Abnormal vehicle detection method and device, electronic equipment and readable storage medium
By analyzing the vehicle captured image data collected by the camera equipment, extracting feature information and calculating abnormal parameters, identifying vehicles that deliberately forged accidents, solving the problem of difficult to identify abnormal vehicles in the prior art, and achieving automatic identification and prevention of intentional accidents .
Patent Information
- Application Number
- CN202310591753.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-22
- Publication Date
- 2024-11-22
AI Technical Summary
The prior art is difficult to identify and solve the problem of deliberately forging abnormal vehicles in collisions between two vehicles, resulting in damage to the interests of car owners and congestion on highways.
By obtaining the vehicle capture image data collected by the camera device, the target feature information of the vehicle, including vehicle appearance, collision, speed and relative position information, the abnormal parameters are calculated, and the potential abnormal vehicles are screened out based on preset thresholds.
It realizes automatic identification of abnormal vehicles that intentionally forged accidents, protects the interests of car owners, and avoids congestion and other hazards on highway roads.
Smart Images

Figure CN119007129A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of monitoring technology, and in particular to an abnormal vehicle detection method, device, electronic equipment and readable storage medium. Background Art
[0002] On highways, some owners of vehicles, especially high-end vehicles, will deliberately fake traffic accidents in which two vehicles collide in order to extort money from the owners of the vehicles involved. This will not only harm the interests of the owners of the vehicles involved, but will also cause congestion on the highway and other hazards.
[0003] Therefore, how to identify abnormal vehicles that intentionally fake a collision between two vehicles is a technical problem that needs to be urgently solved by technicians in related fields. Summary of the invention
[0004] The present invention provides an abnormal vehicle detection method, device, electronic device and readable storage medium, which are used to solve the technical problem of how to identify an abnormal vehicle that intentionally forges a collision accident between two vehicles in the prior art.
[0005] The present invention provides an abnormal vehicle detection method, comprising:
[0006] Acquire captured image data of at least one vehicle to be detected collected by a camera device, wherein the captured image data includes a first captured image of the vehicle to be detected during at least one travel process;
[0007] For each of the vehicles to be detected, based on the first captured image of the vehicle to be detected during each trip, target feature information of the vehicle to be detected is obtained, wherein the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip;
[0008] Based on the target feature information, determining a first abnormal parameter corresponding to the vehicle to be detected during each trip, wherein the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forged a traffic accident;
[0009] Based on the first abnormal parameter corresponding to each of the to-be-detected vehicles during each trip and a preset abnormal screening threshold, a target abnormal vehicle is determined from the at least one to-be-detected vehicle.
[0010] According to an abnormal vehicle detection method provided by the present invention, the target feature information includes vehicle appearance information, vehicle body collision information, driving speed information, and relative position information with other vehicles;
[0011] The determining, based on the target feature information, a first abnormal parameter corresponding to each trip of the vehicle to be detected includes:
[0012] Determining a weighting coefficient of the vehicle type of the vehicle to be detected based on the license plate information and vehicle type information of the vehicle to be detected in the vehicle appearance information;
[0013] Determining a collision weighting coefficient of the vehicle to be detected based on the collision position of the vehicle to be detected in the vehicle body collision information;
[0014] Determining a speed weighting coefficient of the vehicle to be detected based on a first real-time speed and a first average speed of the vehicle to be detected in the driving speed information;
[0015] Determining a following weighting coefficient of the vehicle to be detected based on the relative position between the vehicle to be detected and the other vehicles in the relative position information;
[0016] Based on the vehicle type weighting coefficient, the collision weighting coefficient, the speed weighting coefficient and the following weighting coefficient, a first abnormal parameter corresponding to the vehicle to be detected is determined.
[0017] According to an abnormal vehicle detection method provided by the present invention, the speed weighting coefficient of the vehicle to be detected is determined based on the first real-time speed and the first average speed of the vehicle to be detected in the driving speed information, including:
[0018] Determine a first real-time speed average of the vehicle to be detected based on the first real-time speed of the vehicle to be detected when it passes through at least two camera devices;
[0019] Determining a first quantized mean value of the vehicle to be detected based on a preset quantized value and a first real-time mean value of the speed of the vehicle to be detected;
[0020] When a first average speed of the vehicle to be detected when passing through the at least two camera devices is not greater than the first quantized mean value, the speed weighting coefficient is determined based on a first speed coefficient.
[0021] According to an abnormal vehicle detection method provided by the present invention, determining the speed weighting coefficient based on the first speed coefficient includes:
[0022] In the case where it is determined that the vehicle to be detected has a following driving behavior, obtaining a second real-time speed and a second average speed of the followed vehicle when it passes through the at least two camera devices;
[0023] Determining a second real-time speed average of the followed vehicle based on the second real-time speed of the followed vehicle when the followed vehicle passes through the at least two camera devices;
[0024] Determining a second quantized mean value of the followed vehicle based on a preset quantized value and a second real-time mean value of the speed of the followed vehicle;
[0025] The speed weighting coefficient is determined based on the first speed coefficient and the second speed coefficient when a second average speed of the followed vehicle when passing the at least two imaging devices is not greater than the second quantized mean value.
[0026] According to an abnormal vehicle detection method provided by the present invention, the preset abnormal screening threshold includes a first abnormal threshold and a first quantity threshold;
[0027] The step of determining a target abnormal vehicle from the at least one vehicle to be detected based on the first abnormal parameter corresponding to each of the vehicles to be detected during each trip and a preset abnormal screening threshold comprises:
[0028] For each of the vehicles to be detected, determining a first parameter set corresponding to the vehicle to be detected based on the first abnormal parameter corresponding to the vehicle to be detected during each trip;
[0029] Determine at least one candidate abnormal vehicle based on the first parameter set of each of the vehicles to be detected, the first abnormal threshold, and the first quantity threshold;
[0030] For each candidate abnormal vehicle, obtaining a first driving time of the candidate abnormal vehicle and a second captured image of the passengers when the candidate abnormal vehicle travels within each preset time period;
[0031] Based on the first travel time of each candidate abnormal vehicle when traveling within each preset time period and the second captured image of the passenger, a target abnormal vehicle is determined from the at least one candidate abnormal vehicle.
[0032] According to an abnormal vehicle detection method provided by the present invention, the preset abnormal screening threshold also includes a second abnormal threshold and a second quantity threshold;
[0033] The step of determining a target abnormal vehicle from at least one candidate abnormal vehicle based on a first travel time of each candidate abnormal vehicle and a second captured image of a passenger in each preset time period comprises:
[0034] Determine a first travel coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the first travel time, a first weight coefficient corresponding to each preset time period, a first start time, and a first end time;
[0035] Based on the facial image of the passenger in the second captured image, determine the passenger information corresponding to the candidate abnormal vehicle, and based on the passenger information, determine the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period;
[0036] Determine a second abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the first abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each preset time period, the first travel coefficient, and the passenger coefficient;
[0037] Determine a second parameter set corresponding to the candidate abnormal vehicle based on the second abnormal parameters corresponding to the candidate abnormal vehicle when traveling within each preset time period;
[0038] A target abnormal vehicle is determined from the at least one abnormal vehicle candidate based on the second parameter set of each abnormal vehicle candidate, the second abnormal threshold, and the second quantity threshold.
[0039] According to an abnormal vehicle detection method provided by the present invention, the method comprises:
[0040] Determine the license plate number and model of the current vehicle based on the third captured image of the current vehicle, and obtain the current transaction number and current service life of the current vehicle based on the license plate number of the current vehicle;
[0041] When the current number of transactions is greater than a preset number threshold, the current service life is greater than a preset life threshold, and the model of the current vehicle is the target model, the current vehicle is determined as the vehicle to be detected.
[0042] The present invention also provides an abnormal vehicle detection device, comprising:
[0043] An image acquisition module, used to acquire captured image data of at least one vehicle to be detected captured by a camera device, wherein the captured image data includes a first captured image of the vehicle to be detected during at least one trip;
[0044] An information extraction module is used to obtain target feature information of each vehicle to be detected based on a first captured image of the vehicle to be detected during each trip, wherein the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip;
[0045] A parameter determination module, used to determine, based on the target feature information, a first abnormal parameter corresponding to the vehicle to be detected during each trip, wherein the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forged a traffic accident;
[0046] The abnormal screening module is used to determine a target abnormal vehicle from the at least one vehicle to be detected based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold.
[0047] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the abnormal vehicle detection method as described above is implemented.
[0048] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the abnormal vehicle detection method as described above is implemented.
[0049] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the abnormal vehicle detection method as described above is implemented.
[0050] The abnormal vehicle detection method, device, electronic device and readable storage medium provided by the present invention obtain target feature information of the vehicle to be detected based on the first captured image of the vehicle to be detected during each trip, and the target feature information is used to characterize the abnormal information of the vehicle to be detected during the trip; based on the target feature information, the first abnormal parameter corresponding to the vehicle to be detected during each trip is determined, so as to measure the probability of each vehicle to be detected being an abnormal vehicle that deliberately forges a collision between two vehicles based on the first abnormal parameter, so as to screen out the target abnormal vehicle from at least one vehicle to be detected based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold, so as to achieve the purpose of automatically identifying the abnormal vehicle that deliberately forges a collision between two vehicles, and then take corresponding measures to protect the interests of the vehicle owners involved, and avoid causing congestion on highways and other hazards, thereby solving the technical problem of how to identify the abnormal vehicle that deliberately forges a collision between two vehicles in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0052] Figure 1 This is one of the flow charts of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0053] Figure 2 This is a second flow chart of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0054] Figure 3 is a schematic diagram of the relative positions between the vehicle to be detected and other vehicles in an embodiment of the present invention;
[0055] Figure 4 This is a third flow chart of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0056] Figure 5 This is a fourth flow chart of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0057] Figure 6 This is a fifth flow chart of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0058] Figure 7 This is a sixth flow chart of the abnormal vehicle detection method provided by an embodiment of the present invention;
[0059] Figure 8 This is the seventh flow chart of the abnormal vehicle detection method provided by the embodiment of the present invention;
[0060] Fig. 9 is a schematic structural diagram of an abnormal vehicle detection device provided by an embodiment of the present invention;
[0061] Fig.10 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0062] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0063] Combine the following Figure 1-Figure 8 The abnormal vehicle detection method provided by the present invention is described. Figure 1 As shown, the present invention provides an abnormal vehicle detection method, comprising:
[0064] Step 101 : acquiring captured image data of at least one vehicle to be detected collected by a camera device, wherein the captured image data includes a first captured image of the vehicle to be detected during at least one travel process.
[0065] The captured image data includes a first captured image of the vehicle to be detected collected by the camera at at least one shooting angle, and the number of the first captured image is at least one frame. The travel location corresponding to at least one travel process of the vehicle to be detected includes a highway.
[0066] Step 102 , for each vehicle to be detected, based on the first captured image of the vehicle to be detected during each trip, obtain target feature information of the vehicle to be detected, where the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip.
[0067] Furthermore, the target feature information includes at least one of vehicle appearance information, vehicle body collision information, driving speed information, and relative position information with other vehicles of the vehicle to be detected.
[0068] Furthermore, feature extraction is performed on the first captured image of the vehicle to be detected to obtain vehicle appearance information, vehicle body collision information, and relative position information between the vehicle to be detected and other vehicles.
[0069] Further, based on the displacement information of the vehicle to be detected in at least two frames of the first captured images and the capture time of each frame of the first captured images, the driving speed information of the vehicle to be detected is calculated. Alternatively, based on the speed sensor in the camera device that captures the first captured images, the driving speed information of the vehicle to be detected is obtained.
[0070] Step 103: Based on the target feature information, determine the first abnormal parameter corresponding to the vehicle to be detected during each trip, where the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that fakes a traffic accident.
[0071] Step 104 , based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold, a target abnormal vehicle is determined from at least one vehicle to be detected.
[0072] Furthermore, the higher the parameter value of the first abnormal parameter of the vehicle to be detected is, the greater the probability that the vehicle to be detected is a target abnormal vehicle that intentionally forges a collision accident between two vehicles.
[0073] In the above steps 101 to 104, target feature information of the vehicle to be detected is obtained based on the first captured image of the vehicle to be detected during each trip, and the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip; based on the target feature information, the first abnormal parameter corresponding to the vehicle to be detected during each trip is determined, so as to measure the probability that each vehicle to be detected is an abnormal vehicle that deliberately forges a collision between two vehicles based on the first abnormal parameter, so as to screen out the target abnormal vehicle from at least one vehicle to be detected based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold, so as to achieve the purpose of automatically identifying abnormal vehicles that deliberately forge a collision between two vehicles, and then take corresponding measures to protect the interests of the vehicle owners involved and avoid causing congestion on highways and other hazards, thereby solving the technical problem of how to identify abnormal vehicles that deliberately forge a collision between two vehicles in the prior art.
[0074] In one embodiment, the target feature information includes vehicle appearance information, vehicle body collision information, driving speed information, and relative position information with other vehicles; Figure 2 As shown, in the above step 103, the step of determining the first abnormal parameter corresponding to the vehicle to be detected during each trip based on the target feature information specifically includes:
[0075] Step 201, determining a vehicle model weighting coefficient of the vehicle to be detected based on the license plate information and vehicle model information of the vehicle to be detected in the vehicle appearance information.
[0076] Furthermore, the above step 201 includes the following steps: based on the license plate information of the vehicle to be detected, determining the current service life of the vehicle to be detected, and based on the model information of the vehicle to be detected, determining the new car price of the corresponding model of the vehicle to be detected; wherein the vehicle to be detected can be represented as L car Based on the current age of the vehicle to be tested, the new car price of the corresponding model of the vehicle to be tested, the preset price range and the preset age range, the model weighting coefficient of the vehicle to be tested is determined, where the model weighting coefficient k car As shown in formula (1):
[0077]
[0078] Among them, the current service life can be expressed as T y , the price of a new car can be expressed as R now , the preset price range can be expressed as [R min , R max ], the preset age range can be expressed as [T min , T max ], and T min >X, X means Figure 8 In the illustrated embodiment, a threshold of years is preset.
[0079] It should be noted that if the current age is within the preset age range and the price of a new car is within the preset price range, the weighted coefficient of the vehicle model to be detected is set to 1, otherwise the weighted coefficient of the vehicle model to be detected is set to less than 1. This is because when the current age is within the preset age range and the price of a new car is within the preset price range, the probability that the vehicle to be detected is a target abnormal vehicle that deliberately forges a collision between two vehicles is relatively high.
[0080] Step 202: determining a collision weighting coefficient of the vehicle to be detected based on the collision position of the vehicle to be detected in the vehicle body collision information.
[0081] Optionally, when it is determined based on the vehicle body collision information that the vehicle to be detected has a collision mark, and the collision position corresponding to the collision mark is the left side of the vehicle body to be detected, the collision weighting coefficient of the vehicle to be detected is determined to be the first collision coefficient k mark , and the first collision coefficient k mark Greater than 1. The collision marks include at least one of a broken left rearview mirror, a broken left headlight, and a scratch on the left panel.
[0082] It should be noted that when the target abnormal vehicle is faking a collision between two vehicles, it generally chooses the left side of its body to collide with other vehicles. This is because the driver of the target abnormal vehicle is on the left side of the vehicle, so the left side of its body is chosen to collide with other vehicles in order to control the collision angle and collision force. Therefore, based on the collision position of the left side of the body of the vehicle to be detected, it is possible to more accurately determine whether the vehicle to be detected is the target abnormal vehicle.
[0083] Step 203: Determine a speed weighting coefficient of the vehicle to be detected based on the first real-time speed and the first average speed of the vehicle to be detected in the driving speed information.
[0084] Step 204 : determining a following weighted coefficient of the vehicle to be detected based on the relative position between the vehicle to be detected and other vehicles in the relative position information.
[0085] Optionally, the relative position information includes the relative position between the vehicle to be detected and other vehicles in at least two frames of the first captured images. When the relative position between the vehicle to be detected and other vehicles in at least two frames of the first captured images is determined as the target relative position, the following weighted coefficient of the vehicle to be detected is determined as the first following coefficient k f , and the first following coefficient k f Greater than 1.
[0086] Furthermore, the number of camera devices that capture at least two frames of the first captured images is at least two. The acquisition interval of at least two frames of the first captured images is greater than a preset time threshold. The target relative position includes the lane where the vehicle to be detected is in the right rear lane of other vehicles, and the lane where the vehicle to be detected is located is adjacent to the lane where the other vehicles are located.
[0087] Figure 3 is a schematic diagram of the relative position between the vehicle to be detected and other vehicles in an embodiment of the present invention, such as Figure 3 As shown, the vehicle to be detected is in the position corresponding to the right rear lane of other vehicles, and the vehicle to be detected follows the right rear lane of other vehicles. When other vehicles change lanes to the lane where the vehicle to be detected is located, the vehicle to be detected collides with it and holds the owner of the vehicle involved responsible on the grounds that the other vehicle changed lanes illegally. The owner of the vehicle involved here is the owner of the other vehicle.
[0088] Step 205 , determining a first abnormal parameter corresponding to the vehicle to be detected based on the vehicle type weighting coefficient, the collision weighting coefficient, the speed weighting coefficient and the following weighting coefficient.
[0089] Optionally, based on the preset basic parameter value, the vehicle type weighting coefficient, the collision weighting coefficient, the speed weighting coefficient and the following weighting coefficient, the first abnormal parameter corresponding to the vehicle to be detected is determined, wherein the first abnormal parameter can be expressed by formula (2):
[0090] P first =ΔP*k car *k f *k vm *k mark (2)
[0091] Among them, P first represents the first abnormal parameter, ΔP represents the preset basic parameter value, k car Represents the vehicle model weighting coefficient, k mark represents the collision weighting coefficient, k vm represents the speed weighting coefficient, k f represents the following weighting coefficient. Optionally, the preset basic parameter value ΔP is 10 or 100.
[0092] In one embodiment, Figure 4 As shown, the step of determining the speed weighting coefficient of the vehicle to be detected based on the first real-time speed and the first average speed of the vehicle to be detected in the driving speed information in the above step 203 specifically includes:
[0093] Step 301: Determine a first real-time speed average of the vehicle to be detected based on the first real-time speed of the vehicle to be detected when it passes through at least two camera devices.
[0094] Optionally, taking two camera devices as an example, based on the first real-time speed V of the vehicle to be detected when it passes through the first camera device camn and the first real-time speed V of the vehicle to be detected when it passes the second camera device camn +1, and the first speed average is determined as the first real-time speed average of the vehicle to be detected, that is, the first real-time speed average is
[0095] Step 302: Determine a first quantized mean value of the vehicle to be detected based on a preset quantized value and a first real-time mean value of the speed of the vehicle to be detected.
[0096] Optionally, taking two camera devices as an example, the preset quantization value k and the first real-time speed average of the vehicle to be detected are Multiply them together to get the first quantized mean of the vehicle to be detected. The preset quantization value k is much lower than the quantization value of and satisfies 0 <k<1。
[0097] Step 303: When a first average speed of the vehicle to be detected when passing through at least two camera devices is not greater than a first quantized mean value, determine a speed weighting coefficient based on the first speed coefficient.
[0098] Optionally, the first average speed of the vehicle to be detected when it passes through the two camera devices Not greater than the first quantized mean Right now In the case of v Determined as the speed weighting coefficient, where the first speed coefficient k v Greater than 1.
[0099] It should be noted that, when the first average speed of the vehicle to be detected when passing through the two camera devices is not greater than the first quantized mean, it means that the average speed of the vehicle to be detected is much lower than the average value of the real-time speeds when the two camera devices capture the pictures. Therefore, the speed of the vehicle to be detected on this section of highway is abnormally low, and it is very likely that a deliberate forged collision between the two vehicles has occurred.
[0100] In one embodiment, Figure 5 As shown, the step of determining the speed weighting coefficient based on the first speed coefficient in the above step 303 specifically includes:
[0101] Step 401: When it is determined that the vehicle to be detected has a following driving behavior, a second real-time speed and a second average speed of the followed vehicle when the followed vehicle passes through at least two camera devices are obtained.
[0102] Step 402: Determine an average second real-time speed of the followed vehicle based on the second real-time speed of the followed vehicle when it passes through at least two camera devices.
[0103] Optionally, taking two cameras as an example, based on the second speed average of the first real-time speed v1 of the followed vehicle when it passes the first camera and the first real-time speed v2 of the followed vehicle when it passes the second camera, the second speed average is determined as the second real-time speed average of the followed vehicle.
[0104] Step 403: Determine a second quantized mean value of the followed vehicle based on the preset quantized value and the second real-time mean value of the speed of the followed vehicle.
[0105] Optionally, taking two camera devices as an example, the preset quantization value k is multiplied by the second real-time speed mean of the followed vehicle to obtain the second quantization mean of the followed vehicle, wherein the preset quantization value k is much lower than the quantization value of and satisfies 0 <k<1。
[0106] Step 404: when the second average speed of the followed vehicle passing through at least two camera devices is not greater than the second quantized mean value, determine a speed weighting coefficient based on the first speed coefficient and the second speed coefficient.
[0107] Optionally, when the second average speed of the followed vehicle passing through the two camera devices is not greater than the second quantized mean value, the first speed coefficient k v and the second speed coefficient k vf Multiply them together to obtain the speed weighting coefficient of the detected vehicle, where the second speed coefficient k vf Greater than 1.
[0108] It should be noted that, when the second average speed of the followed vehicle when passing through the two camera devices is not greater than the second quantized mean, it means that the average speed of the followed vehicle is much lower than the average value of the real-time speeds when the two camera devices capture the pictures. Therefore, the speed of the followed vehicle on this section of the highway is abnormally low, and it is very likely that a deliberate forged collision between the two vehicles has occurred.
[0109] In one embodiment, the preset abnormal screening threshold includes a first abnormal threshold and a first quantity threshold; Figure 6 As shown, in the above step 104, based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and the preset abnormal screening threshold, the step of determining a target abnormal vehicle from at least one vehicle to be detected specifically includes:
[0110] Step 501 : for each vehicle to be detected, based on the first abnormal parameter corresponding to the vehicle to be detected during each trip, a first parameter set corresponding to the vehicle to be detected is determined.
[0111] Specifically, based on the first abnormal parameter first corresponding to the vehicle to be detected during each trip, the first parameter set G corresponding to the vehicle to be detected is determined. first_car , that is, G first_car =∑P first .
[0112] Step 502: Determine at least one candidate abnormal vehicle based on the first parameter set, the first abnormal threshold and the first quantity threshold of each vehicle to be detected.
[0113] Specifically, obtain the first parameter set of the vehicle to be detected that is greater than the preset first abnormal threshold P first_threshold The first parameter number C of the first exception parameter first_car , and in the first parameter number C first_car Greater than a preset first quantity threshold C first_threshold In the case of probable_car Optionally, the first abnormal threshold P first_thresholdGreater than the preset basic parameter value ΔP.
[0114] Step 503: For each candidate abnormal vehicle, a first driving moment of the candidate abnormal vehicle and a second captured image of the passengers when the candidate abnormal vehicle travels in each preset time period are obtained.
[0115] Step 504 , based on the first travel time of each candidate abnormal vehicle when traveling within each preset time period and the second captured image of the passenger, a target abnormal vehicle is determined from at least one candidate abnormal vehicle.
[0116] In one embodiment, the preset abnormal screening threshold also includes a second abnormal threshold and a second quantity threshold; Figure 7 As shown, in the above step 504, based on the first travel time of each candidate abnormal vehicle when traveling in each preset time period and the second captured image of the passenger, the step of determining the target abnormal vehicle from at least one candidate abnormal vehicle specifically includes:
[0117] Step 601, based on the first travel time, the first weight coefficient corresponding to each preset time period, the first start time and the first end time, determine the first travel coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period.
[0118] In one embodiment, the preset time period is a time period corresponding to a holiday, including a first time period before the holiday, a second time period during the holiday, and a third time period after the holiday, wherein the first start time represents the start time of the second time period, and the first end time represents the end time of the second time period.
[0119] Further, the first duration of the first time period and the second duration of the third time period are determined based on the duration of the vacation. Generally, the longer the duration of the vacation, the longer the first duration and the second duration.
[0120] Optionally, when the first duration is equal to the second duration, the first duration and the second duration can be collectively referred to as the target duration. The following table (1) illustrates the first weight coefficient corresponding to at least one holiday and the numerical setting method of the target duration, wherein at least one holiday includes the Spring Festival holiday, the Qingming holiday, the May Day holiday, the Dragon Boat Festival holiday, the Mid-Autumn Festival holiday, the National Day holiday and the New Year's Day holiday:
[0121] Table 1 The first weight coefficient and target time length corresponding to at least one holiday
[0122] Holiday Type Spring Festival Qingming May Day Duan Festival Mid-Autumn Festival National Day New Year The first weight coefficient 1.5 1.1 1.3 1.2 1.2 1.4 1.2 Target duration 5 2 3 2 2 3 2
[0123] The first weight coefficient corresponding to the holiday is determined based on the duration of the holiday. Generally, the longer the holiday lasts, the higher the first weight coefficient. In Table 1, the first duration before the holiday is equal to the second duration after the holiday. For example, the first duration before the Spring Festival holiday and the first duration after the Spring Festival holiday are both 5 days.
[0124] Optionally, based on the preset attenuation coefficient, the first driving time, the first weight coefficient corresponding to each holiday, the first start time and the first end time, the first travel coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period is determined, wherein the first travel coefficient can be expressed by formula (3):
[0125]
[0126] Among them, k 节 represents the first trip coefficient, T i represents the first driving moment, k h represents the first weight coefficient, T 始 represents the first starting time, T 终 represents the first end time, α represents the preset attenuation coefficient, and α<1.
[0127] Step 602, based on the facial image of the passenger in the second captured image, determine the passenger information corresponding to the candidate abnormal vehicle, and determine the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the passenger information.
[0128] The passenger information includes the first number of passengers, and at least one of the employment record, gender, height and weight of each passenger on the candidate abnormal vehicle.
[0129] Further, when it is determined that the first number of passengers in the candidate abnormal vehicle is greater than the preset number threshold, the second number of passengers in the candidate abnormal vehicle that meets the preset constraint condition is obtained, and based on the second number of passengers that meets the preset constraint condition and the preset first passenger coefficient, the passenger coefficient corresponding to the candidate abnormal vehicle is determined, wherein the passenger coefficient can be expressed by formula (4):
[0130]
[0131] Among them, k person represents the passenger coefficient corresponding to the candidate abnormal vehicle, k person-basic represents the first personnel coefficient, num represents the second number of passengers that meet the preset constraints. Optionally, the preset constraints include at least one of the following constraints: male gender, weight greater than a preset weight threshold, and no fixed job. Optionally, the first passenger coefficient k person-basicGreater than 1.
[0132] Step 603, based on the first abnormal parameter, the first travel coefficient and the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period, determine the second abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each preset time period.
[0133] Optionally, the first abnormal parameter, the first travel coefficient and the passenger coefficient corresponding to the candidate abnormal vehicle traveling in each preset time period are multiplied to obtain the second abnormal parameter corresponding to the candidate abnormal vehicle traveling in each preset time period, wherein the second abnormal parameter can be expressed by formula (5):
[0134] P second =P first *k 节 *k person (5)
[0135] Among them, P second Represents the second abnormal parameter, P first represents the first abnormal parameter, k 节 represents the first trip coefficient, k person Represents the passenger coefficient.
[0136] Step 604: Determine a second parameter set corresponding to the candidate abnormal vehicle based on the second abnormal parameters corresponding to the candidate abnormal vehicle when the candidate abnormal vehicle travels within each preset time period.
[0137] Specifically, based on the second abnormal parameter P corresponding to the candidate abnormal vehicle when traveling in each preset time period second , determine the second parameter set G corresponding to the candidate abnormal vehicle second_car , that is, G second_car =∑P second .
[0138] Step 605 : determining a target abnormal vehicle from at least one candidate abnormal vehicle based on the second parameter set, the second abnormal threshold, and the second quantity threshold of each candidate abnormal vehicle.
[0139] Specifically, the second parameter set of candidate abnormal vehicles is obtained, which is greater than the preset second abnormal threshold P second_threshold The second parameter number C of the second exception parameter second_car , and in the second parameter the number C second_car Greater than a preset second quantity threshold C second_threshold In the case of suspicion_car Optionally, the second abnormal threshold P second_threshold Greater than the preset basic parameter value ΔP.
[0140] In one embodiment, Figure 8 As shown, the abnormal vehicle detection method provided by the present invention also includes:
[0141] Step 701, determining the license plate number and model of the current vehicle based on the third captured image of the current vehicle, and obtaining the current transaction number and current service life of the current vehicle based on the license plate number of the current vehicle.
[0142] The third captured image of the current vehicle may be a captured image of the current vehicle captured by a camera device at an entrance of a highway.
[0143] Specifically, feature extraction is performed on the third captured image of the current vehicle to obtain vehicle appearance information corresponding to the current vehicle in the third captured image, where the vehicle appearance information includes license plate information, vehicle logo information, color information, shape information, size information and other information.
[0144] Furthermore, the license plate number of the current vehicle is determined based on the license plate information of the current vehicle, and the model of the current vehicle is determined based on the logo information, color information, shape information, size information and other information of the current vehicle.
[0145] The model of the current vehicle is determined based on the vehicle brand corresponding to the current vehicle and the vehicle series and vehicle level corresponding to the vehicle brand. For example, the vehicle series includes A series, Q series, R series, TT series and S series. The vehicle level includes AO level, A level, B level, C level and D level.
[0146] Step 702, when the current number of transactions is greater than a preset number threshold, the current age is greater than a preset age threshold, and the model of the current vehicle is the target model, the current vehicle is determined as a vehicle to be detected.
[0147] The target vehicle model includes a vehicle series under at least one vehicle brand and a vehicle model determined by a vehicle level, wherein the vehicle series and the vehicle level are of at least one type. Optionally, the preset age threshold X can be set to 2 to 4 years. Furthermore, the target vehicle model is a high-end vehicle.
[0148] Furthermore, when the current transaction number includes the first purchase transaction, the preset number threshold may be set to m. When the current transaction number does not include the first purchase transaction, the preset number threshold may be set to m-1. For example, m may be set to 1, and the current transaction number greater than 1 indicates that the current vehicle is a used vehicle.
[0149] In the above steps 701 to 702, the license plate number and model of the current vehicle are determined based on the third captured image of the current vehicle, and the current number of transactions and the current age of use of the current vehicle are obtained based on the license plate number of the current vehicle. When the current number of transactions is greater than the preset number threshold, the current age of use is greater than the preset age threshold, and the model of the current vehicle is the target model, the current vehicle is determined as the vehicle to be detected, so as to screen out high-end vehicles that have changed hands many times as the vehicles to be detected. Since the high-end vehicles that have changed hands many times are likely to be abnormal vehicles that deliberately forged a collision between two vehicles, the detection efficiency and detection success rate of the target abnormal vehicles can be improved.
[0150] The abnormal vehicle detection device provided by the present invention is described below. The abnormal vehicle detection device described below and the abnormal vehicle detection method described above can be referenced to each other.
[0151] like Fig. 9 As shown, the present invention provides an abnormal vehicle detection device, and the abnormal vehicle detection device 100 includes:
[0152] The image acquisition module 1001 is used to obtain captured image data of at least one vehicle to be detected captured by a camera device, where the captured image data includes a first captured image of the vehicle to be detected during at least one travel process.
[0153] The information extraction module 1002 is used to obtain target feature information of each vehicle to be detected based on the first captured image of the vehicle to be detected during each trip, and the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip.
[0154] The parameter determination module 1003 is used to determine the first abnormal parameter corresponding to the vehicle to be detected during each trip based on the target feature information, and the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that fakes a traffic accident.
[0155] The abnormal screening module 1004 is used to determine a target abnormal vehicle from at least one vehicle to be detected based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold.
[0156] In one embodiment, the target feature information includes vehicle appearance information, vehicle body collision information, driving speed information, and relative position information with other vehicles; the parameter determination module 1003 includes a vehicle type coefficient determination unit, a collision coefficient determination unit, a speed coefficient determination unit, a following coefficient determination unit, and an abnormal parameter determination unit, wherein:
[0157] The vehicle model coefficient determination unit is used to determine the vehicle model weighting coefficient of the vehicle to be detected based on the license plate information and vehicle model information of the vehicle to be detected in the vehicle appearance information.
[0158] The collision coefficient determination unit is used to determine the collision weighting coefficient of the vehicle to be detected based on the collision position of the vehicle to be detected in the vehicle body collision information.
[0159] The speed coefficient determination unit is used to determine the speed weighting coefficient of the vehicle to be detected based on the first real-time speed and the first average speed of the vehicle to be detected in the driving speed information.
[0160] The following coefficient determination unit is used to determine the following weighted coefficient of the vehicle to be detected based on the relative position between the vehicle to be detected and other vehicles in the relative position information.
[0161] The abnormal parameter determination unit is used to determine the first abnormal parameter corresponding to the vehicle to be detected based on the vehicle type weighting coefficient, the collision weighting coefficient, the speed weighting coefficient and the following weighting coefficient.
[0162] In one embodiment, the speed coefficient determination unit is also used to determine the first real-time speed average of the vehicle to be detected based on the first real-time speed of the vehicle to be detected when it passes through at least two camera devices; determine the first quantized mean of the vehicle to be detected based on a preset quantized value and the first real-time speed average of the vehicle to be detected; and determine the speed weighting coefficient based on the first speed coefficient when the first average speed of the vehicle to be detected when it passes through at least two camera devices is not greater than the first quantized mean.
[0163] In one embodiment, the speed coefficient determination unit is also used to obtain the second real-time speed and the second average speed of the followed vehicle when it is determined that the vehicle to be detected has a following driving behavior; determine the second real-time speed average of the followed vehicle based on the second real-time speed of the followed vehicle when it passes through at least two camera devices; determine the second quantized mean of the followed vehicle based on a preset quantized value and the second real-time speed average of the followed vehicle; and determine the speed weighting coefficient based on the first speed coefficient and the second speed coefficient when the second average speed of the followed vehicle when it passes through at least two camera devices is not greater than the second quantized mean.
[0164] In one embodiment, the preset abnormal screening threshold includes a first abnormal threshold and a first quantity threshold; the abnormal screening module 1004 is also used to determine, for each vehicle to be detected, a first parameter set corresponding to the vehicle to be detected based on the first abnormal parameter corresponding to the vehicle to be detected during each trip; based on the first parameter set, the first abnormal threshold and the first quantity threshold of each vehicle to be detected, at least one candidate abnormal vehicle is determined; for each candidate abnormal vehicle, a first driving moment of the candidate abnormal vehicle when traveling within each preset time period and a second snapshot image of the passengers are obtained; based on the first driving moment of each candidate abnormal vehicle when traveling within each preset time period and the second snapshot image of the passengers, a target abnormal vehicle is determined from at least one candidate abnormal vehicle.
[0165] In one embodiment, the preset abnormal screening threshold also includes a second abnormal threshold and a second quantity threshold; the abnormal screening module 1004 is also used to determine the first travel coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the first driving time, the first weight coefficient corresponding to each preset time period, the first start time and the first end time; determine the passenger information corresponding to the candidate abnormal vehicle based on the facial image of the passenger in the second captured image, and determine the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the passenger information; determine the second abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each preset time period based on the first abnormal parameter, the first travel coefficient and the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each preset time period; determine the second parameter set corresponding to the candidate abnormal vehicle based on the second abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each preset time period; determine the target abnormal vehicle from at least one candidate abnormal vehicle based on the second parameter set of each candidate abnormal vehicle, the second abnormal threshold and the second quantity threshold.
[0166] In one embodiment, the abnormal vehicle detection device 100 also includes: a vehicle screening module to be inspected, which is used to determine the license plate number and model of the current vehicle based on the third captured image of the current vehicle, and obtain the current transaction number and current service life of the current vehicle based on the license plate number of the current vehicle; when the current transaction number is greater than a preset number threshold, the current service life is greater than a preset age threshold, and the model of the current vehicle is the target model, the current vehicle is determined as a vehicle to be inspected.
[0167] Fig.10 An example of a physical structure diagram of an electronic device is shown in FIG. Fig.10As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the abnormal vehicle detection method provided by the above methods, the method comprising: obtaining the captured image data of at least one vehicle to be detected collected by the camera device, the captured image data comprising the first captured image of the vehicle to be detected during at least one trip; for each vehicle to be detected, based on the first captured image of the vehicle to be detected during each trip, obtaining the target feature information of the vehicle to be detected, the target feature information is used to characterize the abnormal information of the vehicle to be detected during the trip; based on the target feature information, determining the first abnormal parameter corresponding to the vehicle to be detected during each trip, the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forges a traffic accident; based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and the preset abnormal screening threshold, determining the target abnormal vehicle from at least one vehicle to be detected.
[0168] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.
[0169] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the abnormal vehicle detection method provided by the above-mentioned methods, which includes: obtaining captured image data of at least one vehicle to be detected collected by a camera device, the captured image data including a first captured image of the vehicle to be detected during at least one travel process; for each vehicle to be detected, based on the first captured image of the vehicle to be detected during each travel process, obtaining target feature information of the vehicle to be detected, the target feature information is used to characterize abnormal information of the vehicle to be detected during the travel process; based on the target feature information, determining a first abnormal parameter corresponding to the vehicle to be detected during each travel process, the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forges a traffic accident; based on the first abnormal parameter corresponding to each vehicle to be detected during each travel process and a preset abnormal screening threshold, determining a target abnormal vehicle from at least one vehicle to be detected.
[0170] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the abnormal vehicle detection method provided by the above-mentioned methods, the method comprising: obtaining captured image data of at least one vehicle to be detected acquired by a camera device, the captured image data comprising a first captured image of the vehicle to be detected during at least one travel process; for each vehicle to be detected, based on the first captured image of the vehicle to be detected during each travel process, obtaining target feature information of the vehicle to be detected, the target feature information being used to characterize abnormal information of the vehicle to be detected during the travel process; based on the target feature information, determining a first abnormal parameter corresponding to the vehicle to be detected during each travel process, the first abnormal parameter being used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forges a traffic accident; based on the first abnormal parameter corresponding to each vehicle to be detected during each travel process and a preset abnormal screening threshold, determining a target abnormal vehicle from at least one vehicle to be detected.
[0171] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.
[0172] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.
[0173] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for detecting abnormal vehicles, characterized in that: include: Acquire captured image data of at least one vehicle to be detected collected by a camera device, wherein the captured image data includes a first captured image of the vehicle to be detected during at least one travel process; For each of the vehicles to be detected, based on the first captured image of the vehicle to be detected during each trip, obtaining target feature information of the vehicle to be detected, wherein the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip; Based on the target feature information, determining a first abnormal parameter corresponding to the vehicle to be detected during each trip, wherein the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forged a traffic accident; Based on the first abnormal parameter corresponding to each of the to-be-detected vehicles during each trip and a preset abnormal screening threshold, a target abnormal vehicle is determined from the at least one to-be-detected vehicle.
2. The abnormal vehicle detection method according to claim 1, characterized in that: The target feature information includes vehicle appearance information, vehicle body collision information, driving speed information, and relative position information with other vehicles; The determining, based on the target feature information, a first abnormal parameter corresponding to each trip of the vehicle to be detected includes: Determining a weighting coefficient of the vehicle type of the vehicle to be detected based on the license plate information and vehicle type information of the vehicle to be detected in the vehicle appearance information; Determining a collision weighting coefficient of the vehicle to be detected based on the collision position of the vehicle to be detected in the vehicle body collision information; Determining a speed weighting coefficient of the vehicle to be detected based on a first real-time speed and a first average speed of the vehicle to be detected in the driving speed information; Determining a following weighting coefficient of the vehicle to be detected based on the relative position between the vehicle to be detected and the other vehicles in the relative position information; Based on the vehicle type weighting coefficient, the collision weighting coefficient, the speed weighting coefficient and the following weighting coefficient, a first abnormal parameter corresponding to the vehicle to be detected is determined.
3. The abnormal vehicle detection method according to claim 2, characterized in that: The determining of the speed weighting coefficient of the vehicle to be detected based on the first real-time speed and the first average speed of the vehicle to be detected in the driving speed information comprises: Determine a first real-time speed average of the vehicle to be detected based on the first real-time speed of the vehicle to be detected when it passes through at least two camera devices; Determining a first quantized mean value of the vehicle to be detected based on a preset quantized value and a first real-time mean value of the speed of the vehicle to be detected; When a first average speed of the vehicle to be detected when passing through the at least two camera devices is not greater than the first quantized mean value, the speed weighting coefficient is determined based on a first speed coefficient.
4. The abnormal vehicle detection method according to claim 3, characterized in that: The determining the speed weighting coefficient based on the first speed coefficient comprises: In the case where it is determined that the vehicle to be detected has a following driving behavior, obtaining a second real-time speed and a second average speed of the followed vehicle when the followed vehicle passes through the at least two camera devices; Determining a second real-time speed average of the followed vehicle based on the second real-time speed of the followed vehicle when the followed vehicle passes through the at least two camera devices; Determining a second quantized mean value of the followed vehicle based on a preset quantized value and a second real-time mean value of the speed of the followed vehicle; The speed weighting coefficient is determined based on the first speed coefficient and the second speed coefficient when a second average speed of the followed vehicle when passing the at least two imaging devices is not greater than the second quantized mean value.
5. The abnormal vehicle detection method according to any one of claims 1 to 4, characterized in that: The preset abnormal screening threshold includes a first abnormal threshold and a first quantity threshold; The step of determining a target abnormal vehicle from the at least one vehicle to be detected based on the first abnormal parameter corresponding to each of the vehicles to be detected during each trip and a preset abnormal screening threshold comprises: For each of the vehicles to be detected, determining a first parameter set corresponding to the vehicle to be detected based on the first abnormal parameter corresponding to the vehicle to be detected during each trip; Determine at least one candidate abnormal vehicle based on the first parameter set of each of the vehicles to be detected, the first abnormal threshold, and the first quantity threshold; For each candidate abnormal vehicle, obtaining a first driving time of the candidate abnormal vehicle and a second captured image of the passengers when the candidate abnormal vehicle travels in each preset time period; The target abnormal vehicle is determined from the at least one candidate abnormal vehicle based on a first travel time of each candidate abnormal vehicle when traveling within each preset time period and a second captured image of the passenger.
6. The abnormal vehicle detection method according to claim 5, characterized in that: The preset abnormal screening threshold also includes a second abnormal threshold and a second quantity threshold; The step of determining the target abnormal vehicle from the at least one candidate abnormal vehicle based on the first travel time of each candidate abnormal vehicle and the second captured image of the passenger in each preset time period includes: Determine a first travel coefficient corresponding to the candidate abnormal vehicle when traveling in each of the preset time periods based on the first travel time, a first weight coefficient corresponding to each of the preset time periods, a first start time, and a first end time; Based on the facial image of the passenger in the second captured image, determine the passenger information corresponding to the candidate abnormal vehicle, and based on the passenger information, determine the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each of the preset time periods; Determine a second abnormal parameter corresponding to the candidate abnormal vehicle when traveling in each of the preset time periods based on the first abnormal parameter, the first travel coefficient, and the passenger coefficient corresponding to the candidate abnormal vehicle when traveling in each of the preset time periods; Determine a second parameter set corresponding to the candidate abnormal vehicle based on the second abnormal parameters corresponding to the candidate abnormal vehicle when traveling within each of the preset time periods; A target abnormal vehicle is determined from the at least one abnormal vehicle candidate based on the second parameter set of each abnormal vehicle candidate, the second abnormal threshold, and the second quantity threshold.
7. The abnormal vehicle detection method according to any one of claims 1 to 4, characterized in that: The method comprises: Determine the license plate number and model of the current vehicle based on the third captured image of the current vehicle, and obtain the current transaction number and current service life of the current vehicle based on the license plate number of the current vehicle; When the current number of transactions is greater than a preset number threshold, the current service life is greater than a preset life threshold, and the model of the current vehicle is the target model, the current vehicle is determined as the vehicle to be detected.
8. An abnormal vehicle detection device, characterized in that: include: An image acquisition module, used to acquire captured image data of at least one vehicle to be detected captured by a camera device, wherein the captured image data includes a first captured image of the vehicle to be detected during at least one trip; An information extraction module is used to obtain target feature information of each vehicle to be detected based on a first captured image of the vehicle to be detected during each trip, wherein the target feature information is used to characterize abnormal information of the vehicle to be detected during the trip; A parameter determination module, used to determine, based on the target feature information, a first abnormal parameter corresponding to the vehicle to be detected during each trip, wherein the first abnormal parameter is used to characterize the probability that the vehicle to be detected is an abnormal vehicle that forged a traffic accident; The abnormal screening module is used to determine a target abnormal vehicle from the at least one vehicle to be detected based on the first abnormal parameter corresponding to each vehicle to be detected during each trip and a preset abnormal screening threshold.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the abnormal vehicle detection method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the abnormal vehicle detection method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Big data based vehicle point accumulation early warning system and method
CN106875692A
Accident vehicle identification method and terminal device
CN109886820A
Highway vehicle abnormal behavior early warning system based on 5G communication
CN111354193A
Vehicle speed anomaly detection method, device and equipment and storage medium
CN111627224A
Method and device for detecting abnormal lane change of vehicle
CN114627409A
