CN113157817B - Method, device and computer equipment for distinguishing drivers - Google Patents
Method, device and computer equipment for distinguishing drivers Download PDFInfo
- Publication number
- CN113157817B CN113157817B CN202110276146.9A CN202110276146A CN113157817B CN 113157817 B CN113157817 B CN 113157817B CN 202110276146 A CN202110276146 A CN 202110276146A CN 113157817 B CN113157817 B CN 113157817B
- Authority
- CN
- China
- Prior art keywords
- data
- track
- historical driving
- map
- driving track
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000003491 array Methods 0.000 claims abstract description 48
- 238000004364 calculation method Methods 0.000 claims abstract description 30
- 238000003062 neural network model Methods 0.000 claims abstract description 28
- 230000006399 behavior Effects 0.000 claims abstract description 8
- 239000013598 vector Substances 0.000 claims description 58
- 230000006870 function Effects 0.000 claims description 32
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Biophysics (AREA)
- Fuzzy Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Traffic Control Systems (AREA)
Abstract
The application relates to the field of artificial intelligence, and discloses a method for distinguishing drivers, which comprises the following steps: acquiring historical driving track data corresponding to a specified vehicle; matching the historical driving track data into a map grid to form a track map; forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map; inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays; and distinguishing the corresponding appointed running track when the same driver drives in each historical running track data according to the classification weight and the preset threshold value. The historical driving track data and the map grids are overlapped to form a track map, each piece of historical driving track data is marked by the calculation feature data array, the similarity between the feature data arrays of each piece of historical driving track data is calculated according to the twin neural network model, driving behaviors of different drivers are distinguished, and the driver is distinguished and classified.
Description
Technical Field
The present application relates to the field of artificial intelligence, and more particularly to a method, apparatus and computer device for distinguishing drivers.
Background
The insurance premium is different for the designated driver than for the unspecified driver, and the benefit is different. The familiarity of the driver to the vehicle directly influences the driving behavior, and the insurance applicant is that the risk probability of the appointed driver is far lower than that of other temporary drivers, and the premium discount is more. However, in the insurance of car insurance, after the insurance applicant borrows others to drive and deal with the insurance, the phenomenon that the insurance applicant falsely claim to be drives and deals with the insurance easily to form the fraudulent claims of the insurance is easy to occur, and how to effectively judge whether the insurance applicant is in the insurance of car insurance becomes the problem of the insurance company's nuclear claims.
Disclosure of Invention
The application mainly aims to provide a method for distinguishing drivers, which aims to solve the technical problem that the driver cannot be distinguished according to the characteristics of a driving track.
The application provides a method for distinguishing drivers, which comprises the following steps:
acquiring historical driving track data corresponding to a specified vehicle;
matching the historical driving track data into a map grid to form a track map;
Forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays;
and distinguishing the corresponding appointed running track when the same driver drives in each historical running track data according to the classification weight and the preset threshold value.
Preferably, the step of matching the historical driving track data into a map grid to form a track map includes:
Dividing the map into map grids according to a preset dividing mode;
acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data;
Superposing the appointed historical driving track data into the map grid according to the longitude and latitude information;
and according to the mode that the specified historical driving track data are overlapped in the map grid, all the historical driving track data are overlapped in the map grid in a one-to-one correspondence mode, and the track map is formed.
Preferably, the feature data array includes track feature data, and the step of forming feature data arrays corresponding to the historical driving track data respectively according to the track map includes:
Acquiring each map grid interval occupied by the appointed historical driving track data and the corresponding speed limit label of each map grid interval;
Counting track characteristic data corresponding to the appointed historical driving track data according to speed limit labels respectively corresponding to each map grid interval, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio;
and counting the track characteristic data corresponding to all the historical driving track data respectively according to the counting mode of the track characteristic data corresponding to the appointed historical driving track data.
Preferably, the feature data array includes driving feature data of a driver, and the step of forming feature data arrays corresponding to the historical driving track data according to the track map includes:
acquiring time data corresponding to each track point corresponding to the appointed historical driving track data;
calculating driver driving characteristic data corresponding to the appointed historical driving track data according to the time data, wherein the driver driving characteristic data comprises preset time interval driving data, driving smoothness and fatigue driving data;
And calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
Preferably, before the step of inputting each of the feature data arrays into the twin neural network model, the step of calculating the classification weights between the feature data arrays comprises:
Inputting training samples into a twin neural network model, and mapping the training samples into a high-dimensional space through a specified function to obtain space vectors corresponding to the training samples respectively, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples;
Calculating a first space vector distance corresponding to a first sample and a second sample with the same category label and a second space vector distance corresponding to a third sample and a fourth sample with different category labels according to a first calculation formula, wherein the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance;
adjusting parameters of the specified function to enable the distance of the first space vector to be smaller, and enabling the distance of the second space vector to be larger;
Judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum;
if yes, determining the parameter of the appointed function as a fixed parameter.
Preferably, the step of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data according to the classification weight and the preset threshold value includes:
dividing the historical driving track data with the classification weight greater than or equal to the preset threshold into a first set and a second set which correspond to two drivers respectively, and merging the historical driving track data with the classification weight smaller than the preset threshold into a set corresponding to the same driver;
and respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
Preferably, after the step of distinguishing the specified driving trajectories corresponding to the same driver driving in each of the historical driving trajectory data according to the classification weight and the preset threshold, the step of distinguishing the specified driving trajectories includes:
screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located;
taking the appointed set as a running track set corresponding to the applicant;
judging whether the current dangerous running track is contained in a running track set corresponding to the applicant or not;
If yes, judging that the type of the risk of the appointed driver is judged, otherwise, judging that the type of the risk of the appointed driver is not judged.
The application also provides a device for distinguishing drivers, which comprises:
the first acquisition module is used for acquiring historical driving track data corresponding to the appointed vehicle;
The first forming module is used for matching the historical driving track data into a map grid to form a track map;
The second forming module is used for forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
the first calculation module is used for inputting each characteristic data array into the twin neural network model, and calculating the classification weight between the characteristic data arrays;
And the distinguishing module is used for distinguishing the corresponding appointed driving track when the same driver drives in the historical driving track data according to the classification weight and the preset threshold value.
The application also provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the above method when executing the computer program.
The application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method described above.
According to the method, the historical driving track data and the map grids are overlapped to form the track map, each piece of historical driving track data is marked according to the track map calculation feature data array, then the similarity between the feature data arrays of each piece of historical driving track data is calculated according to the twin neural network model, driving behaviors of different drivers are distinguished, the drivers are distinguished and classified, and one-to-one correspondence between the driving track data and the drivers is achieved.
Drawings
FIG. 1 is a flow chart of a method for distinguishing drivers according to an embodiment of the present application;
FIG. 2 is a schematic flow diagram of a system for distinguishing drivers according to an embodiment of the present application;
FIG. 3 is a schematic diagram showing an internal structure of a computer device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Referring to fig. 1, a method of distinguishing drivers according to an embodiment of the present application includes:
S1: acquiring historical driving track data corresponding to a specified vehicle;
s2: matching the historical driving track data into a map grid to form a track map;
s3: forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
S4: inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays;
s5: and distinguishing the corresponding appointed running track when the same driver drives in each historical running track data according to the classification weight and the preset threshold value.
In the embodiment of the application, the historical driving track data corresponding to the appointed vehicle is obtained through an insurance APP account registered by an owner of the appointed vehicle, and the insurance APP collects the navigation track of each insurance driver as the historical driving track data in a mode of setting rewards or insurance conditions. The navigation track is a vector cluster in units of seconds, and comprises: time, longitude and latitude, altitude, direction, speed and the like. The map grid is obtained by applying a division algorithm on the existing map data of an open source, wherein the map data of the open source comprises the longitude and latitude of a road, the type of the road, the speed limit of the road and the like. The above-mentioned division algorithm includes an algorithm of dividing the map grid by a distance of specified distance, for example, dividing the map grid by a distance of 10 km or 20 km to form a map grid of 10 x 10 square km or 20 x 20 square km; or a dividing algorithm for dividing the map grids according to each administrative region in the map, such as Shenzhen Fufield grids, shenzhen mountain grids and the like. By superposing the historical driving track data and the map grid, a data sample for analyzing driving track characteristics is formed and is used for calculating characteristic data arrays respectively corresponding to the historical driving track data, such as, but not limited to, night driving data, peak period driving data, driving smoothness and the like.
According to the method, the characteristic data arrays corresponding to the two pieces of historical driving track data are input into the twin neural network model, the similarity of the two pieces of historical driving track data is calculated through the functions contained in the twin neural network model, and then the probability that the two pieces of historical driving track data conform to the driving characteristics of the same driver is judged, and further whether a specified vehicle is driven by the same driver registered by insurance is determined.
According to the method, the historical driving track data and the map grids are overlapped to form the track map, each piece of historical driving track data is marked according to the track map calculation feature data array, then the similarity between the feature data arrays of each piece of historical driving track data is calculated according to the twin neural network model, driving behaviors of different drivers are distinguished, the drivers are distinguished and classified, and one-to-one correspondence between the driving track data and the drivers is achieved.
Further, the step S2 of matching the historical driving track data to a map grid to form a track map includes:
S21: dividing the map into map grids according to a preset dividing mode;
S22: acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data;
s23: superposing the appointed historical driving track data into the map grid according to the longitude and latitude information;
S24: and according to the mode that the specified historical driving track data are overlapped in the map grid, all the historical driving track data are overlapped in the map grid in a one-to-one correspondence mode, and the track map is formed.
In the embodiment of the application, the specified historical driving track data is superimposed into the corresponding grids according to the longitude and latitude data to form the track map with one-to-one correspondence between the historical driving track data and the map grids, and the track map is displayed as a composite of the map grids and the track lines corresponding to the historical driving track data respectively. The background data of the track map is represented by the combination of a data table corresponding to the historical driving track data and a data table corresponding to the map grid.
Further, the feature data array includes track feature data, and the step S3 of forming feature data arrays corresponding to the historical driving track data respectively according to the track map includes:
S31, acquiring each map grid interval occupied by the specified historical driving track data and the corresponding speed limit label of each map grid interval;
S32: counting track characteristic data corresponding to the appointed historical driving track data according to speed limit labels respectively corresponding to each map grid interval, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio;
S33: and counting the track characteristic data corresponding to all the historical driving track data respectively according to the counting mode of the track characteristic data corresponding to the appointed historical driving track data.
In the embodiment of the application, the track characteristic data takes each map grid as a calculation interval, and the interval value corresponds to the driving mileage of the driving track in the interval. The track characteristic data comprises a slow driving interval ratio and an overdrive interval ratio, wherein the overdrive interval ratio is subdivided into an interval ratio exceeding the highest speed limit, a road type speed limit interval ratio and an extreme overdrive interval ratio. The number of the driving record intervals with the slow driving interval ratio and the overspeed driving interval ratio being lower than 60kph is the ratio of the number of the driving record intervals to the total interval number corresponding to the whole driving track; the highest speed limit interval ratio is equal to the ratio of the number of overspeed record intervals with the vehicle speed exceeding 120kph to the total interval number; the road type speed limit interval ratio is equal to the proportion of the number of recording intervals with the speed exceeding the road type speed limit to the total number of intervals; the above-mentioned extreme overdrive interval ratio is equal to the ratio of the number of recorded intervals of extreme overdrive to the total number of intervals, and the extreme overdrive is defined over 20% of the speed limit.
Further, the feature data array includes driving feature data of the driver, and the step S3 of forming feature data arrays corresponding to the historical driving track data according to the track map includes:
S301: acquiring time data corresponding to each track point corresponding to the appointed historical driving track data;
S302: calculating driver driving characteristic data corresponding to the appointed historical driving track data according to the time data, wherein the driver driving characteristic data comprises preset time interval driving data, driving smoothness and fatigue driving data;
S303: and calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
In the embodiment of the application, in order to increase the accuracy of distinguishing different drivers according to the corresponding driving data characteristics of the drivers, the characteristic data array not only comprises track characteristic data, but also comprises driving characteristic data of the drivers, and the characteristic data array is a group of 8-dimensional data. The driving characteristic data is obtained by analyzing driving habits of a driver, and includes, for example, night driving data, rush hour driving data, driving smoothness, fatigue driving data, and the like. And acquiring the driving characteristic data by acquiring time data of the driving track data. For example, the driving time from 11 pm to 5 pm is recorded as 1, which represents night, and 0 is non-night, and the total number of intervals of the current driving track is counted to be used as night driving data by counting the number of intervals of the driving track with night; counting the number of intervals driven in the peak period, wherein the number of intervals is equal to the total number of intervals of the current driving track, the record mark is 1 as peak period driving data, the record mark is 1 to indicate that peak period driving exists, otherwise, the record mark is 0 to indicate that off-peak period driving exists, and the peak period comprises the early and late peak periods of working days; the driving smoothness is calculated by counting the times of rapid acceleration and rapid deceleration in the current driving track, and the speed change exceeds 100kph/10S, namely the rapid acceleration or rapid deceleration event; the fatigue driving data is obtained by counting the number of intervals with the fatigue driving marks, and the travel marks which are more than 2.5 hours of continuous driving are obtained by comparing the number of the intervals with the current counted running track.
Further, before the step S4 of inputting each of the feature data arrays into the twin neural network model and calculating the classification weights between the feature data arrays, the method includes:
s41: inputting training samples into a twin neural network model, and mapping the training samples into a high-dimensional space through a specified function to obtain space vectors corresponding to the training samples respectively, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples;
S42: calculating a first space vector distance corresponding to a first sample and a second sample with the same category label and a second space vector distance corresponding to a third sample and a fourth sample with different category labels according to a first calculation formula, wherein the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance;
S43: adjusting parameters of the specified function to enable the distance of the first space vector to be smaller, and enabling the distance of the second space vector to be larger;
S44: judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum;
S45: if yes, determining the parameter of the appointed function as a fixed parameter.
In the embodiment of the application, the parameters of the appointed function in the twin neural network are trained by taking the historical driving tracks respectively corresponding to different drivers in the statistical insurance APP as training samples and taking the characteristic data arrays respectively corresponding to the training samples as input data. The specified function is used for distinguishing the categories of different characteristic data arrays, and further distinguishing different drivers corresponding to the categories of the different characteristic data arrays. The twin neural network comprises two inputs and two networks which are designed in parallel, and can be divided into true and false twin networks according to whether the two networks share parameter weights or not. The twin neural network learns a similarity metric relationship from the input data and then uses the learned similarity metric relationship to compare and match samples of new unknown classes.
According to the embodiment of the application, the input data are mapped to a high-dimensional space by the appointed function in the twin neural network to form the space vectors corresponding to the input data respectively, and then the similarity of the two input data is judged by calculating the distance of the space vectors. According to the method, the parameters in the appointed function are determined by minimizing the loss function values of a pair of samples from the same category on the training samples and maximizing the loss function values of a stack of samples from different categories, and then after the parameters in the appointed function are determined, the classification information of the input data is obtained by receiving the similarity of the input data of the non-expressed category labels.
Further, the step S5 of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data according to the classification weight and the preset threshold value includes:
S51: dividing the historical driving track data with the classification weight greater than or equal to the preset threshold into a first set and a second set which correspond to two drivers respectively, and merging the historical driving track data with the classification weight smaller than the preset threshold into a set corresponding to the same driver;
s52: and respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
In the embodiment of the application, the historical driving track data corresponding to the appointed vehicle is classified by classification, each class is stored in one set, the corresponding set type, the set type quantity and the driving track quantity respectively corresponding to the set types are associated in the map grid, and the set type containing the maximum driving track quantity is corresponding to the appointed driver of insurance registration. The output value interval of the twin neural network model is a value between [ -1, 1], the absolute value of the output value is taken as the classification weight, the preset threshold value is 0.8, and the condition that the preset threshold value is smaller than 0.8 is considered to be that the same driver drives the appointed vehicle, otherwise, the preset threshold value is not the formal track data of the same driver driving the appointed vehicle. The smaller the value, the higher the probability of being presumed to be accurate for the same driver on the basis of the classification weight being smaller than 0.8.
In the embodiment of the application, the input characteristic data array is a group of 8-dimensional vectors, and the formula for calculating the similarity of two input data is as follows: Wherein/> Is a similarity value,/>Representing the first vector,/>Representing the second vector,/>I-th dimension representing a first vector,/>Representing the i-th dimension of the second vector.
Further, after step S5 of distinguishing the specified driving trajectories corresponding to the same driver driving in each of the historical driving trajectory data according to the classification weights and the preset threshold, the method includes:
S501: screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located;
s502: taking the appointed set as a running track set corresponding to the applicant;
S503: judging whether the current dangerous running track is contained in a running track set corresponding to the applicant or not;
s504: if yes, judging that the type of the risk of the appointed driver is judged, otherwise, judging that the type of the risk of the appointed driver is not judged.
In the embodiment of the application, the set type containing the maximum number of running tracks is corresponding to the designated driver registered by insurance, then the designated map grid corresponding to the insurance place and the designated set corresponding to the designated map grid are determined as judging standards, whether the current running track is contained in the running track set corresponding to the insurance applicant or not is judged by comparing the relation between the classification weight of the current running track and the characteristic data array in the designated set and the preset threshold value, if the classification weight of the characteristic data array of the current running track and the characteristic data array in the designated set is smaller than the preset threshold value, the current running track is contained in the running track set corresponding to the insurance applicant, and is the driving behavior of the designated driver registered by the insurance, belongs to the insurance type of the designated driver, and otherwise, the current running track is not the insurance type of the designated driver and is the insurance fraud.
Referring to fig. 2, an apparatus for distinguishing a driver according to an embodiment of the present application includes:
A first obtaining module 1, configured to obtain historical driving track data corresponding to a specified vehicle;
The first forming module 2 is used for matching the historical driving track data into a map grid to form a track map;
The second forming module 3 is used for forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
the first calculation module 4 is used for inputting each characteristic data array into the twin neural network model, and calculating the classification weight between the characteristic data arrays;
And the distinguishing module 5 is used for distinguishing the corresponding appointed running track when the same driver drives in the historical running track data according to the classification weight and the preset threshold value.
The explanation of the corresponding method part is applicable to the relevant explanation of the embodiment of the application, and is not repeated.
Further, the first forming module 2 includes:
the dividing unit is used for dividing the map into map grids according to a preset dividing mode;
The first acquisition unit is used for acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data;
the first superposition unit is used for superposing the appointed historical driving track data into the map grid according to the longitude and latitude information;
and the second superposition unit is used for superposing all the historical driving track data into the map grid in a one-to-one correspondence mode according to the mode that the specified historical driving track data is superposed into the map grid, so as to form the track map.
Further, the feature data array includes track feature data, and the second forming module 3 includes:
the second acquisition unit is used for acquiring each map grid interval occupied by the specified historical driving track data and the corresponding speed limit label of each map grid interval;
the first statistics unit is used for counting track characteristic data corresponding to the appointed historical driving track data according to the speed limit labels respectively corresponding to the map grid intervals, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio;
And the second statistical unit is used for counting the track characteristic data corresponding to all the historical driving track data respectively according to the statistical mode of the track characteristic data corresponding to the appointed historical driving track data.
Further, the feature data array includes driver driving feature data, and the second forming module 3 includes:
A third obtaining unit, configured to obtain time data corresponding to each track point corresponding to the specified historical driving track data;
a first calculation unit, configured to calculate driver driving feature data corresponding to the specified historical driving track data according to the time data, where the driver driving feature data includes preset time interval driving data, driving smoothness and fatigue driving data;
And the second calculation unit is used for calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
Further, the device for distinguishing drivers includes:
The input module is used for inputting training samples into the twin neural network model, mapping the training samples into a high-dimensional space through a specified function to obtain space vectors respectively corresponding to the training samples, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples;
The second calculation module is configured to calculate a first space vector distance corresponding to a first sample and a second sample with the same type of labels and a second space vector distance corresponding to a third sample and a fourth sample with different type of labels according to a first calculation formula, where the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance;
The adjusting module is used for adjusting the parameters of the specified function to enable the first space vector distance to be smaller, and meanwhile, the second space vector distance to be larger;
The first judging module is used for judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum;
And the determining module is used for determining that the parameter of the designated function is a fixed parameter if the parameters are simultaneously maximum.
Further, the differentiating module 5 includes:
The distinguishing unit is used for distinguishing the historical driving track data with the classification weight greater than or equal to the preset threshold value into a first set and a second set which are respectively corresponding to two drivers, and merging the historical driving track data with the classification weight smaller than the preset threshold value into a set corresponding to the same driver;
and the association unit is used for respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
Further, the device for distinguishing drivers includes:
The screening module is used for screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located;
the module is used for taking the appointed set as a running track set corresponding to the applicant;
the second judging module is used for judging whether the current dangerous running track is contained in the running track set corresponding to the applicant;
And the judging module is used for judging that the driver is appointed for the type of the risk if the driver is included in the running track set corresponding to the applicant, otherwise, the driver is not appointed for the type of the risk.
Referring to fig. 3, in an embodiment of the present application, there is further provided a computer device, which may be a server, and an internal structure thereof may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store all the data needed to distinguish the driver's procedure. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by the processor to implement a method of distinguishing drivers.
The processor executes the method for distinguishing drivers, and the method comprises the following steps: acquiring historical driving track data corresponding to a specified vehicle; matching the historical driving track data into a map grid to form a track map; forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map; inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays; and distinguishing the corresponding appointed running track when the same driver drives in each historical running track data according to the classification weight and the preset threshold value.
According to the computer equipment, the historical driving track data and the map grids are overlapped to form the track map, each piece of historical driving track data is marked according to the track map calculation feature data array, then the similarity between the feature data arrays of each piece of historical driving track data is calculated according to the twin neural network model, driving behaviors of different drivers are distinguished, the drivers are distinguished and classified, and one-to-one correspondence correlation between the driving track data and the drivers is achieved.
In one embodiment, the step of matching the historical driving track data into a map grid by the processor to form a track map includes: dividing the map into map grids according to a preset dividing mode; acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data; superposing the appointed historical driving track data into the map grid according to the longitude and latitude information; and according to the mode that the specified historical driving track data are overlapped in the map grid, all the historical driving track data are overlapped in the map grid in a one-to-one correspondence mode, and the track map is formed.
In one embodiment, the feature data array includes track feature data, and the step of forming, by the processor, feature data arrays corresponding to the historical driving track data respectively according to the track map includes: acquiring each map grid interval occupied by the appointed historical driving track data and the corresponding speed limit label of each map grid interval; counting track characteristic data corresponding to the appointed historical driving track data according to speed limit labels respectively corresponding to each map grid interval, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio; and counting the track characteristic data corresponding to all the historical driving track data respectively according to the counting mode of the track characteristic data corresponding to the appointed historical driving track data.
In one embodiment, the feature data array includes driving feature data of a driver, and the step of forming, by the processor, feature data arrays corresponding to the historical driving track data respectively according to the track map includes: acquiring time data corresponding to each track point corresponding to the appointed historical driving track data; calculating driver driving characteristic data corresponding to the appointed historical driving track data according to the time data, wherein the driver driving characteristic data comprises preset time interval driving data, driving smoothness and fatigue driving data; and calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
In one embodiment, before the step of inputting each of the feature data arrays into the twin neural network model and calculating the classification weights between the feature data arrays, the processor includes: inputting training samples into a twin neural network model, and mapping the training samples into a high-dimensional space through a specified function to obtain space vectors corresponding to the training samples respectively, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples; calculating a first space vector distance corresponding to a first sample and a second sample with the same type of labels and a second space vector distance corresponding to a third sample and a fourth sample with different types of labels according to a first calculation formula, wherein the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance), adjusting parameters of the specified function to enable the first space vector distance to be smaller and the second space vector distance to be larger, judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum, and determining that the parameters of the specified function are fixed parameters if the first space vector distance is minimum.
In one embodiment, the step of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data by the processor according to the classification weight and the preset threshold includes: dividing the historical driving track data with the classification weight greater than or equal to the preset threshold into a first set and a second set which correspond to two drivers respectively, and merging the historical driving track data with the classification weight smaller than the preset threshold into a set corresponding to the same driver; and respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
In one embodiment, the step of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data by the processor according to the classification weight and the preset threshold includes: screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located; taking the appointed set as a running track set corresponding to the applicant; judging whether the current dangerous running track is contained in a running track set corresponding to the applicant or not; if yes, judging that the type of the risk of the appointed driver is judged, otherwise, judging that the type of the risk of the appointed driver is not judged.
It will be appreciated by those skilled in the art that the architecture shown in fig. 3 is merely a block diagram of a portion of the architecture in connection with the present inventive arrangements and is not intended to limit the computer devices to which the present inventive arrangements are applicable.
An embodiment of the present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of distinguishing drivers, comprising: acquiring historical driving track data corresponding to a specified vehicle; matching the historical driving track data into a map grid to form a track map; forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map; inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays; and distinguishing the corresponding appointed running track when the same driver drives in each historical running track data according to the classification weight and the preset threshold value.
The computer readable storage medium is used for superposing the historical driving track data and the map grid to form a track map, marking each piece of historical driving track data according to a track map calculation feature data array, calculating the similarity between feature data arrays of each piece of historical driving track data according to a twin neural network model, distinguishing driving behaviors of different drivers, distinguishing and classifying the drivers, and realizing one-to-one correspondence between the driving track data and the drivers.
In one embodiment, the step of matching the historical driving track data into a map grid by the processor to form a track map includes: dividing the map into map grids according to a preset dividing mode; acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data; superposing the appointed historical driving track data into the map grid according to the longitude and latitude information; and according to the mode that the specified historical driving track data are overlapped in the map grid, all the historical driving track data are overlapped in the map grid in a one-to-one correspondence mode, and the track map is formed.
In one embodiment, the feature data array includes track feature data, and the step of forming, by the processor, feature data arrays corresponding to the historical driving track data respectively according to the track map includes: acquiring each map grid interval occupied by the appointed historical driving track data and the corresponding speed limit label of each map grid interval; counting track characteristic data corresponding to the appointed historical driving track data according to speed limit labels respectively corresponding to each map grid interval, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio; and counting the track characteristic data corresponding to all the historical driving track data respectively according to the counting mode of the track characteristic data corresponding to the appointed historical driving track data.
In one embodiment, the feature data array includes driving feature data of a driver, and the step of forming, by the processor, feature data arrays corresponding to the historical driving track data respectively according to the track map includes: acquiring time data corresponding to each track point corresponding to the appointed historical driving track data; calculating driver driving characteristic data corresponding to the appointed historical driving track data according to the time data, wherein the driver driving characteristic data comprises preset time interval driving data, driving smoothness and fatigue driving data; and calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
In one embodiment, before the step of inputting each of the feature data arrays into the twin neural network model and calculating the classification weights between the feature data arrays, the processor includes: inputting training samples into a twin neural network model, and mapping the training samples into a high-dimensional space through a specified function to obtain space vectors corresponding to the training samples respectively, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples; calculating a first space vector distance corresponding to a first sample and a second sample with the same type of labels and a second space vector distance corresponding to a third sample and a fourth sample with different types of labels according to a first calculation formula, wherein the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance), adjusting parameters of the specified function to enable the first space vector distance to be smaller and the second space vector distance to be larger, judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum, and determining that the parameters of the specified function are fixed parameters if the first space vector distance is minimum.
In one embodiment, the step of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data by the processor according to the classification weight and the preset threshold includes: dividing the historical driving track data with the classification weight greater than or equal to the preset threshold into a first set and a second set which correspond to two drivers respectively, and merging the historical driving track data with the classification weight smaller than the preset threshold into a set corresponding to the same driver; and respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
In one embodiment, the step of distinguishing the specified driving track corresponding to the same driver driving in each of the historical driving track data by the processor according to the classification weight and the preset threshold includes: screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located; taking the appointed set as a running track set corresponding to the applicant; judging whether the current dangerous running track is contained in a running track set corresponding to the applicant or not; if yes, judging that the type of the risk of the appointed driver is judged, otherwise, judging that the type of the risk of the appointed driver is not judged.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided by the present application and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that comprises the element.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the application, and all equivalent structures or equivalent processes using the descriptions and drawings of the present application or directly or indirectly applied to other related technical fields are included in the scope of the application.
Claims (7)
1. A method of distinguishing drivers, comprising:
Acquiring historical driving track data corresponding to a specified vehicle; the method comprises the steps that historical driving track data corresponding to a specified vehicle are obtained through an insurance APP account registered by a vehicle owner of the specified vehicle, and the insurance APP collects the navigation track of each insurance driver as the historical driving track data in a mode of setting rewards or insurance conditions;
matching the historical driving track data into a map grid to form a track map;
Forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
inputting each characteristic data array into a twin neural network model, and calculating the classification weight between the characteristic data arrays;
According to the classification weight and a preset threshold value, distinguishing the corresponding appointed running track when the same driver drives in each historical running track data;
After the step of distinguishing the specified driving track corresponding to the same driver in the historical driving track data according to the classification weight and the preset threshold value, the method comprises the following steps:
screening a specified set with the largest data volume in specified map grids, wherein the specified map grids are grids where the vehicle insurance places are located;
taking the appointed set as a running track set corresponding to the applicant;
judging whether the current dangerous running track is contained in a running track set corresponding to the applicant or not;
if yes, judging that the type of the risk of the appointed driver is judged, otherwise, judging that the type of the risk of the appointed driver is not judged;
if the classification weight of the characteristic data array of the current dangerous running track and the characteristic data array in the appointed set is smaller than a preset threshold value, the current dangerous running track is indicated to be contained in the running track set corresponding to the applicant, and is the driving behavior of the appointed driver registered by the insurance of the vehicle, belongs to the dangerous type of the appointed driver, or not belongs to the dangerous type of the appointed driver, and is dangerous fraud;
The step of matching the historical driving track data into a map grid to form a track map comprises the following steps:
Dividing the map into map grids according to a preset dividing mode;
acquiring longitude and latitude information corresponding to each track point in specified historical driving track data, wherein the specified historical driving track data belongs to any piece of all the historical driving track data;
Superposing the appointed historical driving track data into the map grid according to the longitude and latitude information;
according to the mode that the specified historical driving track data are overlapped into the map grid, all the historical driving track data are overlapped into the map grid in a one-to-one correspondence mode, and the track map is formed;
the step of distinguishing the corresponding appointed driving track in each historical driving track data when the same driver drives according to the classification weight and the preset threshold value comprises the following steps:
dividing the historical driving track data with the classification weight greater than or equal to the preset threshold into a first set and a second set which correspond to two drivers respectively, and merging the historical driving track data with the classification weight smaller than the preset threshold into a set corresponding to the same driver;
and respectively associating the set type and the running track data quantity corresponding to each set type in each map grid.
2. The method for distinguishing drivers according to claim 1, wherein the feature data array includes track feature data, and the step of forming feature data arrays respectively corresponding to the historical driving track data according to the track map includes:
Acquiring each map grid interval occupied by the appointed historical driving track data and the corresponding speed limit label of each map grid interval;
Counting track characteristic data corresponding to the appointed historical driving track data according to speed limit labels respectively corresponding to each map grid interval, wherein the track characteristic data comprises a slow driving interval ratio and an overspeed driving interval ratio;
and counting the track characteristic data corresponding to all the historical driving track data respectively according to the counting mode of the track characteristic data corresponding to the appointed historical driving track data.
3. The method for distinguishing a driver according to claim 1, wherein the feature data array includes driver driving feature data, and the step of forming feature data arrays respectively corresponding to the historical driving track data according to the track map includes:
acquiring time data corresponding to each track point corresponding to the appointed historical driving track data;
calculating driver driving characteristic data corresponding to the appointed historical driving track data according to the time data, wherein the driver driving characteristic data comprises preset time interval driving data, driving smoothness and fatigue driving data;
And calculating the driving characteristic data of the drivers corresponding to all the historical driving track data according to the calculation mode of the driving characteristic data of the drivers corresponding to the specified historical driving track data.
4. The method of distinguishing drivers according to claim 1, wherein before the step of inputting each of the feature data arrays into the twin neural network model and calculating the classification weights between the feature data arrays, the method comprises:
Inputting training samples into a twin neural network model, and mapping the training samples into a high-dimensional space through a specified function to obtain space vectors corresponding to the training samples respectively, wherein the specified function is Gw (X), w represents a parameter, and X represents the training samples;
Calculating a first space vector distance corresponding to a first sample and a second sample with the same category label and a second space vector distance corresponding to a third sample and a fourth sample with different category labels according to a first calculation formula, wherein the first calculation formula is Ew (X 1,X2)＝||Gw(X1)-Gw(X2)||,X1,X2 represents a training sample and Ew (X 1,X2) represents a space vector distance;
adjusting parameters of the specified function to enable the distance of the first space vector to be smaller, and enabling the distance of the second space vector to be larger;
Judging whether the second space vector distance is simultaneously maximum when the first space vector distance is minimum;
if yes, determining the parameter of the appointed function as a fixed parameter.
5. A driver differentiating apparatus for performing the driver differentiating method according to any one of claims 1 to 4, comprising:
the first acquisition module is used for acquiring historical driving track data corresponding to the appointed vehicle;
The first forming module is used for matching the historical driving track data into a map grid to form a track map;
The second forming module is used for forming characteristic data arrays corresponding to the historical driving track data respectively according to the track map;
the first calculation module is used for inputting each characteristic data array into the twin neural network model, and calculating the classification weight between the characteristic data arrays;
And the distinguishing module is used for distinguishing the corresponding appointed driving track when the same driver drives in the historical driving track data according to the classification weight and the preset threshold value.
6. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 4 when the computer program is executed.
7. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 4.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110276146.9A CN113157817B (en) | 2021-03-15 | 2021-03-15 | Method, device and computer equipment for distinguishing drivers |
| PCT/CN2021/091711 WO2022193416A1 (en) | 2021-03-15 | 2021-04-30 | Method and apparatus for distinguishing driver, and computer device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110276146.9A CN113157817B (en) | 2021-03-15 | 2021-03-15 | Method, device and computer equipment for distinguishing drivers |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113157817A CN113157817A (en) | 2021-07-23 |
| CN113157817B true CN113157817B (en) | 2024-05-10 |
Family
ID=76887126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110276146.9A Active CN113157817B (en) | 2021-03-15 | 2021-03-15 | Method, device and computer equipment for distinguishing drivers |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113157817B (en) |
| WO (1) | WO2022193416A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022142075A (en) * | 2021-03-16 | 2022-09-30 | サトーホールディングス株式会社 | Display system, control apparatus, and control program |
| CN114860859A (en) * | 2022-04-14 | 2022-08-05 | 郑州天迈科技股份有限公司 | Bus track similarity calculation method based on twin neural network |
| CN117421572B (en) * | 2023-10-30 | 2024-08-13 | 北京九栖科技有限责任公司 | Network taxi driver number identification method and system based on communication data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106314438A (en) * | 2016-08-15 | 2017-01-11 | 西北工业大学 | Method and system for detecting abnormal track in driver driving track |
| CN110443185A (en) * | 2019-07-31 | 2019-11-12 | 京东城市(北京)数字科技有限公司 | Driver's recognition methods, driver identification device, electronic equipment and storage medium |
| CN110889444A (en) * | 2019-11-22 | 2020-03-17 | 南京邮电大学 | Driving track feature classification method based on convolutional neural network |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9898759B2 (en) * | 2014-03-28 | 2018-02-20 | Joseph Khoury | Methods and systems for collecting driving information and classifying drivers and self-driving systems |
| US11244254B2 (en) * | 2014-12-31 | 2022-02-08 | The City And County Of San Francisco | Application-based commercial ground transportation clearinghouse system |
| DE102016210029A1 (en) * | 2016-06-07 | 2017-12-07 | Robert Bosch Gmbh | Method Device and system for wrong driver identification |
| CN107784046B (en) * | 2016-11-14 | 2021-05-04 | 平安科技（深圳）有限公司 | POI information processing method and device |
| US11106969B2 (en) * | 2017-01-19 | 2021-08-31 | International Business Machines Corporation | Method and apparatus for driver identification leveraging telematics data |
| EP3679552B1 (en) * | 2017-09-06 | 2024-11-06 | Swiss Reinsurance Company Ltd. | Electronic logging and track identification system for mobile telematics devices, and corresponding method thereof |
| CN109145982A (en) * | 2018-08-17 | 2019-01-04 | 上海汽车集团股份有限公司 | The personal identification method and device of driver, storage medium, terminal |
| US11577734B2 (en) * | 2018-12-20 | 2023-02-14 | Nauto, Inc. | System and method for analysis of driver behavior |
| US10820166B1 (en) * | 2019-12-04 | 2020-10-27 | Toyota Connected North America, Inc. | Systems and methods for obtaining location intelligence |
| CN112417273B (en) * | 2020-11-17 | 2022-04-19 | 平安科技（深圳）有限公司 | Region portrait image generation method, region portrait image generation device, computer equipment and storage medium |
-
2021
- 2021-03-15 CN CN202110276146.9A patent/CN113157817B/en active Active
- 2021-04-30 WO PCT/CN2021/091711 patent/WO2022193416A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106314438A (en) * | 2016-08-15 | 2017-01-11 | 西北工业大学 | Method and system for detecting abnormal track in driver driving track |
| CN110443185A (en) * | 2019-07-31 | 2019-11-12 | 京东城市(北京)数字科技有限公司 | Driver's recognition methods, driver identification device, electronic equipment and storage medium |
| CN110889444A (en) * | 2019-11-22 | 2020-03-17 | 南京邮电大学 | Driving track feature classification method based on convolutional neural network |
Non-Patent Citations (2)
| Title |
|---|
| Driver Identification and Verification From Smartphone Accelerometers Using Deep Neural Networks;Sara Hernández Sánchez et al;《IEEE》;第97-108页 * |
| Galina Malykhina et al.Application of Siamese Neural Networks for the Type of Emergency Determination.《CSIS'2019: Proceedings of the XI International Scientific Conference Communicative Strategies of the Information Society》.2019,第1-5页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113157817A (en) | 2021-07-23 |
| WO2022193416A1 (en) | 2022-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113157817B (en) | Method, device and computer equipment for distinguishing drivers | |
| US11407410B2 (en) | Method and system for estimating an accident risk of an autonomous vehicle | |
| CN110268454B (en) | Determining a customized safe speed for a vehicle | |
| Jiang et al. | Traffic and vehicle speed prediction with neural network and hidden markov model in vehicular networks | |
| CN108022012A (en) | Vehicle location Forecasting Methodology based on deep learning | |
| CN107784587A (en) | A kind of driving behavior evaluation system | |
| CN107784251A (en) | The method evaluated based on image recognition technology driving behavior | |
| CN108877213B (en) | Traffic accident identification method and system | |
| CN114493191B (en) | Driving behavior modeling analysis method based on network about vehicle data | |
| Wang et al. | Vehicle reidentification with self-adaptive time windows for real-time travel time estimation | |
| CN117238126A (en) | Traffic accident risk assessment method under continuous flow road scene | |
| Siddique et al. | State-dependent self-adaptive sampling (SAS) method for vehicle trajectory data | |
| Jia et al. | A microscopic model of vehicle co₂ emissions based on deep learning—A spatiotemporal analysis of taxicabs in Wuhan, China | |
| CN108268678A (en) | Driving behavior analysis method, apparatus and system | |
| CN113723529B (en) | Traffic information credible identification method based on speed prediction algorithm | |
| Zong et al. | Understanding taxi drivers’ multi-day cruising patterns | |
| Mortabit et al. | Instantaneous Vehicle Fuel Consumption Estimation Using Neural Networks | |
| CN118334863B (en) | Vehicle scheduling method, device and equipment | |
| Milardo et al. | An unsupervised approach for driving behavior analysis of professional truck drivers | |
| CN116665342B (en) | New energy automobile driving behavior analysis method, system and equipment | |
| US20240355152A1 (en) | Method and system for analyzing and predicting vehicle stay behavior based on multi-task learning | |
| Morshed | Enhanced Methods for Utilization of Data to Support Multi-Scenario Analysis and Multi-Resolution Modeling | |
| CN114565049A (en) | Method and device for estimating instantaneous energy consumption of electric automobile | |
| CN118629210A (en) | Urban road travel time prediction method, system, equipment and medium | |
| Seraj | The Classification of Short and Long-term Driving Behavior for an Advanced Driver Assistance System by Analyzing Bidirectional Driving Features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |