CN118015843B - Vehicle detection method and system based on pressure sensor time sequence data - Google Patents
Vehicle detection method and system based on pressure sensor time sequence data Download PDFInfo
- Publication number
- CN118015843B CN118015843B CN202410424607.6A CN202410424607A CN118015843B CN 118015843 B CN118015843 B CN 118015843B CN 202410424607 A CN202410424607 A CN 202410424607A CN 118015843 B CN118015843 B CN 118015843B
- Authority
- CN
- China
- Prior art keywords
- list
- peak
- time
- moment
- sub
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- 238000001914 filtration Methods 0.000 claims abstract description 19
- 238000005070 sampling Methods 0.000 claims abstract description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 2
- 238000005286 illumination Methods 0.000 abstract description 3
- 238000005303 weighing Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 2
- 238000000540 analysis of variance Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Landscapes
- Measuring Fluid Pressure (AREA)
Abstract
The invention discloses a vehicle detection method and a system based on pressure sensor time sequence data, which comprises the steps of firstly carrying out smooth filtering on original time sequence data, traversing data at each moment to carry out variance analysis, detecting abrupt peak according to variance change conditions, and further determining actual peak moment according to absolute values to obtain a moment list; then dividing the continuous peak values according to vehicles, and merging peak values generated by the same vehicle to obtain a series of sub-lists corresponding to the vehicles; correcting each sub-list according to the actual service condition to obtain a final peak value moment list; and finally, for each sub-list, selecting a first element as a starting point of vehicle passing, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data. The accuracy of the detection method is not influenced by natural environments such as weather, illumination and the like, and is also not influenced by the passing speed and the flow of vehicles.
Description
Technical Field
The invention relates to a vehicle detection method and system based on time sequence data of a pressure sensor, and belongs to the field of Internet of vehicles.
Background
In the technology of internet of vehicles, the technology of V2X (Vehicle to Everything) relies on the basic environment data of roads to make intelligent decisions. In road base environment data, the passing data of vehicles is the basis of all decisions of intelligent traffic. The existing road vehicle data detection methods include the following: 1. by deploying the weighing system at a specific location of the lane, the data changes of the weighing system are utilized to detect the passing information of the vehicle. 2. The video acquisition equipment is deployed at the key position of the road, and the video stream is analyzed by using an image detection algorithm to obtain the vehicle information of the position of the road section.
The existing vehicle detection has the following defects: 1. the detection method based on the weighing system comprises the following steps: the scale table type and bent plate type weighing system has low precision under the high-speed condition, and is only suitable for detecting vehicles passing at low speed; while the piezoelectric weighing system can detect the measurement passing through at high speed, the overall accuracy is low and the statistical error is large. 2. Traffic flow statistics based on video is greatly affected by weather. Under the conditions of bad weather, poor illumination and the like, the detection error is large, and even the condition of undetectable occurs.
Disclosure of Invention
The invention aims to: in view of the above problems in the prior art, an object of the present invention is to provide a method and a system for detecting vehicles based on pressure sensor time series data, which analyze sensor time series data by using a pressure sensor preset in a road to detect passing information of vehicles.
The technical scheme is as follows: in order to achieve the aim of the invention, the invention adopts the following technical scheme:
a vehicle detection method based on pressure sensor time sequence data comprises the following steps:
smoothing and filtering the original time sequence data;
Traversing the filtered data step by step, respectively calculating variances of the first X time points and the last X time points of the subsequent 2X time points for the data of each moment, and detecting abrupt peaks according to variance change conditions to obtain a peak moment list A to be confirmed;
Traversing each moment in the moment list A, selecting the moment corresponding to the data with the largest absolute value in the front and rear preset ranges in the original time sequence data, and obtaining an actual peak moment list B;
merging the continuous peaks in the first threshold value in the time list B into peaks of the same waveform to divide the peaks generated by different vehicles to obtain sub-lists corresponding to the vehicles, merging the continuous peaks in the second threshold value in each sub-list, and selecting the point with the largest absolute value as the point of the merged peak to merge the peaks generated by the same vehicle to obtain a time list C;
correcting the time list C, removing sub-lists of which the peak point number and the numerical value do not meet the actual service, and obtaining a time list E;
And for each sub-list in E, selecting a first element as a starting point of vehicle passing, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data.
Preferably, if l= (V i+x-Vi)/Vi is greater than a preset threshold, then adding the i time to the time list a to be confirmed, where V i is the variance of X time points after the i time and V i+x is the variance of X time points after the i+x time.
Preferably, the method for correcting the time list C is as follows:
filtering out a sub-list with only one peak value point in C to obtain a list C1;
Filtering out sub-lists with peak points larger than six in C1 to obtain a list C2;
Traversing the sensor value corresponding to the peak time for each sub-list in C2, and removing the peak time to obtain a list C3 if the sensor value is larger than the reference value of the sensor under the action of no external force;
and filtering out the sub-list with only one peak point in C3 to obtain a list E.
Preferably, the reference value is determined according to the following method: for each sub-list in C2, selecting data in a preset time from a first peak point, calculating an average value, and taking the average value as a reference value of the sensor under the action of no external force; and no vehicle passes through the first peak point in the preset time.
Further, the vehicle type is determined based on the number of peaks in each sub-list in list E.
Based on the same inventive concept, the present invention provides a vehicle detection system based on pressure sensor time series data, comprising:
The abrupt peak detection module is used for carrying out smooth filtering on the original time sequence data; traversing the filtered data step by step, respectively calculating variances of the first X time points and the last X time points of the subsequent 2X time points for the data of each moment, and detecting abrupt peaks according to variance change conditions to obtain a peak moment list A to be confirmed;
The actual peak detection module is used for traversing each moment in the moment list A, selecting the moment corresponding to the data with the largest absolute value in the front and rear preset ranges in the original time sequence data, and obtaining an actual peak moment list B;
The waveform segmentation module is used for merging the continuous peaks in the first threshold value in the moment list B into the peaks of the same waveform so as to segment the peaks generated by different vehicles to obtain sub-lists corresponding to the vehicles, merging the continuous peaks in the second threshold value in each sub-list, and selecting the point with the largest absolute value as the point of the merged peak so as to merge the peaks generated by the same vehicle to obtain a moment list C;
the correction module is used for correcting the time list C, removing the sub-list of which the peak point number and the numerical value do not meet the actual service, and obtaining a time list E;
and the vehicle detection module is used for selecting a first element as a starting point of vehicle passing for each sub-list in the E, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data.
Based on the same inventive concept, the invention provides a computer system comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the computer program is loaded to the processor to realize the steps of the vehicle detection method based on the time sequence data of the pressure sensor.
Based on the same inventive concept, the present invention provides a computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the vehicle detection method based on pressure sensor time series data.
The beneficial effects are that: compared with the prior art, the invention has the following advantages: 1. the invention realizes an analysis of variance method for sensor time sequence data, and can detect the passing information of a vehicle by analyzing the change of sensor pressure data. 2. The invention realizes the specific waveform detection of the time sequence pressure data, and further analyzes and generalizes the vehicle type through the waveform, thereby improving the detection accuracy; 3. the accuracy of the detection method is not influenced by natural environments such as weather, illumination and the like, and is also not influenced by the passing speed and the flow of vehicles.
Drawings
FIG. 1 is a flow chart of a method of an embodiment of the present invention.
FIG. 2 is a flow chart of peak data correction in an embodiment of the invention.
FIG. 3 is a schematic diagram of a system module according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments.
The embodiment of the invention discloses a vehicle detection method based on time sequence data of a pressure sensor, which utilizes a pressure sensor (such as DTC-DP 200) preset in a road, and the vehicle can generate downward pressure to the sensor when passing through the road surface, so that abrupt change of the data of the sensor is caused. The number of abrupt peaks generated by the sensor is detected by analyzing the sensor timing data. According to the abrupt change data, the passing information of the vehicle can be detected; the number of peaks that are abrupt represents the difference in the type of vehicle that passes. Thus enabling determination of vehicle throughput, as well as vehicle type.
As shown in fig. 1, the specific service execution flow includes:
Step 1, performing smooth filtering on original time sequence data F to obtain filtered data D= { [ T 1,D1],[T2,D2 ] and … }; filtering can reduce the effect of sensor noise on the detection result.
And step 2, detecting abrupt peak values of the filtered data. For the filtered data D, all data is traversed step by step in a step 1 fashion. For data D i at a certain time i, the following 2X time points (X may be selected in combination with the sensor sampling rate and the vehicle speed), where the time range corresponding to the X time points is smaller than the time range from the reference value to the peak value of the waveform, for example, considering that the vehicle speed is generally smaller than 100km/h, the sampling rate is 500hz, and X may take 10), the variances V i and V i+x.Vi of the data are calculated using the previous X data, and V i+x is calculated using the following X data. The calculation method of V i is as follows:
For the above calculated V i and V i+x, l= (V i+x-Vi)/Vi) is calculated, if L is greater than a specified threshold, adding the instant i to the instant list a to be confirmed, where each data in a represents that there may be a vehicle passing near the instant, thus causing a severe change in the data change of the sensor.
And step 3, searching an actual peak value in the original data according to the abrupt peak value. And traversing all data subsets (such as Y=10) of which the position index of each index is within a range [ index-Y, index+Y ] in the data set D aiming at the time list A= { index1, index2, index … }, and selecting the index corresponding to the data with the largest absolute value in the subsets as an actual peak time list B.
And 4, dividing peaks generated by different vehicles. And (3) merging continuous peaks within a threshold T seconds into peaks of the same waveform according to the sampling frequency and the actual service requirement aiming at the data analyzed in the step (B) so as to divide peaks generated by different vehicles and obtain a sub-list corresponding to the vehicles. The waveform peak time list after segmentation is expressed as C, C= { { { C11, C12,..C1n }, { C21, C22, … C2m }, …, { Co1, co2, …, cop }, each sub-list corresponds to the peak time when one vehicle passes through, and n, m and p are the peak point numbers of the 1 st, 2 nd and o th vehicles respectively. Let b= {1000,1010,1090,7000, 7070}, threshold t=2 seconds, sample rate sample=1000. Then the peaks within T sample=2×1000, i.e. 2000, of the index differences are considered to belong to the same waveform. For example B, 2 actual waveforms, c= { {1000,1010,1090}, {7000, 7070}, can be resolved.
And 5, merging peaks passing by the same vehicle. And combining the continuous peaks within the threshold T1 seconds according to the sampling frequency and the actual service requirement. Assuming that the sampling rate sample=is 1000, the threshold t1=0.05 seconds (the value of T1 corresponds to the time interval of different wheels passing through the sensor, so the speed is not too fast according to the road speed limit in real life, for example, only 100km/h is allowed, and a reasonable value of T1 can be estimated). For peak intervals less than T1 sample=0.05×1000, i.e. all peaks within 50 are considered to be the same peak. For example c= { {1000,1010,1090}, {7000, 7070}, in step 4, in waveform {1000,1010,1090}, the spacing between 1000 and 1010 is less than 50, combined into the same peak. Comparing the absolute values abs of D1000 and D1010, and selecting the point with the large absolute value as the peak point after combination. Let abs (D1000) > abs (D1010), combined c= { {1000,1090}, {7000, 7070 }.
And 6, correcting the detected vehicle peak value data. And correcting the merged time list C according to the actual service. As shown in fig. 2, the method specifically includes:
Step 6.1, when each vehicle passes through the sensor, the front and rear wheels generate at least two waveforms, so at least 2 peaks are corresponding. Filtering out a sub-list with only one peak point in C, and recording the obtained data as C1;
Step 6.2, the actual vehicle has at most 6 axes, so for the waveform, there are at most 6 peak data. Filtering out sub-lists with peak points larger than 6 points in C1, and recording the obtained data as C2;
and 6.3, selecting data in a preset time T3 (for example, 3 seconds, the probability that two vehicles pass through the same point in the T3 time is considered to be 0) before the first peak point of each waveform of the waveform data in the C2, and marking the data as Dbase. For all data in Dbase, the average value X thereof was calculated, X being the reference value of the sensor without being subjected to external force. Each peak generated by the sensor should be smaller than its corresponding reference value X. For each waveform (sub-list) in C2, the sensor value Di corresponding to peak time i is traversed, and if Di > X, the peak time is removed from the waveform. The obtained data is marked as C3;
Step 6.4, traversing each sub-list in C3, and removing a sub-list if the sub-list has only one peak value. The resulting list is denoted as E.
And 7, calculating the vehicle passing time. Each sub-list in list E represents the passing information of the vehicle. For each vehicle Ei, a first point Ei1 in ei= { Ei1, ei2, … Eiq } is selected as a starting point of vehicle passing, and according to the time offset Foffset and the sampling rate sample of the original time sequence data F, the actual time Tpass of vehicle passing is calculated, where tpass=ei 1/sample+foffset. Assuming that the time offset of the data is foffset=2023/11/23 09:09:10 and ei1=12000 in a certain detection, tpass=2023/11/23/09:09:10+ (12000/1000) =2023/11/23/09:09:22.
And 8, judging the type of the passing vehicle according to the peak value number. Each sub-list in list E is traversed with a different number of peaks representing vehicle types of different wheelbases. The type of vehicle passing can be determined based on the number of peaks, e.g., two peaks for two-axis vehicle passing and three peaks for three-axis vehicle passing.
Based on the same inventive concept, the vehicle detection system based on time series data of the pressure sensor disclosed in the embodiment of the invention, as shown in fig. 3, includes: the abrupt peak detection module is used for carrying out smooth filtering on the original time sequence data; traversing the filtered data step by step, respectively calculating variances of the first X time points and the last X time points of the subsequent 2X time points for the data of each moment, and detecting abrupt peaks according to variance change conditions to obtain a peak moment list A to be confirmed; the actual peak detection module is used for traversing each moment in the moment list A, selecting the moment corresponding to the data with the largest absolute value in the front and rear preset ranges in the original time sequence data, and obtaining an actual peak moment list B; the waveform segmentation module is used for merging the continuous peaks in the first threshold value in the moment list B into the peaks of the same waveform so as to segment the peaks generated by different vehicles to obtain sub-lists corresponding to the vehicles, merging the continuous peaks in the second threshold value in each sub-list, and selecting the point with the largest absolute value as the point of the merged peak so as to merge the peaks generated by the same vehicle to obtain a moment list C; the correction module is used for correcting the time list C, removing the sub-list of which the peak point number and the numerical value do not meet the actual service, and obtaining a time list E; and the vehicle detection module is used for selecting a first element as a starting point of vehicle passing for each sub-list in the E, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data.
Based on the same inventive concept, the embodiment of the invention discloses a computer system, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program is loaded to the processor to realize the steps of the vehicle detection method based on the time sequence data of the pressure sensor.
Based on the same inventive concept, an embodiment of the present invention discloses a computer readable storage medium storing a computer program, which when executed by a processor, implements the steps of the vehicle detection method based on pressure sensor time series data.
Claims (6)
1. The vehicle detection method based on the time sequence data of the pressure sensor is characterized by comprising the following steps of:
smoothing and filtering the original time sequence data;
traversing the filtered data step by step, respectively calculating variances of the first X time points and the last X time points of the subsequent 2X time points for the data of each moment, and detecting abrupt peaks according to variance change conditions to obtain a peak moment list A to be confirmed; if l= (V i+x-Vi)/Vi is greater than a preset threshold, adding the i moment to the moment list a to be confirmed, wherein V i is the variance of X time points after the i moment, and V i+x is the variance of X time points after the i+x moment;
Traversing each moment in the moment list A, selecting the moment corresponding to the data with the largest absolute value in the front and rear preset ranges in the original time sequence data, and obtaining an actual peak moment list B;
merging the continuous peaks in the first threshold value in the time list B into peaks of the same waveform to divide the peaks generated by different vehicles to obtain sub-lists corresponding to the vehicles, merging the continuous peaks in the second threshold value in each sub-list, and selecting the point with the largest absolute value as the point of the merged peak to merge the peaks generated by the same vehicle to obtain a time list C;
Correcting the time list C, removing sub-lists of which the peak point number and the numerical value do not meet the actual service, and obtaining a time list E; comprising the following steps: filtering out a sub-list with only one peak value point in C to obtain a list C1; filtering out sub-lists with peak points larger than six in C1 to obtain a list C2; traversing the sensor value corresponding to the peak time for each sub-list in C2, and removing the peak time to obtain a list C3 if the sensor value is larger than the reference value of the sensor under the action of no external force; filtering out a sub-list with only one peak value point in C3 to obtain a list E;
And for each sub-list in E, selecting a first element as a starting point of vehicle passing, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data.
2. The vehicle detection method based on pressure sensor time series data according to claim 1, wherein the reference value is determined according to the following method: for each sub-list in C2, selecting data in a preset time from a first peak point, calculating an average value, and taking the average value as a reference value of the sensor under the action of no external force; and no vehicle passes through the first peak point in the preset time.
3. The method for detecting a vehicle based on time series data of a pressure sensor according to claim 1, wherein the type of the vehicle is judged based on the number of peaks in each sub-list in the list E.
4. A vehicle detection system based on pressure sensor timing data, comprising:
The abrupt peak detection module is used for carrying out smooth filtering on the original time sequence data; traversing the filtered data step by step, respectively calculating variances of the first X time points and the last X time points of the subsequent 2X time points for the data of each moment, and detecting abrupt peaks according to variance change conditions to obtain a peak moment list A to be confirmed; if l= (V i+x-Vi)/Vi is greater than a preset threshold, adding the i moment to the moment list a to be confirmed, wherein V i is the variance of X time points after the i moment, and V i+x is the variance of X time points after the i+x moment;
The actual peak detection module is used for traversing each moment in the moment list A, selecting the moment corresponding to the data with the largest absolute value in the front and rear preset ranges in the original time sequence data, and obtaining an actual peak moment list B;
The waveform segmentation module is used for merging the continuous peaks in the first threshold value in the moment list B into the peaks of the same waveform so as to segment the peaks generated by different vehicles to obtain sub-lists corresponding to the vehicles, merging the continuous peaks in the second threshold value in each sub-list, and selecting the point with the largest absolute value as the point of the merged peak so as to merge the peaks generated by the same vehicle to obtain a moment list C;
The correction module is used for correcting the time list C, removing the sub-list of which the peak point number and the numerical value do not meet the actual service, and obtaining a time list E; comprising the following steps: filtering out a sub-list with only one peak value point in C to obtain a list C1; filtering out sub-lists with peak points larger than six in C1 to obtain a list C2; traversing the sensor value corresponding to the peak time for each sub-list in C2, and removing the peak time to obtain a list C3 if the sensor value is larger than the reference value of the sensor under the action of no external force; filtering out a sub-list with only one peak value point in C3 to obtain a list E;
and the vehicle detection module is used for selecting a first element as a starting point of vehicle passing for each sub-list in the E, and calculating the actual passing time according to the time offset and the sampling rate of the original time sequence data.
5. A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when loaded to the processor realizes the steps of the method for vehicle detection based on pressure sensor time series data according to any one of claims 1-3.
6. A computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the vehicle detection method based on pressure sensor time series data according to any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410424607.6A CN118015843B (en) | 2024-04-10 | 2024-04-10 | Vehicle detection method and system based on pressure sensor time sequence data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410424607.6A CN118015843B (en) | 2024-04-10 | 2024-04-10 | Vehicle detection method and system based on pressure sensor time sequence data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118015843A CN118015843A (en) | 2024-05-10 |
| CN118015843B true CN118015843B (en) | 2024-06-28 |
Family
ID=90950406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410424607.6A Active CN118015843B (en) | 2024-04-10 | 2024-04-10 | Vehicle detection method and system based on pressure sensor time sequence data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118015843B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101377524A (en) * | 2007-08-30 | 2009-03-04 | 北京佳讯飞鸿电气股份有限公司 | Vehicle speed measuring method based on steel rail deformation / stress parameters |
| CN101794515A (en) * | 2010-03-29 | 2010-08-04 | 河海大学 | Target detection system and method based on covariance and binary-tree support vector machine |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103366562B (en) * | 2012-09-12 | 2016-04-06 | 北京纳米能源与系统研究所 | traffic monitoring sensor and detection method |
| JP2017122986A (en) * | 2016-01-05 | 2017-07-13 | 日本電信電話株式会社 | Location identity determination system and control method thereof |
| EP3652721A1 (en) * | 2017-09-04 | 2020-05-20 | NNG Software Developing and Commercial LLC | A method and apparatus for collecting and using sensor data from a vehicle |
| CN117238170A (en) * | 2023-09-05 | 2023-12-15 | 中国联合网络通信集团有限公司 | Vehicle monitoring method, device and storage medium |
-
2024
- 2024-04-10 CN CN202410424607.6A patent/CN118015843B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101377524A (en) * | 2007-08-30 | 2009-03-04 | 北京佳讯飞鸿电气股份有限公司 | Vehicle speed measuring method based on steel rail deformation / stress parameters |
| CN101794515A (en) * | 2010-03-29 | 2010-08-04 | 河海大学 | Target detection system and method based on covariance and binary-tree support vector machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118015843A (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111985494B (en) | Road disease detection and classification method based on lightweight vehicle-mounted terminal | |
| CN110991272B (en) | Multi-target vehicle track recognition method based on video tracking | |
| CN109050535B (en) | Rapid terrain condition identification method based on vehicle attitude | |
| CN114879695B (en) | Track matching method, device, equipment and medium | |
| US20120166080A1 (en) | Method, system and computer-readable medium for reconstructing moving path of vehicle | |
| CN112433203A (en) | Lane alignment detection method based on millimeter wave radar data | |
| JP7051424B2 (en) | Anomaly detection device | |
| CN110569855A (en) | Long-time target tracking algorithm based on correlation filtering and feature point matching fusion | |
| US6711508B2 (en) | Method and apparatus for estimating tire air pressure | |
| CN112710273A (en) | Crowd-sourced road surface pothole detection method based on smart phone sensor and machine learning | |
| CN111599182B (en) | Automatic lane changing behavior extraction method based on natural driving data | |
| CN117173913B (en) | Traffic control method and system based on traffic flow analysis at different time periods | |
| CN113221759A (en) | Road scattering identification method and device based on anomaly detection model | |
| CN111707476B (en) | Longitudinal driving capability detection method for automatic driving automobile | |
| CN116206451A (en) | Intelligent traffic flow data analysis method | |
| CN118015843B (en) | Vehicle detection method and system based on pressure sensor time sequence data | |
| CN117809458A (en) | Real-time assessment method and system for traffic accident risk | |
| CN116805409A (en) | Method for identifying road surface state and evaluating flatness by using driving video | |
| CN115909241A (en) | Lane line detection method, system, electronic device and storage medium | |
| CN114169247A (en) | Method, device and equipment for generating simulated traffic flow and computer readable storage medium | |
| CN107038711A (en) | Adaptive evolution type vehicle lamp signal detection tracking and identification system and method | |
| CN113044046B (en) | Driver style recognition method, system, electronic device and readable storage medium | |
| CN116091437B (en) | Axle number detection method based on 3D point cloud | |
| CN114333318B (en) | Motorcycle collision detection method based on sensor space angle | |
| CN115271001B (en) | Vehicle driving condition identification method and device, vehicle and storage medium |
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 |