a preset traffic accident image in the environmental image information. For example, by pre-collecting a large number of / sufficient image samples, dividing all image samples into a training set and a test set, and labeling each image sample with an artificial label (i.e., corresponding to whether it is a preset traffic accident image) to train it into a neural network model with high accuracy, and then using the trained neural network model to predict / detect it. This method of detecting through a neural network is a mature technology in the prior art and will not be elaborated here.
[0065] S4. When there is the preset traffic accident image in the environmental image information, control the camera to enter the first video recording mode. After waiting for the first preset duration, send the environmental image information recorded by the camera within the first preset duration and / or the first warning instruction to the first target object. Among them, the first preset duration can be set to 10s, and the first target object can be one or more of the user corresponding to the electric bicycle, the background management system, the traffic management department system, the medical system, and the shared electric bicycle operation and maintenance personnel. The first warning instruction can be to send the environmental image information, the real-time position and time information corresponding to the electric bicycle, etc. to the first target object.
[0066] The technical solution of this application obtains the real-time status information of the electric bicycle, and determines whether the electric bicycle is in an on state according to the real-time status information. When the electric bicycle is in an on state, control the camera installed in front of the electric bicycle helmet to enter the capture mode, and obtain the environmental image information captured by the camera at the first preset time interval. Determine whether there is a preset traffic accident image in the environmental image information. When there is the preset traffic accident image in the environmental image information, control the camera to enter the first video recording mode. After waiting for the first preset duration, send the environmental image information recorded by the camera within the first preset duration and / or the first warning instruction to the first target object. The present invention can detect whether a traffic accident occurs in the external environment of the electric bicycle and itself, with high timeliness and high detection efficiency.
[0067] As a preferred embodiment of the present invention, the step S4 specifically includes the following steps:
[0068] S41. When the preset traffic accident image exists in the environmental image information, determine the traffic accident assessment level corresponding to the preset traffic accident image according to the preset traffic accident image, and judge whether the traffic accident assessment level is greater than a preset severe level; when the traffic accident assessment level is greater than the preset severe level, execute steps S42 - S44; when the traffic accident assessment level is less than or equal to the preset severe level, execute steps S45 - S47. Specifically, the traffic accident assessment level corresponding to the preset traffic accident image can be determined by calculating physical evidence that can reflect a severe traffic accident (such as vehicle residues, vehicle broken creases, bloodstains, images of people falling to the ground, etc.), so as to determine the traffic accident assessment level corresponding to the traffic accident image. Or it can also be determined by using a neural network algorithm. For example, collect a sufficient number of historical sample pictures, label each image with an artificial label (the corresponding accident level), divide the historical sample pictures into a training set and a test set, and train to obtain a trained neural network model. The current traffic accident image can be detected through this neural network model.
[0069] S42. Control the camera to enter the first video recording mode, and obtain the current running speed of the electric bicycle and the distance between the electric bicycle and the traffic accident point. It should be noted that the traffic accident point can be selected as the most severe area in the traffic accident image as the traffic accident point.
[0070] S43. Determine the theoretically calculated time for the electric bicycle to move to the traffic accident point according to the distance and the current running speed, and use the theoretically calculated time as the first preset duration. The theoretically calculated time is the time required for the electric bicycle to move from the current position to the traffic accident point. In other words, the recording duration of the camera is the time from the current moment to the position where the traffic accident occurs at the traffic accident point. During this time, sufficient image information of the traffic accident scene can be collected.
[0071] S44. After waiting for the first preset duration, send the environmental information recorded by the camera within the first preset duration, the traffic accident assessment level, and the current position of the electric bicycle to the first target object. By sending the environmental information recorded by the camera within the first preset duration, the traffic accident assessment level, and the current position of the electric bicycle to the first target object, the traffic accident scene image can be sent to the target object in a timely manner for timely recording, tracking, and subsequent processing.
[0072] S45. Control the camera to enter the first video recording mode, record the environmental information video obtained by the camera within the second preset duration (30 s) after entering the first video recording mode, and determine whether the environmental information node video corresponding to the first time node at the end of the second preset duration has a preset traffic accident image. It should be noted that when the traffic accident assessment level is less than or equal to the preset severe level, it indicates that the camera may not have detected a traffic accident at this time or the traffic accident is relatively minor. In this case, such situations need to be further detected. That is, record the environmental information video obtained by the camera within the second preset duration after entering the first video recording mode, and determine whether the environmental information node video corresponding to the first time node at the end of the second preset duration has a preset traffic accident image. In other words, continuously record and detect within 30 s, intercept the environmental information node video again at the end of 30 s, and determine whether the environmental information node video has the preset traffic accident image, indicating that the traffic accident still exists within 30 s. To further determine the severity of the accident, perform subsequent steps for detection.
[0073] S46. When the environmental information node video corresponding to the first time node at the end of the second preset duration has a preset traffic accident image, extend the second preset duration by a third preset duration (1 min) as the first preset duration. When the environmental information node video corresponding to the first time node at the end of the second preset duration has a preset traffic accident image, it indicates that the traffic accident at this time is relatively serious, and the detection time is extended.
[0074] S47. After waiting for the first preset duration, send the environmental image information recorded by the camera within the first preset duration, the traffic accident assessment level, and the current position of the electric vehicle to the first target object. In other words, after the delay, the first preset duration becomes one minute and thirty seconds, thereby increasing the detection duration and recording more on-site information.
[0075] As a preferred embodiment of the present invention, after step S3, the following steps are further included:
[0076] S51. When there is no preset traffic accident image in the environmental image information, obtain the real-time running speed of the electric bicycle, and determine whether the electric bicycle is in a stationary state according to the real-time running speed. It should be noted that when there is no preset traffic accident image in the environmental image information, there are still situations where the electric bicycle itself may have a traffic accident or not. To accurately determine whether the electric bicycle itself has a traffic accident, in this embodiment, the real-time running speed is obtained, and whether the electric bicycle is in a stationary state is determined according to the real-time running speed. For example, when the real-time running speed is 0, it can be determined that the electric bicycle is in a stationary state.
[0077] S52. When the electric bicycle is in a stationary state, obtain the vehicle body attitude information detected by the vehicle body inclination detection device and obtain the vertical distance between the camera and the ground. The vehicle body inclination detection device is arranged inside the vehicle body of the electric bicycle. It can be understood that existing shared electric bicycles are usually equipped with inclination detection devices (such as gyroscopes, etc.), which can detect the vehicle body attitude information of the electric bicycle, for example, can determine whether the electric bicycle is in a toppled state. According to existing mature image processing algorithms, for example, the vertical distance between the camera and the ground can be determined by the Euclidean distance method. In addition, it should be noted that if there is no ground image in the environmental image information, it means that the distance between the camera and the ground is infinite.
[0078] S53. Determine whether the electric bicycle is in a toppled state according to the vehicle body attitude information and determine whether the vertical distance is less than a preset threshold.
[0079] S54. When any one of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, that is, the electric bicycle is already in a toppled state and the helmet has fallen to the ground, at this time, control the camera to enter the second video recording mode, obtain the environmental information video obtained by the camera within the fourth preset duration (for example, 2 minutes) after entering the second video recording mode, and at the second time node when the fourth preset duration ends, determine whether the electric bicycle is in a toppled state and determine whether the vertical distance between the camera and the ground is less than the preset threshold. That is, after 2 minutes, if the electric bicycle is still in a toppled state, or the helmet is still in a state of falling to the ground, at this time, it can be judged that the electric bicycle may be in a state of random parking or has had a traffic accident. At this time, subsequent steps need to be executed to further confirm these two states.
[0080] S55, when any one of the following conditions holds true: the motorcycle is in a tilted state and the vertical distance is less than a preset threshold, the environmental image information and / or the second warning instruction recorded by the camera within the fourth preset time period is sent to the selected operation and maintenance personnel; that is, when the motorcycle is parked in a disorderly manner for a long time or the helmet falls on the ground for a long time, no matter what the situation is, the environmental image information and / or the second warning instruction recorded by the camera within the fourth preset time period is sent to the selected operation and maintenance personnel, so that the selected operation and maintenance personnel can deal with the situation in time.
[0081] S56, when the motorcycle is in a tipping state and the vertical distance is less than a preset threshold, the camera is maintained in the capture mode. That is, the motorcycle is not in a tipping state and the helmet is not on the ground, indicating that the motorcycle has not had a traffic accident exceeding the preset level, and at this time, the camera only needs to be maintained in the capture mode.
[0082] In this embodiment, the posture of the motorcycle is detected and whether the helmet is in a grounded state is detected to comprehensively judge whether the motorcycle is involved in a traffic accident greater than a preset level. Different situations are handled separately. If the motorcycle is in a tilted state for a long period of time or the helmet is in a grounded state for a long period of time, instructions are immediately sent to the selected operation and maintenance personnel for timely processing, thereby improving the operation and maintenance management efficiency of the motorcycle.
[0083] As another preferred implementation, the step S54 further includes the following steps:
[0084] S541, when two of the conditions that the motorcycle is in a tipping state and the vertical distance is less than a preset threshold value are simultaneously met, it is determined that the motorcycle is in a serious traffic accident level, and the environmental image information recorded by the camera within the fourth preset time period and the third warning instruction are sent to the second target object.
[0085] In this embodiment, when the motorcycle is in a tilted state and the vertical distance is less than the preset threshold, that is, the motorcycle body is in a tilted state for a long time and the helmet is in a grounded state for a long time, in a preferred example, by detecting whether the helmet is in a worn state, and when the helmet is in a worn state and the helmet is in a grounded state, it can be determined that the motorcycle is in a serious traffic accident level. At this time, the environmental image information recorded by the camera within the fourth preset time and the third warning instruction can be sent to the second target object.
[0086] As another preferred embodiment, the step S3 further includes the following steps:
[0087] S301. When there is no preset traffic accident image in the environmental image information, obtain the angular change rate of the helmet detected by the helmet inclination detection device and the speed change rate of the electric bicycle body, and determine whether the angular change rate is greater than a first preset threshold within a first set time, and determine whether the speed change rate of the electric bicycle is greater than a second preset threshold within a second set time;
[0088] S302. When any one of the conditions that the angular change rate is greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle is greater than the second preset threshold within the second set time holds, determine that the electric bicycle is in the severe traffic accident level, and send the environmental image information recorded by the camera within the set time and a third warning instruction to a second target object;
[0089] S303. When neither of the two conditions that the angular change rate is greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle is greater than the second preset threshold within the second set time holds, return to step S301.
[0090] Similarly, when there is no preset traffic accident image in the environmental image information, the user himself may or may not have a traffic accident. This embodiment provides another detection method. Specifically, obtain the angular change rate of the helmet detected by the helmet inclination detection device and the speed change rate of the electric bicycle body, and determine whether the angular change rate is greater than a first preset threshold within a first set time, and determine whether the speed change rate of the electric bicycle is greater than a second preset threshold within a second set time; It can be understood that in the event of a severe traffic accident, that is, after the electric bicycle is violently impacted, the angle of the helmet of the electric bicycle and the speed of the electric bicycle change greatly in a short time. At this time, by judging the angular change rate and the speed change rate, it can be determined whether the electric bicycle has a severe traffic accident.
[0091] When any one of the conditions that the angular change rate is greater than the first preset threshold within a first set time (for example, 1 s) and the speed change rate of the electric bicycle is greater than the second preset threshold within a second set time (for example, 2 s) holds, determine that the electric bicycle is in the severe traffic accident level, and send the environmental image information recorded by the camera within the set time and a third warning instruction to a second target object; wherein, the second target object may be one or more of a background management system, a selected operation and maintenance personnel, the rider corresponding to the electric bicycle, and a medical system; the third warning instruction may be to send an alarm instruction, the severity level of the accident, and the current location and time of the electric bicycle, etc. to the second target object. Thus, rapid response can be achieved, and relevant processing can be carried out on the electric bicycle and the user involved in the traffic accident.
[0092] In addition, when neither of the two conditions that the rate of change of the angle is greater than the first preset threshold within the first set time and whether the rate of change of the speed of the electric motorcycle is greater than the second preset threshold within the second set time holds, it indicates that no traffic accident has occurred or the traffic accident is minor. Return to step S301 to continue the subsequent detection.
[0093] As a preferred embodiment, after step S3, the following steps are further included:
[0094] S31, when there is no preset traffic accident image in the environmental image information, determine whether there is a target obstacle in the environmental image information;
[0095] S32, when there is the target obstacle in the environmental image information, adjust the first preset time interval to a second preset time interval; wherein, the second preset time interval is less than the first preset time interval.
[0096] It should be noted that when there is no preset traffic accident image in the environmental image information, at this time, the user wearing a helmet may have a target obstacle between the vehicle and the traffic accident point, such as a large truck, a building, etc., or the user only pays attention to the traffic accident for a while. In order to be able to detect whether a traffic accident has occurred in this case as much as possible, in this embodiment, the environmental image information is further processed. By determining whether there is a target obstacle in the environmental image information, the target obstacle can be set in advance, such as a large truck, a building. The image recognition / detection algorithm is already very mature and will not be elaborated here. For example, through the image matching algorithm, it can be matched whether there is a target obstacle in the current environmental image information. Or, in other embodiments, a deep learning method can also be used, such as using the YOLO V5 model. By comparing the current environmental image information with the pre-trained YOLO V5 model, it can be detected whether the label corresponding to the current environmental image information contains the target obstacle.
[0097] In this embodiment, when there is the target obstacle in the environmental image information, the first preset time interval is adjusted to the second preset time interval. In other words, when there is a target obstacle, the camera will detect at a higher frequency to detect to the greatest extent whether the user captures a traffic accident image during riding. As a specific example, the first preset time interval can be set to 2s, and the second preset time interval can be set to 0.6s.
[0098] Further, after step S32, the following steps are further included:
[0099] S33. At a third time node at the end of a fifth preset duration after the target obstacle is first detected, determine whether the corresponding environmental image information has the target obstacle;
[0100] S34. When the environmental image information does not have the target obstacle, adjust the second preset time interval to the first time interval;
[0101] S35. When the environmental image information has the target obstacle, maintain the camera to continue capturing images at the second preset time interval.
[0102] In this embodiment, to avoid the situation where the user does not detect an obstacle and does not detect a traffic accident for a long time during cycling, and this scenario will cause the camera to be in a high-frequency detection mode. On the one hand, too many useless images are detected, and on the other hand, the invalid workload of the camera is increased, reducing the service life of the camera. In this implementation, at a third time node at the end of a fifth preset duration (for example, 30s) after the target obstacle is first detected, determine whether the corresponding environmental image information has the target obstacle. When the environmental image information does not have the target obstacle, adjust the second preset time interval to the first time interval, that is, after the camera does not capture the target obstacle for a long time, switch the camera back to the low-frequency detection mode. When the environmental image information has the target obstacle, maintain the camera to continue capturing images at the second preset time interval.
[0103] As a preferred implementation manner, after step S4, the following steps are further included:
[0104] S401. Obtain the historical position data of the electric bicycle within a sixth preset duration before the current moment and determine the vector motion path of the electric bicycle according to the historical position data set;
[0105] S402. According to the current position of the electric bicycle at the current moment and the vector motion path, determine whether there is a target shared vehicle in an area behind the current electric bicycle and within a preset range from the current position; wherein, the target shared vehicle is communicatively connected to the electric bicycle;
[0106] S403. When there is a target shared vehicle in an area behind the current electric bicycle and within a preset range from the current position, send a fourth warning instruction to the target shared vehicle.
[0107] It should be noted that in this embodiment, after it is determined that the preset traffic accident image exists in the environmental image information, in order to be able to give a reminder and warning to the following vehicles, in this embodiment, the historical position data of the electric bicycle within the sixth preset time period is obtained, and the vector motion path of the electric bicycle is determined according to the historical position data set. For example, the historical position data set within 3 s before the current moment is obtained, so that the driving direction of the electric bicycle at the current moment can be defined. Furthermore, the vehicles within the area behind the current electric bicycle and within a preset range from the current position are determined through the driving direction. For example, the electric bicycles within 50 m behind the current electric bicycle are given the fourth warning instruction. Among them, the target shared vehicle and the electric bicycle can be communicatively connected. For example, vehicles of the same model, or vehicles that can establish a communication connection between different models can all be used as the target shared vehicle. Among them, the fourth warning instruction can be to send reminder deceleration information, detour driving information, etc. to the target shared vehicle. In this way, it is possible to give an early warning reminder to the following vehicles and improve the safety of traffic travel.
[0108] Further, the "selected operation and maintenance personnel" in step S55 is obtained through the following steps:
[0109] S551, obtain the real-time position of the electric bicycle, and obtain the operation and maintenance task amounts of multiple alternative operation and maintenance personnel within a preset range from the real-time position according to the real-time position of the electric bicycle, and the distances between the alternative operation and maintenance personnel and the real-time position;
[0110] S552, determine the alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distances are within a preset range and whose operation and maintenance task amounts are less than the preset task amount as the selected operation and maintenance personnel;
[0111] S553, when the number of alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distances are within a preset range and whose operation and maintenance task amounts are less than the preset task amount is multiple, randomly select one of the alternative operation and maintenance personnel as the selected operation and maintenance personnel.
[0112] It should be noted that in this embodiment, the spatial distance between the electric bicycle and the operation and maintenance personnel is considered, and at the same time, the operation and maintenance task amount of each operation and maintenance personnel is considered. The alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distances are within a preset range and whose operation and maintenance task amounts are less than the preset task amount are used as the selected operation and maintenance personnel. When the number of alternative operation and maintenance personnel among the multiple alternative operation and maintenance personnel whose distances are within a preset range and whose operation and maintenance task amounts are less than the preset task amount is multiple, randomly select one of the alternative operation and maintenance personnel as the selected operation and maintenance personnel. In this way, it is possible to ensure that a better alternative operation and maintenance personnel is selected as the selected operation and maintenance personnel, thereby improving the management efficiency of the electric bicycle and shortening the disposal time as much as possible.
[0113] The present invention also provides an electric bicycle, which includes a vehicle body, a helmet, and a control system disposed in the vehicle body; wherein, an inclination sensing device for detecting the attitude information of the vehicle body is further disposed in the vehicle body, the inclination sensor is connected to the control system, the helmet is provided with a communication module communicatively connected to the control system, a camera for acquiring an image of the riding environment is installed in front of the helmet, the control system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the method for detecting and processing traffic accidents based on an electric bicycle helmet as described above are implemented.
[0114] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.
Claims
1. A method for detecting and processing traffic accidents based on an electric bicycle helmet, characterized in that, Including the steps: S1. Obtain the real-time status information of the electric bicycle, and determine whether the electric bicycle is in an on state according to the real-time status information; S2. When the electric bicycle is in an on state, control the camera mounted in front of the electric bicycle helmet to enter the capture mode, and obtain the environmental image information captured by the camera at a first preset time interval; S3. Determine whether there is a preset traffic accident image in the environmental image information; S4. When there is the preset traffic accident image in the environmental image information, control the camera to enter the first video recording mode, wait for a first preset duration, and send the environmental image information and / or the first warning instruction recorded by the camera within the first preset duration to a first target object; The step S4 specifically includes the following steps: S41. When there is the preset traffic accident image in the environmental image information, determine the traffic accident assessment level corresponding to the preset traffic accident image according to the preset traffic accident image, and judge whether the traffic accident assessment level is greater than a preset severe level; when the traffic accident assessment level is greater than the preset severe level, execute steps S42 to S44; when the traffic accident assessment level is less than or equal to the preset severe level, execute steps S45 to S47; S42. Control the camera to enter the first video recording mode, and obtain the current running speed of the electric bicycle and the distance between the electric bicycle and the traffic accident point; S43. Determine the theoretically calculated time for the electric bicycle to move to the traffic accident point according to the distance and the current running speed, and use the theoretically calculated time as the first preset duration; S44. After waiting for the first preset duration, send the environmental information, the traffic accident assessment level, and the current position of the electric bicycle recorded by the camera within the first preset duration to the first target object; S45. Control the camera to enter the first video recording mode, record the environmental information video obtained by the camera within a second preset duration after entering the first video recording mode, and judge whether there is a preset traffic accident image in the environmental information node video corresponding to the first time node at the end of the second preset duration; S46. When there is a preset traffic accident image in the environmental information node video corresponding to the first time node at the end of the second preset duration, extend the second preset duration by a third preset duration and use it as the first preset duration; S47. After waiting for the first preset duration, send the environmental image information, the traffic accident assessment level, and the current position of the electric bicycle recorded by the camera within the first preset duration to the first target object.
2. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 1, wherein After the step S3, the following steps are further included: S51. When there is no preset traffic accident image in the environmental image information, obtain the real-time running speed of the electric bicycle, and judge whether the electric bicycle is in a stationary state according to the real-time running speed; S52. When the electric bicycle is in a stationary state, obtain the vehicle body attitude information detected by the vehicle body inclination detection device and obtain the vertical distance between the camera and the ground; wherein, the vehicle body inclination detection device is arranged inside the vehicle body of the electric bicycle. S53. According to the vehicle body attitude information, judge whether the electric bicycle is in a toppled state and judge whether the vertical distance is less than a preset threshold. S54. When any one of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, control the camera to enter the second video recording mode, obtain the environmental information video obtained by the camera within a fourth preset time period after entering the second video recording mode, and at the second time node when the fourth preset time period ends, judge whether the electric bicycle is in a toppled state and judge whether the vertical distance between the camera and the ground is less than the preset threshold. S55. When any one of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, send the environmental image information and / or the second warning instruction recorded by the camera within the fourth preset time period to the selected operation and maintenance personnel. S56. When neither of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold is established, maintain the camera in the capture mode.
3. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 2, wherein, After the step S54, the following step is further included: S541. When both of the electric bicycle being in a toppled state and the vertical distance being less than the preset threshold are established, determine that the electric bicycle is in a serious traffic accident level, and send the environmental image information and the third warning instruction recorded by the camera within the fourth preset time period to the second target object.
4. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 1, wherein, After the step S3, the following step is further included: S301. When the preset traffic accident image is not present in the environmental image information, obtain the angular change rate of the helmet detected by the helmet inclination detection device and obtain the speed change rate of the electric bicycle body, and judge whether the angular change rate is greater than a first preset threshold within a first set time and judge whether the speed change rate of the electric bicycle is greater than a second preset threshold within a second set time. S302. When any one of the angular change rate being greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle being greater than the second preset threshold within the second set time is established, determine that the electric bicycle is in a serious traffic accident level, and send the environmental image information and the third warning instruction recorded by the camera within the set time to the second target object. S303. When neither of the angular change rate being greater than the first preset threshold within the first set time and the speed change rate of the electric bicycle being greater than the second preset threshold within the second set time is established, return to step S301.
5. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 1, characterized in that, After the step S3, the following step is further included: S31. When the preset traffic accident image is not present in the environmental image information, judge whether there is a target obstacle in the environmental image information. S32. When the target obstacle exists in the environmental image information, adjust the first preset time interval to a second preset time interval, where the second preset time interval is less than the first preset time interval.
6. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 5, wherein, After the step S32, the following steps are further included: S33. At a third time node when a fifth preset duration after the target obstacle is first detected ends, determine whether the corresponding environmental image information has the target obstacle. S34. When the environmental image information does not have the target obstacle, adjust the second preset time interval to the first preset time interval. S35. When the environmental image information has the target obstacle, maintain the camera to continue capturing images at the second preset time interval.
7. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 1, wherein, After the step S4, the following steps are further included: S401. Obtain the historical position data of the electric bicycle within a sixth preset duration before the current moment and determine the vector motion path of the electric bicycle according to the historical position data set. S402. According to the current position where the electric bicycle is located at the current moment and the vector motion path, determine whether there is a target shared vehicle in the area behind the current electric bicycle and within a preset range from the current position; where the target shared vehicle is communicatively connected to the electric bicycle. S403. When there is a target shared vehicle in the area behind the current electric bicycle and within a preset range from the current position, send a fourth warning instruction to the target shared vehicle.
8. The method for detecting and processing traffic accidents based on an electric bicycle helmet according to claim 2, wherein The "selected operation and maintenance personnel" in the step S55 is obtained through the following steps: S551. Obtain the real-time position of the electric bicycle, and according to the real-time position of the electric bicycle, obtain the operation and maintenance task amounts of multiple alternative operation and maintenance personnel within a preset range from the real-time position, and the distances between the alternative operation and maintenance personnel and the real-time position. S552. Determine the alternative operation and maintenance personnel within the preset range and with an operation and maintenance task amount less than the preset task amount among the multiple alternative operation and maintenance personnel as the selected operation and maintenance personnel. S553. When the number of alternative operation and maintenance personnel within the preset range and with an operation and maintenance task amount less than the preset task amount among the multiple alternative operation and maintenance personnel is multiple, randomly select one of the alternative operation and maintenance personnel as the selected operation and maintenance personnel.
9. An electric bicycle, characterized in that, It includes a vehicle body, a helmet, and a control system arranged in the vehicle body; where an inclination sensing device for detecting the vehicle body attitude information is further arranged in the vehicle body, the inclination sensing device is connected to the control system, the helmet is provided with a communication module communicatively connected to the control system, a camera for obtaining a riding environment image is installed in front of the helmet, the control system includes a memory, a processor, and a computer program stored in the memory and operable on the processor, and when the processor executes the computer program, it implements the steps of the method for detecting and processing traffic accidents based on an electric bicycle helmet according to any one of claims 1 to 8.
Citation Information
Patent Citations
An outomatic sensing system for traffic accident and method thereof
KR1020100119476A
Smart Mobility accident risk detection system
KR102141299B1