CN117975382B - Violation detection traffic control method and system based on deep learning - Google Patents
Violation detection traffic control method and system based on deep learning Download PDFInfo
- Publication number
- CN117975382B CN117975382B CN202410386339.3A CN202410386339A CN117975382B CN 117975382 B CN117975382 B CN 117975382B CN 202410386339 A CN202410386339 A CN 202410386339A CN 117975382 B CN117975382 B CN 117975382B
- Authority
- CN
- China
- Prior art keywords
- function
- intelligent auxiliary
- intelligent
- vehicle
- violation detection
- 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 132
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000013135 deep learning Methods 0.000 title claims abstract description 28
- 230000006870 function Effects 0.000 claims abstract description 384
- 230000010354 integration Effects 0.000 claims abstract description 25
- 238000001914 filtration Methods 0.000 claims abstract description 14
- 230000004913 activation Effects 0.000 claims description 86
- 238000010276 construction Methods 0.000 claims description 66
- 230000003213 activating effect Effects 0.000 claims description 45
- 230000008859 change Effects 0.000 claims description 39
- 238000004364 calculation method Methods 0.000 claims description 33
- 238000002360 preparation method Methods 0.000 claims description 25
- 230000003993 interaction Effects 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 abstract 1
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 229910052717 sulfur Inorganic materials 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a violation detection traffic control method and system based on deep learning. The invention relates to the technical field of traffic control, in particular to a traffic control method and a traffic control system for violation detection based on deep learning, which construct seven intelligent driving assistance functions including violation detection, and improve the functional integration level and the overall usability of traffic control; the method of combining deep learning with Kalman filtering is adopted to detect violations, so that the performance of the violations detection function and the effect of processing complex running environments are improved; the function classification and the priority are set, so that functions are integrated, and the function cooperation effectiveness and the overall usability of intelligent auxiliary control are improved.
Description
Technical Field
The invention relates to the technical field of traffic control, in particular to a traffic control method and system for detecting violations based on deep learning.
Background
The method is mainly used for improving the efficiency and the safety of traffic management, reducing the influence of traffic illegal behaviors on traffic order and public safety, improving the intelligent level of traffic management, realizing the rapid and accurate identification and treatment of traffic illegal behaviors, maintaining the traffic order, improving the road safety and effectively reducing the occurrence rate of traffic accidents.
However, in the existing traffic control method for detecting the violations, the detection of the violations is often independent of the traffic control auxiliary system, mainly exists as a single functional module, and cannot be effectively combined with other intelligent driving auxiliary functions; in the existing traffic control method for detecting violations, the existing method for detecting violations exists, and the performance of the existing method for detecting violations is poor when a classical Kalman filtering method is used for processing high-dimensional data and a nonlinear system, so that the accuracy of detecting violations is required to be improved; in the existing traffic control method for detecting violations, the technical problem that the coordination among functions is to be improved under the condition that multiple functions exist in the existing traffic control method exists.
Disclosure of Invention
Aiming at the technical problems that in the existing traffic control method for detecting violations, the violations are often independent of a traffic control auxiliary system and mainly exist as independent functional modules and cannot be effectively combined with other intelligent driving auxiliary functions, seven intelligent driving auxiliary functions including violations are creatively constructed, and intelligent auxiliary control and intelligent information reminding are performed by setting independent activating conditions for each function, so that the functional integration level and the overall usability of traffic control are improved; aiming at the technical problems that in the existing traffic control method for detecting violations, the existing method for detecting violations exists, and the performance is poor when a classical Kalman filtering method is used for processing high-dimensional data and a nonlinear system, so that the accuracy of detecting violations is to be improved; aiming at the technical problem that the coordination among functions is to be improved under the condition that multiple functions exist in the existing traffic control method for detecting violations, the scheme creatively sets the function classification and the priority, integrates the functions and improves the function cooperation effectiveness and the overall usability of intelligent auxiliary control.
The technical scheme adopted by the invention is as follows: the invention provides a violation detection traffic control method based on deep learning, which comprises the following steps:
step S1: constructing an intelligent auxiliary control coordinate system of the vehicle;
step S2: intelligent auxiliary function selection;
step S3: constructing intelligent auxiliary functions;
Step S4: activating an intelligent auxiliary function;
step S5: setting the priority of the intelligent auxiliary function;
Step S6: intelligent auxiliary function integration;
step S7: and detecting the vehicle violation and controlling intelligent traffic.
Further, in step S1, the vehicle intelligent auxiliary control coordinate system is constructed, and is used for constructing a vehicle scene coordinate system required by vehicle intelligent auxiliary control, specifically, a vehicle intelligent interaction coordinate system is constructed;
The intelligent interaction coordinate system of the vehicle specifically comprises an intelligent auxiliary control vehicle body and a surrounding vehicle entity;
The intelligent auxiliary control vehicle body and surrounding vehicle entities share a motion state through a computer vision technology, wherein the motion state specifically comprises a speed, an acceleration, a course angle and a geometric center.
Further, in step S2, the intelligent auxiliary function selection is used for defining an available intelligent auxiliary function, selecting a specific traffic control item for detecting violations from the intelligent auxiliary functions, specifically defining a traffic control preparation option for detecting violations, performing intelligent auxiliary function selection, and performing subsequent intelligent auxiliary function construction, activation, priority setting, integration and traffic control for detecting violations according to the traffic control preparation option for detecting violations;
the traffic control preparation and selection function items for the violation detection specifically comprise a collision warning function, a lane change warning function, a curve overspeed warning function, an emergency notification function, a violation detection warning function, a traffic guiding function and a real-time information reminding function.
Further, in step S3, the intelligent auxiliary function construction is configured to perform intelligent auxiliary function basic construction according to the function item selected by the intelligent auxiliary function, specifically perform intelligent auxiliary function construction according to the traffic control preparation option according to the violation detection, and obtain an intelligent auxiliary function item set, and includes the following steps:
step S31: the collision warning function is constructed, specifically, a collision event index set is constructed, the collision event index set is used for judging whether collision occurs or not, the collision warning function comprises a rear-end collision index, a forward collision index and a side collision index, and the collision event index set is constructed, and the method comprises the following steps:
Step S311: the rear-end collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC y1 is a rear-end collision index for determining a condition for the occurrence of a rear-end collision, y O is a projection of a geometric center O n of the intelligent auxiliary control vehicle body n to a vertical coordinate axis y, l s is a vehicle length parameter of a surrounding vehicle entity s, w s is a vehicle width parameter of the surrounding vehicle entity s, Is the included angle parameter of the intelligent auxiliary control vehicle body n and the surrounding vehicle entities s, l n is the vehicle length parameter of the intelligent auxiliary control vehicle body n,/>Is a vehicle speed vector parameter of surrounding vehicle entities s,/>Is a vehicle speed vector parameter for intelligently assisting in controlling the vehicle body n, and I/I is a modulo operator;
step S312: the side collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC j is a side collision index for determining a condition for occurrence of a side collision, x O is a projection of a geometric center O n of the intelligent auxiliary control vehicle body n onto a horizontal coordinate axis x, and w n is a vehicle width parameter of the intelligent auxiliary control vehicle body n;
Step S313: the forward collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC y2 is a forward collision index for judging conditions for rear-end collision;
Step S32: the construction of the lane change warning function is specifically to construct a lane change event index which is used as a condition for judging the potential risk of the lane change, and the calculation formula of the lane change event index is as follows:
；
wherein TTC y3 is a lane change event index;
Step S33: the construction of the curve overspeed warning function, in particular to the construction of curve overspeed event indexes, which are used as conditions for judging the potential risk of curve overspeed, wherein the calculation formula for constructing the curve overspeed event indexes is as follows:
；
Where v cst is the minimum between the critical side slip velocity and the critical side turn velocity on the curved road, used as a condition for discriminating the potential risk of curve overspeed, Is the critical sideslip velocity on curved roads,/>Is the critical d rollover speed on curved roads, |·| is the modulo operator;
Step S34: the construction of the emergency notification function, specifically, collecting road facilities and traffic environment data through a satellite navigation system, and carrying out the construction of the emergency notification function and the special event notification function;
Step S35: the construction of the violation detection warning function, specifically, the construction of a computer vision violation detection model by adopting a deep learning method, and the violation detection by the violation detection model, comprises the following steps:
step S351: the method comprises the steps of data acquisition, namely acquiring violation detection original data from crossroad camera image records;
step S352: the data preprocessing is specifically to perform image denoising and contrast enhancement operation on the violation detection original data to obtain violation detection optimized data;
Step S353: deep learning target detection and feature extraction, specifically, performing target detection on the violation detection optimization data by adopting a YOLOv model to obtain vehicle target data, and performing feature extraction on the violation detection optimization data by constructing a convolutional neural network to obtain violation feature data;
step S354: the Kalman filtering vehicle tracking is specifically carried out according to the vehicle target data and the violation feature data, specifically, the vehicle position in the vehicle target data and the violation feature data are used as measurement input of Kalman filtering, each detected vehicle is tracked by using a Kalman filter, vehicle tracking information is obtained, and the vehicle tracking information is used for carrying out vehicle violation prediction by obtaining the vehicle tracking information;
Step S355: training a violation detection Model, namely training the violation detection Model through the data acquisition, the data preprocessing, the deep learning target detection, the feature extraction and the Kalman filtering vehicle tracking to obtain a violation detection Model V;
Step S356: the construction of the violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning;
step S36: the traffic guiding function is constructed, specifically, the expected speed required to be kept when the following front vehicle moves is calculated, and the calculation formula is as follows:
；
Where V exp is the expected speed to be maintained when following the movement of the preceding vehicle, X exp is a dynamic distance parameter, K is a vehicle speed correction coefficient for the intelligent auxiliary control vehicle body n, L min is the minimum safe distance, V rec is a speed reference value, V max is the maximum safe speed, tanh (. Cndot.) is a hyperbolic tangent function, Is a first expected speed correction parameter,/>Is a second desired speed correction parameter,/>Is the rate of update of the speed;
step S37: the real-time information reminding function is constructed by providing weather and temperature information through a weather application interface, and returning real-time weather data through the weather application interface.
Further, in step S4, the intelligent auxiliary function is activated, and is used for activating the intelligent auxiliary function, specifically activating the functions in the traffic control preparation option function item for detecting the violation according to the activation condition, so as to obtain intelligent function activation information, which includes the following steps:
Step S41: the method comprises the steps of activating a collision warning function, namely setting a threshold value for activating the collision warning function, and activating the collision warning function when an activation condition is met, wherein intelligent auxiliary control of the automobile is carried out when the collision warning function is activated;
The specific step of activating the collision warning function comprises the following steps:
step S411: setting a collision warning function activation threshold, wherein the calculation formula is as follows:
；
Wherein, TTC CY is a collision warning function activation threshold, wherein, TTC Cy1 is a rear-end collision activation threshold, TTC Cj is a lateral collision activation threshold, and TTC Cy2 is a forward collision activation threshold;
step S412: the collision warning activation condition is constructed, and the calculation formula is as follows:
；
Wherein TTC Y is a collision event index set, TTC CY is a collision warning function activation threshold, Y is a collision event index, the value range of the collision event index Y is { Y1, j, Y2}, and TTC is a collision warning function activation threshold;
Step S42: the lane change warning function is activated, specifically, a lane change warning function activation threshold is set, and when the activation condition is met, lane change warning function activation is carried out, and when the lane change warning function is activated, intelligent auxiliary control of the automobile is carried out;
The specific step of activating the collision warning function comprises the following steps:
step S421: setting the activation threshold of the lane change warning function ；
Step S422: the collision warning activation condition is constructed, and the calculation formula is as follows:
；
wherein TTC y3 is a lane change event index;
step S43: the method comprises the steps of activating a curve overspeed warning function, specifically setting a curve overspeed warning activation condition, and when the activation condition is met, activating the curve overspeed warning function, and when the curve overspeed warning function is activated, performing intelligent auxiliary control on an automobile, wherein the calculation formula of the curve overspeed warning activation condition is as follows:
；
In the method, in the process of the invention, Is the road horizontal curvature;
Step S44: the method comprises the steps of activating an emergency notification function, specifically, activating the emergency notification function through road event information returned by a satellite navigation system, and carrying out intelligent auxiliary reminding when the emergency notification function is activated;
Step S45: the method comprises the steps of activating a violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning, and carrying out violation detection warning function activation through returned real-time violation detection warning, wherein when the violation detection warning function is activated, intelligent auxiliary control of the automobile is carried out;
Step S46: the traffic guiding function is activated, specifically, traffic guiding is carried out by obtaining an expected speed V exp required to be kept when the following front vehicle moves, and the traffic guiding function is activated by returning an expected speed V exp required to be kept when the following front vehicle moves, and intelligent auxiliary reminding is carried out when the traffic guiding function is activated;
Step S47: the real-time information reminding function is activated, specifically, real-time information reminding is carried out by returning real-time weather data through a weather application interface, the real-time information reminding function is activated through the returned real-time weather data, and intelligent auxiliary reminding is carried out when the real-time information reminding function is activated.
Further, in step S5, the intelligent auxiliary function priority is set, so as to classify and assign priorities to the intelligent auxiliary functions, specifically, classify the traffic control preparation option function items for detecting violations, obtain intelligent auxiliary function information categories, and perform intelligent auxiliary control and intelligent information reminding according to the classification sequence of the intelligent auxiliary function information categories;
the intelligent auxiliary function information category specifically comprises a violation detection safety risk category, a driving efficiency optimization category and an information service auxiliary category;
the violation detection safety risk comprises a collision warning function, a lane change warning function, a curve overspeed warning function and a violation detection warning function;
The driving efficiency optimization class specifically comprises an emergency notification function and a traffic dispersion function;
the information service auxiliary class specifically comprises a real-time information reminding function;
The classification sequence specifically sets the security risk class for detecting violations as a first priority, the driving efficiency optimization class as a second priority, and the information service auxiliary class as a third priority.
Further, in step S6, the intelligent auxiliary function integration is used for integrating all intelligent auxiliary functions, specifically, information interaction integration is performed through a satellite navigation system, a short-range communication system and a computer terminal, so as to obtain the traffic control integration unit for detecting violations.
Further, in step S7, the intelligent traffic control for detecting the vehicle violation is configured to use the integrated unit to perform intelligent traffic control, specifically, use the integrated unit for detecting the vehicle violation to perform intelligent auxiliary control on the vehicle, so as to complete intelligent traffic control for detecting the vehicle violation.
The invention provides a violation detection traffic control system based on deep learning, which comprises a vehicle intelligent auxiliary control coordinate system construction module, an intelligent auxiliary function selection module, an intelligent auxiliary function construction module, an intelligent auxiliary function activation module, an intelligent auxiliary function priority setting module, an intelligent auxiliary function integration module and a vehicle violation detection intelligent traffic control module;
The vehicle intelligent auxiliary control coordinate system construction module is used for constructing a vehicle intelligent auxiliary control coordinate system, and a vehicle intelligent interaction coordinate system is obtained through the vehicle intelligent auxiliary control coordinate system construction and is used for constructing and selecting intelligent auxiliary functions;
The intelligent auxiliary function selection module is used for intelligent auxiliary function selection, obtaining traffic control preparation selection function items for violation detection through intelligent auxiliary function selection, wherein the traffic control preparation selection function items for violation detection are used for intelligent auxiliary function construction module, activation, priority setting, integration and intelligent traffic control for violation detection;
The intelligent auxiliary function construction module is used for intelligent auxiliary function construction, and an intelligent auxiliary function project set is obtained through the intelligent auxiliary function construction and used for intelligent auxiliary function activation;
the intelligent auxiliary function activation module is used for activating the intelligent auxiliary function, obtaining intelligent function activation information through the intelligent auxiliary function activation and setting the priority of the intelligent auxiliary function;
The intelligent auxiliary function priority setting module is used for setting the priority of the intelligent auxiliary function, obtaining the information category of the intelligent auxiliary function through the priority setting of the intelligent auxiliary function, and carrying out intelligent auxiliary control and intelligent information reminding according to the priority setting of the intelligent auxiliary function;
The intelligent auxiliary function integration module is used for integrating all intelligent auxiliary functions to obtain a traffic control integrated unit for detecting violations, and the traffic control integrated unit for detecting violations of vehicles is used for intelligent traffic control;
The intelligent traffic control module is used for intelligent traffic control of vehicle violation detection and is used for intelligent auxiliary control of the vehicle through the intelligent traffic control of vehicle violation detection.
By adopting the scheme, the beneficial effects obtained by the invention are as follows:
(1) Aiming at the technical problems that in the existing traffic control method for detecting violations, the detection violations are often independent of the traffic control auxiliary system and mainly exist as independent functional modules and cannot be effectively combined with other intelligent driving auxiliary functions, seven intelligent driving auxiliary functions including the detection violations are creatively constructed, and by setting independent activation conditions for each function, intelligent auxiliary control and intelligent information reminding are carried out, so that the functional integration level and the overall usability of traffic control are improved;
(2) Aiming at the technical problems that in the existing traffic control method for detecting violations, the existing method for detecting violations exists, and the performance is poor when a classical Kalman filtering method is used for processing high-dimensional data and a nonlinear system, so that the accuracy of detecting violations is to be improved;
(3) Aiming at the technical problem that the coordination among functions is to be improved under the condition that multiple functions exist in the existing traffic control method for detecting violations, the scheme creatively sets the function classification and the priority, integrates the functions and improves the function cooperation effectiveness and the overall usability of intelligent auxiliary control.
Drawings
FIG. 1 is a schematic flow chart of a traffic control method for detecting violations based on deep learning;
FIG. 2 is a schematic diagram of a traffic control system for detecting violations based on deep learning provided by the invention;
FIG. 3 is a schematic diagram of the intelligent interaction coordinate system of the vehicle in step S1;
FIG. 4 is a flow chart of the intelligent auxiliary function construction in step S3;
fig. 5 is a flow chart illustrating the construction of the violation detection warning function in step S35.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; all other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," "outer," and the like indicate orientation or positional relationships based on those shown in the drawings, merely to facilitate description of the invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the invention.
Referring to fig. 1, the method for detecting traffic control violations based on deep learning provided by the invention comprises the following steps:
step S1: constructing an intelligent auxiliary control coordinate system of the vehicle;
step S2: intelligent auxiliary function selection;
step S3: constructing intelligent auxiliary functions;
Step S4: activating an intelligent auxiliary function;
step S5: setting the priority of the intelligent auxiliary function;
Step S6: intelligent auxiliary function integration;
step S7: and detecting the vehicle violation and controlling intelligent traffic.
In the second embodiment, referring to fig. 1, fig. 2 and fig. 3, in step S1, the vehicle intelligent auxiliary control coordinate system is constructed, and is used for constructing a vehicle scene coordinate system required by vehicle intelligent auxiliary control, specifically, a vehicle intelligent interaction coordinate system is constructed;
The intelligent interaction coordinate system of the vehicle specifically comprises an intelligent auxiliary control vehicle body and a surrounding vehicle entity;
The intelligent auxiliary control vehicle body and surrounding vehicle entities share a motion state through a computer vision technology, wherein the motion state specifically comprises speed, acceleration, course angle and geometric center;
the specific steps for constructing the intelligent interaction coordinate system of the vehicle comprise the following steps:
Step S11: the intelligent auxiliary control vehicle body n is constructed to be a positive rectangle, and the geometric point positions of the intelligent auxiliary control vehicle body n are defined, wherein the intelligent auxiliary control vehicle body comprises a geometric center and four corners, and the calculation formula is as follows:
N={On,An,Bn,Cn,Dn}；
Wherein N is a geometric point of the intelligent auxiliary control vehicle body N, O n is a geometric center of the intelligent auxiliary control vehicle body N, a n is an upper right corner of the intelligent auxiliary control vehicle body N, B n is an upper left corner of the intelligent auxiliary control vehicle body N, C n is a lower left corner of the intelligent auxiliary control vehicle body N, D n is a lower right corner of the intelligent auxiliary control vehicle body N, wherein N is an intelligent auxiliary control vehicle body identifier;
Step S12: the surrounding vehicle entity s is constructed to be in progress, and the geometric point positions of the surrounding vehicle entity s are defined, wherein the geometric point positions comprise geometric centers and four corners, and the calculation formula is as follows:
S={Os,As,Bs,Cs,Ds}；
Where S is the geometric point of the surrounding vehicle entity S, O s is the geometric center of the surrounding vehicle entity S, A s is the upper right hand corner of the surrounding vehicle entity S, B s is the upper left hand corner of the surrounding vehicle entity S, C s is the lower left hand corner of the surrounding vehicle entity S, and D s is the lower right hand corner of the surrounding vehicle entity S, where S is the surrounding vehicle entity identifier;
Step S13: defining a vehicle parameter set, wherein the calculation formula is as follows:
；
Wherein P is a vehicle parameter set, w i is a vehicle width parameter, i is a vehicle entity identifier, and i has a value range of N is the intelligent auxiliary control vehicle body identifier, s is the surrounding vehicle entity identifier, l i is the vehicle length parameter,/>Is a vehicle speed vector parameter,/>Is the included angle parameter of the intelligent auxiliary control vehicle body n and the surrounding vehicle entity s, and the included angle parameter/>The value range of (2) is/>The included angle parameter/>In particular by vehicle speed vector parameters/>Determining the heading angle difference of the vehicle;
Step S14: defining a local coordinate system, specifically, taking a geometric center O s of the surrounding vehicle entity s as a coordinate system origin, taking rays in a parallel direction along the geometric center O s and the side C nDn of the surrounding vehicle entity s as a horizontal coordinate axis x, taking rays in a parallel direction along the geometric center O s and the side B nCn of the surrounding vehicle entity s as a vertical coordinate axis y, representing a projection from the geometric center O n of the intelligent auxiliary control vehicle body n to the horizontal coordinate axis x as x O, and representing a projection from the geometric center O n of the intelligent auxiliary control vehicle body n to the vertical coordinate axis y as y O;
Step S15: defining a predicted collision type of the illegal collision, specifically defining a forward collision, a side collision and a rear-end collision;
The forward collision specifically refers to the collision occurring at any point on the edge B nAn except for A n and B n;
The side impact, specifically, the impact at any point on side B nCn except B n and C n, and the impact at any point on side a nDn except a n and D n;
The rear-end collision specifically refers to a collision occurring at any point on the edge C nDn except for C n and D n.
An embodiment III, referring to FIG. 1 and FIG. 2, in which, in step S2, the intelligent auxiliary function selection is used to define available intelligent auxiliary functions, and select specific traffic control items for detecting violations from the intelligent auxiliary functions, specifically, by defining traffic control preparation options for detecting violations, the intelligent auxiliary function selection is performed, and the subsequent intelligent auxiliary function construction, activation, priority setting, integration and intelligent traffic control for detecting violations are performed according to the traffic control preparation options for detecting violations;
the traffic control preparation and selection function items for the violation detection specifically comprise a collision warning function, a lane change warning function, a curve overspeed warning function, an emergency notification function, a violation detection warning function, a traffic guiding function and a real-time information reminding function.
In step S3, the intelligent auxiliary function construction is used for performing intelligent auxiliary function basic construction according to the function item selected by the intelligent auxiliary function, specifically performing intelligent auxiliary function construction according to the traffic control preparation option for detecting the violation, and obtaining an intelligent auxiliary function item set, and the method comprises the following steps:
step S31: the collision warning function is constructed, specifically, a collision event index set is constructed, the collision event index set is used for judging whether collision occurs or not, the collision warning function comprises a rear-end collision index, a forward collision index and a side collision index, and the collision event index set is constructed, and the method comprises the following steps:
Step S311: the rear-end collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC y1 is a rear-end collision index for determining a condition for the occurrence of a rear-end collision, y O is a projection of a geometric center O n of the intelligent auxiliary control vehicle body n to a vertical coordinate axis y, l s is a vehicle length parameter of a surrounding vehicle entity s, w s is a vehicle width parameter of the surrounding vehicle entity s, Is the included angle parameter of the intelligent auxiliary control vehicle body n and the surrounding vehicle entities s, l n is the vehicle length parameter of the intelligent auxiliary control vehicle body n,/>Is a vehicle speed vector parameter of surrounding vehicle entities s,/>Is a vehicle speed vector parameter for intelligently assisting in controlling the vehicle body n, and I/I is a modulo operator;
step S312: the side collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC j is a side collision index for determining a condition for occurrence of a side collision, x O is a projection of a geometric center O n of the intelligent auxiliary control vehicle body n onto a horizontal coordinate axis x, and w n is a vehicle width parameter of the intelligent auxiliary control vehicle body n;
Step S313: the forward collision index is constructed, and the calculation formula is as follows:
；
Wherein TTC y2 is a forward collision index for judging conditions for rear-end collision;
Step S32: the construction of the lane change warning function is specifically to construct a lane change event index which is used as a condition for judging the potential risk of the lane change, and the calculation formula of the lane change event index is as follows:
；
wherein TTC y3 is a lane change event index;
Step S33: the construction of the curve overspeed warning function, in particular to the construction of curve overspeed event indexes, which are used as conditions for judging the potential risk of curve overspeed, wherein the calculation formula for constructing the curve overspeed event indexes is as follows:
；
Where v cst is the minimum between the critical side slip velocity and the critical side turn velocity on the curved road, used as a condition for discriminating the potential risk of curve overspeed, Is the critical sideslip velocity on curved roads,/>Is the critical d rollover speed on curved roads, |·| is the modulo operator;
Step S34: the construction of the emergency notification function, specifically, collecting road facilities and traffic environment data through a satellite navigation system, and carrying out the construction of the emergency notification function and the special event notification function;
Step S35: the construction of the violation detection warning function, specifically, the construction of a computer vision violation detection model by adopting a deep learning method, and the violation detection by the violation detection model, comprises the following steps:
step S351: the method comprises the steps of data acquisition, namely acquiring violation detection original data from crossroad camera image records;
step S352: the data preprocessing is specifically to perform image denoising and contrast enhancement operation on the violation detection original data to obtain violation detection optimized data;
Step S353: deep learning target detection and feature extraction, specifically, performing target detection on the violation detection optimization data by adopting a YOLOv model to obtain vehicle target data, and performing feature extraction on the violation detection optimization data by constructing a convolutional neural network to obtain violation feature data;
step S354: the Kalman filtering vehicle tracking is specifically carried out according to the vehicle target data and the violation feature data, specifically, the vehicle position in the vehicle target data and the violation feature data are used as measurement input of Kalman filtering, each detected vehicle is tracked by using a Kalman filter, vehicle tracking information is obtained, and the vehicle tracking information is used for carrying out vehicle violation prediction by obtaining the vehicle tracking information;
Step S355: training a violation detection Model, namely training the violation detection Model through the data acquisition, the data preprocessing, the deep learning target detection, the feature extraction and the Kalman filtering vehicle tracking to obtain a violation detection Model V;
Step S356: the construction of the violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning;
step S36: the traffic guiding function is constructed, specifically, the expected speed required to be kept when the following front vehicle moves is calculated, and the calculation formula is as follows:
；
Where V exp is the expected speed to be maintained when following the movement of the preceding vehicle, X exp is a dynamic distance parameter, K is a vehicle speed correction coefficient for the intelligent auxiliary control vehicle body n, L min is the minimum safe distance, V rec is a speed reference value, V max is the maximum safe speed, tanh (. Cndot.) is a hyperbolic tangent function, Is a first expected speed correction parameter,/>Is a second desired speed correction parameter,/>Is the rate of update of the speed;
step S37: the real-time information reminding function is constructed by providing weather and temperature information through a weather application interface, and returning real-time weather data through the weather application interface.
By executing the above operation, aiming at the technical problems that in the existing traffic control method for detecting violations, the detection violations often exist independently of the traffic control auxiliary system and mainly exist as independent functional modules and cannot be effectively combined with other intelligent driving auxiliary functions, seven intelligent driving auxiliary functions including the detection violations are creatively constructed, and by setting independent activation conditions for each function, intelligent auxiliary control and intelligent information reminding are carried out, so that the functional integration level and the overall usability of traffic control are improved;
aiming at the technical problems that in the existing traffic control method for detecting violations, the existing method for detecting violations exists, and the performance of the classical Kalman filtering method is poor when the high-dimensional data and a nonlinear system are processed, so that the accuracy of detecting violations is to be improved.
In a fifth embodiment, referring to fig. 1, fig. 2 and fig. 3, the embodiment is based on the above embodiment, and in step S4, the intelligent auxiliary function is activated, and is used for activating the intelligent auxiliary function, specifically activating a function in the traffic control preparation option function item for detecting the violation according to an activation condition, so as to obtain intelligent function activation information, and the method includes the following steps:
Step S41: the method comprises the steps of activating a collision warning function, namely setting a threshold value for activating the collision warning function, and activating the collision warning function when an activation condition is met, wherein intelligent auxiliary control of the automobile is carried out when the collision warning function is activated;
The specific step of activating the collision warning function comprises the following steps:
step S411: setting a collision warning function activation threshold, wherein the calculation formula is as follows:
；
Wherein, TTC CY is a collision warning function activation threshold, wherein, TTC Cy1 is a rear-end collision activation threshold, TTC Cj is a lateral collision activation threshold, and TTC Cy2 is a forward collision activation threshold;
step S412: the collision warning activation condition is constructed, and the calculation formula is as follows:
；
Wherein TTC Y is a collision event index set, TTC CY is a collision warning function activation threshold, Y is a collision event index, the value range of the collision event index Y is { Y1, j, Y2}, and TTC is a collision warning function activation threshold;
Step S42: the lane change warning function is activated, specifically, a lane change warning function activation threshold is set, and when the activation condition is met, lane change warning function activation is carried out, and when the lane change warning function is activated, intelligent auxiliary control of the automobile is carried out;
The specific step of activating the collision warning function comprises the following steps:
step S421: setting the activation threshold of the lane change warning function ；
Step S422: the collision warning activation condition is constructed, and the calculation formula is as follows:
；
wherein TTC y3 is a lane change event index;
step S43: the method comprises the steps of activating a curve overspeed warning function, specifically setting a curve overspeed warning activation condition, and when the activation condition is met, activating the curve overspeed warning function, and when the curve overspeed warning function is activated, performing intelligent auxiliary control on an automobile, wherein the calculation formula of the curve overspeed warning activation condition is as follows:
；
In the method, in the process of the invention, Is the road horizontal curvature;
Step S44: the method comprises the steps of activating an emergency notification function, specifically, activating the emergency notification function through road event information returned by a satellite navigation system, and carrying out intelligent auxiliary reminding when the emergency notification function is activated;
Step S45: the method comprises the steps of activating a violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning, and carrying out violation detection warning function activation through returned real-time violation detection warning, wherein when the violation detection warning function is activated, intelligent auxiliary control of the automobile is carried out;
Step S46: the traffic guiding function is activated, specifically, traffic guiding is carried out by obtaining an expected speed V exp required to be kept when the following front vehicle moves, and the traffic guiding function is activated by returning an expected speed V exp required to be kept when the following front vehicle moves, and intelligent auxiliary reminding is carried out when the traffic guiding function is activated;
Step S47: the real-time information reminding function is activated, specifically, real-time information reminding is carried out by returning real-time weather data through a weather application interface, the real-time information reminding function is activated through the returned real-time weather data, and intelligent auxiliary reminding is carried out when the real-time information reminding function is activated.
An embodiment six, referring to fig. 1 and fig. 2, based on the above embodiment, in step S5, the intelligent auxiliary function priority setting is used for classifying and allocating the intelligent auxiliary functions, specifically classifying the traffic control preparation option for detecting the violations, to obtain intelligent auxiliary function information categories, and performing intelligent auxiliary control and intelligent information reminding according to the classification sequence of the intelligent auxiliary function information categories;
the intelligent auxiliary function information category specifically comprises a violation detection safety risk category, a driving efficiency optimization category and an information service auxiliary category;
the violation detection safety risk comprises a collision warning function, a lane change warning function, a curve overspeed warning function and a violation detection warning function;
The driving efficiency optimization class specifically comprises an emergency notification function and a traffic dispersion function;
the information service auxiliary class specifically comprises a real-time information reminding function;
The classification sequence specifically sets the security risk class for detecting violations as a first priority, the driving efficiency optimization class as a second priority, and the information service auxiliary class as a third priority.
By executing the above operation, aiming at the technical problem that the coordination among functions is to be improved under the condition that multiple functions exist in the existing traffic control method for detecting the violations, the scheme creatively sets the function classification and the priority, integrates the functions and improves the function cooperation effectiveness and the overall usability of intelligent auxiliary control.
In a seventh embodiment, referring to fig. 1 and fig. 2, in step S6, the intelligent auxiliary functions are integrated, and are used for integrating all intelligent auxiliary functions, specifically, information interaction integration is performed through a satellite navigation system, a short-range communication system and a computer terminal, so as to obtain a traffic control integrated unit for detecting violations.
In the eighth embodiment, referring to fig. 1 and fig. 2, in step S7, the vehicle violation detection intelligent traffic control is configured to use an integrated unit to perform intelligent traffic control, specifically, use the violation detection traffic control integrated unit to perform vehicle intelligent auxiliary control, so as to complete vehicle violation detection intelligent traffic control.
An embodiment nine, referring to fig. 1 and fig. 2, based on the above embodiment, the present invention provides a traffic control system for detecting violations based on deep learning, which includes a vehicle intelligent auxiliary control coordinate system construction module, an intelligent auxiliary function selection module, an intelligent auxiliary function construction module, an intelligent auxiliary function activation module, an intelligent auxiliary function priority setting module, an intelligent auxiliary function integration module, and a vehicle traffic control module for detecting violations;
The vehicle intelligent auxiliary control coordinate system construction module is used for constructing a vehicle intelligent auxiliary control coordinate system, and a vehicle intelligent interaction coordinate system is obtained through the vehicle intelligent auxiliary control coordinate system construction and is used for constructing and selecting intelligent auxiliary functions;
The intelligent auxiliary function selection module is used for intelligent auxiliary function selection, obtaining traffic control preparation selection function items for violation detection through intelligent auxiliary function selection, wherein the traffic control preparation selection function items for violation detection are used for intelligent auxiliary function construction module, activation, priority setting, integration and intelligent traffic control for violation detection;
The intelligent auxiliary function construction module is used for intelligent auxiliary function construction, and an intelligent auxiliary function project set is obtained through the intelligent auxiliary function construction and used for intelligent auxiliary function activation;
the intelligent auxiliary function activation module is used for activating the intelligent auxiliary function, obtaining intelligent function activation information through the intelligent auxiliary function activation and setting the priority of the intelligent auxiliary function;
The intelligent auxiliary function priority setting module is used for setting the priority of the intelligent auxiliary function, obtaining the information category of the intelligent auxiliary function through the priority setting of the intelligent auxiliary function, and carrying out intelligent auxiliary control and intelligent information reminding according to the priority setting of the intelligent auxiliary function;
The intelligent auxiliary function integration module is used for integrating all intelligent auxiliary functions to obtain a traffic control integrated unit for detecting violations, and the traffic control integrated unit for detecting violations of vehicles is used for intelligent traffic control;
The intelligent traffic control module is used for intelligent traffic control of vehicle violation detection and is used for intelligent auxiliary control of the vehicle through the intelligent traffic control of vehicle violation detection.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made hereto without departing from the spirit and principles of the present invention.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.
Claims (6)
1. A violation detection traffic control method based on deep learning is characterized in that: the method comprises the following steps:
Step S1: the vehicle intelligent auxiliary control coordinate system is constructed and used for constructing a vehicle scene coordinate system required by vehicle intelligent auxiliary control, in particular to a vehicle intelligent interaction coordinate system;
Step S2: the intelligent auxiliary function selection is used for defining available intelligent auxiliary functions, selecting specifically used traffic control items for violation detection from the intelligent auxiliary functions, specifically preparing and selecting function items by defining the traffic control for violation detection, and carrying out intelligent auxiliary function selection;
The traffic control preparation and selection function items for the violation detection specifically comprise a collision warning function, a lane change warning function, a curve overspeed warning function, an emergency notification function, a violation detection warning function, a traffic guiding function and a real-time information reminding function;
Step S3: the intelligent auxiliary function construction is used for carrying out intelligent auxiliary function basic construction according to the function item selected by the intelligent auxiliary function, specifically carrying out intelligent auxiliary function construction according to the traffic control preparation and selection function item detected by the violation detection, and obtaining an intelligent auxiliary function item set, and comprises the following steps:
Step S31: the collision warning function construction specifically comprises a collision event index set, wherein the collision event index set is used for judging whether collision occurs or not and comprises a rear-end collision index, a forward collision index and a side collision index, and the collision event index set is constructed specifically by the following steps:
Step S311: the rear-end collision index is constructed, and the calculation formula is as follows:
Wherein TTC y1 is a rear-end collision index for determining a condition of a rear-end collision, y O is a projection of a geometric center O n of an intelligent auxiliary control vehicle body n to a vertical coordinate axis y, l s is a vehicle length parameter of a surrounding vehicle entity s, w s is a vehicle width parameter of the surrounding vehicle entity s, α is an angle parameter of the intelligent auxiliary control vehicle body n and the surrounding vehicle entity s, l n is a vehicle length parameter of the intelligent auxiliary control vehicle body n, Is a vehicle speed vector parameter of surrounding vehicle entities s,/>Is a vehicle speed vector parameter for intelligently assisting in controlling the vehicle body n, and I/I is a modulo operator;
step S312: the side collision index is constructed, and the calculation formula is as follows:
Wherein TTC j is a side collision index for determining a condition for occurrence of a side collision, x O is a projection of a geometric center O n of the intelligent auxiliary control vehicle body n onto a horizontal coordinate axis x, and w n is a vehicle width parameter of the intelligent auxiliary control vehicle body n;
Step S313: the forward collision index is constructed, and the calculation formula is as follows:
Wherein TTC y2 is a forward collision index for judging conditions for rear-end collision;
Step S32: the construction of the lane change warning function, in particular to the construction of lane change event indexes, which are used as conditions for judging the lane change potential risk, wherein the calculation formula for constructing the lane change event indexes is as follows:
wherein TTC y3 is a lane change event index;
Step S33: the construction of the curve overspeed warning function, in particular to the construction of curve overspeed event indexes, which are used as conditions for judging the potential risk of curve overspeed, wherein the calculation formula for constructing the curve overspeed event indexes is as follows:
Where v cst is the minimum between the critical side slip velocity and the critical side turn velocity on the curved road, used as a condition for discriminating the potential risk of curve overspeed, Is the critical sideslip velocity on curved roads,/>Is the critical d rollover speed on curved roads, |·| is the modulo operator;
Step S34: the construction of the emergency notification function, specifically, collecting road facilities and traffic environment data through a satellite navigation system, and carrying out the construction of the emergency notification function and the special event notification function;
Step S35: the construction of the violation detection warning function is realized by adopting a deep learning method to construct a computer vision violation detection model, and the violation detection is carried out through the violation detection model, and the method specifically comprises the following steps of:
step S351: the method comprises the steps of data acquisition, namely acquiring violation detection original data from crossroad camera image records;
step S352: the data preprocessing is specifically to perform image denoising and contrast enhancement operation on the violation detection original data to obtain violation detection optimized data;
Step S353: deep learning target detection and feature extraction, specifically, performing target detection on the violation detection optimization data by adopting a YOLOv model to obtain vehicle target data, and performing feature extraction on the violation detection optimization data by constructing a convolutional neural network to obtain violation feature data;
step S354: the Kalman filtering vehicle tracking is specifically carried out according to the vehicle target data and the violation feature data, specifically, the vehicle position in the vehicle target data and the violation feature data are used as measurement input of Kalman filtering, each detected vehicle is tracked by using a Kalman filter, vehicle tracking information is obtained, and the vehicle tracking information is used for carrying out vehicle violation prediction by obtaining the vehicle tracking information;
Step S355: training a violation detection Model, namely training the violation detection Model through the data acquisition, the data preprocessing, the deep learning target detection, the feature extraction and the Kalman filtering vehicle tracking to obtain a violation detection Model V;
Step S356: the construction of the violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning;
step S36: the traffic guiding function is constructed, specifically, the expected speed required to be kept when the following front vehicle moves is calculated, and the calculation formula is as follows:
Wherein V exp is an expected speed to be maintained when following the movement of the preceding vehicle, X exp is a dynamic distance parameter, K is a vehicle speed correction coefficient of the intelligent auxiliary control vehicle body n, L min is a minimum safe distance, V rec is a speed reference value, V max is a maximum safe speed, tanh (·) is a hyperbolic tangent function, λ 1 is a first expected speed correction parameter, λ 2 is a second expected speed correction parameter, Δt is a speed update rate;
Step S37: the real-time information reminding function construction is specifically implemented by providing weather and temperature information through a weather application interface, and returning real-time weather data through the weather application interface to carry out the real-time information reminding function construction;
Step S4: the intelligent auxiliary function activation is used for activating the intelligent auxiliary function, specifically activating the function in the traffic control preparation option function item for detecting the violation according to the activation condition to obtain intelligent function activation information, and specifically comprises the following steps:
Step S41: the method comprises the steps of activating a collision warning function, namely setting a threshold value for activating the collision warning function, and activating the collision warning function when an activation condition is met, wherein intelligent auxiliary control of the automobile is carried out when the collision warning function is activated;
Step S42: the lane change warning function is activated, specifically, a lane change warning function activation threshold is set, and when the activation condition is met, lane change warning function activation is carried out, and when the lane change warning function is activated, intelligent auxiliary control of the automobile is carried out;
step S43: the method comprises the steps of activating a curve overspeed warning function, namely setting a curve overspeed warning activation condition, and activating the curve overspeed warning function when the activation condition is met, wherein intelligent auxiliary control of the automobile is carried out when the curve overspeed warning function is activated;
Step S44: the method comprises the steps of activating an emergency notification function, specifically, activating the emergency notification function through road event information returned by a satellite navigation system, and carrying out intelligent auxiliary reminding when the emergency notification function is activated;
Step S45: the method comprises the steps of activating a violation detection warning function, specifically, using the violation detection Model V to carry out real-time violation detection warning, and carrying out violation detection warning function activation through returned real-time violation detection warning, wherein when the violation detection warning function is activated, intelligent auxiliary control of the automobile is carried out;
Step S46: the traffic guiding function is activated, specifically, traffic guiding is carried out by obtaining an expected speed V exp required to be kept when the following front vehicle moves, and the traffic guiding function is activated by returning an expected speed V exp required to be kept when the following front vehicle moves, and intelligent auxiliary reminding is carried out when the traffic guiding function is activated;
Step S47: the method comprises the steps of activating a real-time information reminding function, specifically, returning real-time weather data by using a weather application interface, carrying out real-time information reminding, and carrying out intelligent auxiliary reminding when the real-time information reminding function is activated by the returned real-time weather data;
Step S5: the intelligent auxiliary function priority is used for classifying and distributing the intelligent auxiliary functions, specifically classifying the traffic control preparation and selection function items for detecting the violations to obtain intelligent auxiliary function information categories, and carrying out intelligent auxiliary control and intelligent information reminding according to the classification sequence of the intelligent auxiliary function information categories;
the intelligent auxiliary function information category specifically comprises a violation detection safety risk category, a driving efficiency optimization category and an information service auxiliary category;
the violation detection safety risk comprises a collision warning function, a lane change warning function, a curve overspeed warning function and a violation detection warning function;
The driving efficiency optimization class specifically comprises an emergency notification function and a traffic dispersion function;
the information service auxiliary class specifically comprises a real-time information reminding function;
the classification sequence specifically sets the security risk class for detecting violations as a first priority, sets the driving efficiency optimization class as a second priority, and sets the information service auxiliary class as a third priority;
step S6: the intelligent auxiliary function integration is used for integrating all intelligent auxiliary functions, and particularly, the intelligent auxiliary function integration is carried out through information interaction integration of a satellite navigation system, a short-range communication system and a computer terminal to obtain a traffic control integrated unit for detecting violations;
step S7: the intelligent traffic control system comprises an integrated unit, a traffic control system and a traffic control system.
2. The deep learning-based traffic control method for detecting violations according to claim 1, wherein: in step S1, the vehicle intelligent auxiliary control coordinate system is constructed, and is used for constructing a vehicle scene coordinate system required by vehicle intelligent auxiliary control, specifically, a vehicle intelligent interaction coordinate system is constructed;
The intelligent interaction coordinate system of the vehicle specifically comprises an intelligent auxiliary control vehicle body and a surrounding vehicle entity;
The intelligent auxiliary control vehicle body and surrounding vehicle entities share a motion state through a computer vision technology, wherein the motion state specifically comprises a speed, an acceleration, a course angle and a geometric center.
3. The deep learning-based traffic control method for violation detection of traffic according to claim 2, wherein: in step S2, the intelligent auxiliary function selection is used for defining available intelligent auxiliary functions, selecting specifically used traffic control items for violation detection from the intelligent auxiliary functions, specifically, selecting intelligent auxiliary functions by defining traffic control preparation option items for violation detection, and performing subsequent intelligent auxiliary function construction, activation, priority setting, integration and intelligent traffic control for violation detection according to the traffic control preparation option items for violation detection.
4. The deep learning-based traffic control method for detecting violations according to claim 3, wherein: in step S4, the intelligent auxiliary function is activated, which is used for activating the intelligent auxiliary function, specifically activating the functions in the traffic control preparation and selection function item for detecting the violation according to the activation condition, so as to obtain intelligent function activation information, and the method includes the following steps:
In step S41, the specific step of activating the collision warning function includes:
step S411: setting a collision warning function activation threshold, wherein the calculation formula is as follows:
TTCCY＝{TTCCy1,TTCCj,TTCCy2}；
Wherein, TTC CY is a collision warning function activation threshold, wherein, TTC Cy1 is a rear-end collision activation threshold, TTC Cj is a lateral collision activation threshold, and TTC Cy2 is a forward collision activation threshold;
step S412: the collision warning activation condition is constructed, and the calculation formula is as follows:
TTCY≤TTCCYandTTCCY＞0；
Wherein TTC Y is a collision event index set, TTC CY is a collision warning function activation threshold, Y is a collision event index, the value range of the collision event index Y is { Y1, j, Y2}, and TTC is a collision warning function activation threshold;
in step S42, the specific step of activating the collision warning function includes:
Step S421: setting a lane change warning function activation threshold TTC Cy3;
Step S422: the collision warning activation condition is constructed, and the calculation formula is as follows:
TTCy3≤TTCCy3andTTCy3＞TTCCy3；
wherein TTC y3 is a lane change event index;
In step S43, the calculation formula of the curve overspeed warning activation condition is:
Where ρ is the road horizontal curvature.
5. A deep learning-based traffic control system for implementing a deep learning-based traffic control method for detecting violations as claimed in any of claims 1-4, characterized in that: the intelligent traffic control system comprises a vehicle intelligent auxiliary control coordinate system construction module, an intelligent auxiliary function selection module, an intelligent auxiliary function construction module, an intelligent auxiliary function activation module, an intelligent auxiliary function priority setting module, an intelligent auxiliary function integration module and a vehicle violation detection intelligent traffic control module.
6. The deep learning based traffic control system for violation detection of claim 5, wherein: the vehicle intelligent auxiliary control coordinate system construction module is used for constructing a vehicle intelligent auxiliary control coordinate system, and a vehicle intelligent interaction coordinate system is obtained through the vehicle intelligent auxiliary control coordinate system construction and is used for constructing and selecting intelligent auxiliary functions;
The intelligent auxiliary function selection module is used for intelligent auxiliary function selection, obtaining traffic control preparation selection function items for violation detection through intelligent auxiliary function selection, wherein the traffic control preparation selection function items for violation detection are used for intelligent auxiliary function construction module, activation, priority setting, integration and intelligent traffic control for violation detection;
The intelligent auxiliary function construction module is used for intelligent auxiliary function construction, and an intelligent auxiliary function project set is obtained through the intelligent auxiliary function construction and used for intelligent auxiliary function activation;
the intelligent auxiliary function activation module is used for activating the intelligent auxiliary function, obtaining intelligent function activation information through the intelligent auxiliary function activation and setting the priority of the intelligent auxiliary function;
The intelligent auxiliary function priority setting module is used for setting the priority of the intelligent auxiliary function, obtaining the information category of the intelligent auxiliary function through the priority setting of the intelligent auxiliary function, and carrying out intelligent auxiliary control and intelligent information reminding according to the priority setting of the intelligent auxiliary function;
The intelligent auxiliary function integration module is used for integrating all intelligent auxiliary functions to obtain a traffic control integrated unit for detecting violations, and the traffic control integrated unit for detecting violations of vehicles is used for intelligent traffic control;
The intelligent traffic control module is used for intelligent traffic control of vehicle violation detection and is used for intelligent auxiliary control of the vehicle through the intelligent traffic control of vehicle violation detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410386339.3A CN117975382B (en) | 2024-04-01 | 2024-04-01 | Violation detection traffic control method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410386339.3A CN117975382B (en) | 2024-04-01 | 2024-04-01 | Violation detection traffic control method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117975382A CN117975382A (en) | 2024-05-03 |
| CN117975382B true CN117975382B (en) | 2024-06-11 |
Family
ID=90865034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410386339.3A Active CN117975382B (en) | 2024-04-01 | 2024-04-01 | Violation detection traffic control method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117975382B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015022567A1 (en) * | 2013-08-14 | 2015-02-19 | AGHAJANZADEH, Naser | Assistance system for automated, intelligent management of traffic regulations |
| CN108154683A (en) * | 2017-12-26 | 2018-06-12 | 河北农业大学 | Intelligent traffic administration system method and system |
| DE102018007087A1 (en) * | 2018-09-07 | 2019-03-07 | Daimler Ag | Driver assistance process |
| CN111583659A (en) * | 2020-05-08 | 2020-08-25 | 济南博观智能科技有限公司 | Vehicle violation determination method, device, equipment and medium |
| CN112071058A (en) * | 2020-08-14 | 2020-12-11 | 深延科技(北京)有限公司 | Road traffic monitoring and vehicle abnormity, contraband and fire detection method and system based on deep learning |
| CN112477847A (en) * | 2020-12-11 | 2021-03-12 | 清华大学苏州汽车研究院（吴江） | Traffic jam auxiliary control method and system |
| CN113593250A (en) * | 2021-07-12 | 2021-11-02 | 浙江工贸职业技术学院 | Illegal parking detection system based on visual identification |
| CN115123297A (en) * | 2022-06-24 | 2022-09-30 | 阿波罗智联(北京)科技有限公司 | Vehicle control method, device, equipment, vehicle and storage medium |
| CN117238142A (en) * | 2023-07-28 | 2023-12-15 | 武汉盛信鸿通科技有限公司 | Intelligent video analysis-based intersection non-motor vehicle violation identification method |
| CN117612364A (en) * | 2023-10-13 | 2024-02-27 | 深圳市综合交通与市政工程设计研究总院有限公司 | Method and device for detecting motor vehicle violation, electronic equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150061752A (en) * | 2013-11-28 | 2015-06-05 | 현대모비스 주식회사 | Device for driving assist and method for activating the function automatically by the device |
| US10296794B2 (en) * | 2016-12-20 | 2019-05-21 | Jayant Rtti | On-demand artificial intelligence and roadway stewardship system |
| KR102485348B1 (en) * | 2017-12-28 | 2023-01-05 | 현대자동차주식회사 | Apparatus and method for controlling the driving assist function of vehicle automatically |
-
2024
- 2024-04-01 CN CN202410386339.3A patent/CN117975382B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015022567A1 (en) * | 2013-08-14 | 2015-02-19 | AGHAJANZADEH, Naser | Assistance system for automated, intelligent management of traffic regulations |
| CN108154683A (en) * | 2017-12-26 | 2018-06-12 | 河北农业大学 | Intelligent traffic administration system method and system |
| DE102018007087A1 (en) * | 2018-09-07 | 2019-03-07 | Daimler Ag | Driver assistance process |
| CN111583659A (en) * | 2020-05-08 | 2020-08-25 | 济南博观智能科技有限公司 | Vehicle violation determination method, device, equipment and medium |
| CN112071058A (en) * | 2020-08-14 | 2020-12-11 | 深延科技(北京)有限公司 | Road traffic monitoring and vehicle abnormity, contraband and fire detection method and system based on deep learning |
| CN112477847A (en) * | 2020-12-11 | 2021-03-12 | 清华大学苏州汽车研究院（吴江） | Traffic jam auxiliary control method and system |
| CN113593250A (en) * | 2021-07-12 | 2021-11-02 | 浙江工贸职业技术学院 | Illegal parking detection system based on visual identification |
| CN115123297A (en) * | 2022-06-24 | 2022-09-30 | 阿波罗智联(北京)科技有限公司 | Vehicle control method, device, equipment, vehicle and storage medium |
| CN117238142A (en) * | 2023-07-28 | 2023-12-15 | 武汉盛信鸿通科技有限公司 | Intelligent video analysis-based intersection non-motor vehicle violation identification method |
| CN117612364A (en) * | 2023-10-13 | 2024-02-27 | 深圳市综合交通与市政工程设计研究总院有限公司 | Method and device for detecting motor vehicle violation, electronic equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| Deep learning based vehicle violation detection system;Rui Xu等;《2021 6th International Conference on Intelligent Computing and Signal Processing (ICSP)》;20210426;第796-799页 * |
| 列车数据高速转储和司机违章检测***研究;夏珊珊;《全国优秀硕士学位论文数据集》;20230215;第1-72页 * |
| 深度学习中的单阶段车辆检测算法综述;杨锦帆等;《计算机工程与应用》;20211202;第55-67页 * |
| 车联网环境下实时交通信息对驾驶行为的影响研究;赵文静;《全国优秀博士学位论文数据集》;20231215;第1-209页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117975382A (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11380193B2 (en) | Method and system for vehicular-related communications | |
| CN110362077B (en) | Unmanned vehicle emergency hedge decision making system, method and medium | |
| Bila et al. | Vehicles of the future: A survey of research on safety issues | |
| Nidamanuri et al. | A progressive review: Emerging technologies for ADAS driven solutions | |
| Cheng | Autonomous intelligent vehicles: theory, algorithms, and implementation | |
| EP3032454B1 (en) | Method and system for adaptive ray based scene analysis of semantic traffic spaces and vehicle equipped with such system | |
| US20210380115A1 (en) | Devices and methods for predicting collisions and/or intersection violations | |
| US20140354684A1 (en) | Symbology system and augmented reality heads up display (hud) for communicating safety information | |
| US20210216798A1 (en) | Using captured video data to identify pose of a vehicle | |
| DE102020127043A1 (en) | SECURITY SYSTEM FOR ONE VEHICLE | |
| CN111891061B (en) | Vehicle collision detection method and device and computer equipment | |
| US11685406B2 (en) | Vehicle control device, vehicle control method, and storage medium | |
| US20220242453A1 (en) | Detection System for Predicting Information on Pedestrian | |
| JP7289760B2 (en) | electronic controller | |
| JP6756908B2 (en) | Methods and devices for detecting pedestrians around the vehicle | |
| US20210312193A1 (en) | Devices and methods for predicting intersection violations and/or collisions | |
| US20230166731A1 (en) | Devices and methods for assisting operation of vehicles based on situational assessment fusing expoential risks (safer) | |
| Kim et al. | Lane change intention classification of surrounding vehicles utilizing open set recognition | |
| JP5593217B2 (en) | Vehicle external recognition device and vehicle system using the same | |
| CN117975382B (en) | Violation detection traffic control method and system based on deep learning | |
| CN115123291B (en) | Behavior prediction method and device based on obstacle recognition | |
| JP6171608B2 (en) | Object detection device | |
| US20210309221A1 (en) | Devices and methods for determining region of interest for object detection in camera images | |
| EP4239616A1 (en) | Object tracking device and object tracking method | |
| CN110349425B (en) | Important target generation method for vehicle-road cooperative automatic driving system |
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 |