CN118015844A - Traffic dynamic control method and system based on deep learning network - Google Patents
Traffic dynamic control method and system based on deep learning network Download PDFInfo
- Publication number
- CN118015844A CN118015844A CN202410425775.7A CN202410425775A CN118015844A CN 118015844 A CN118015844 A CN 118015844A CN 202410425775 A CN202410425775 A CN 202410425775A CN 118015844 A CN118015844 A CN 118015844A
- Authority
- CN
- China
- Prior art keywords
- dynamic target
- vertical motion
- grid
- risk threshold
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013135 deep learning Methods 0.000 title claims abstract description 20
- 230000033001 locomotion Effects 0.000 claims abstract description 560
- 238000012502 risk assessment Methods 0.000 claims abstract description 28
- 238000012216 screening Methods 0.000 claims description 36
- 230000005540 biological transmission Effects 0.000 claims description 33
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 abstract description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Traffic Control Systems (AREA)
Abstract
The application belongs to the technical field of road traffic, and particularly relates to a traffic dynamic control method and system based on a deep learning network, wherein a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence are acquired, each dynamic target sequence in each sequence group of a vertical motion sequence group and a horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, and the dynamic target sequences in each sequence group are sequentially detected according to vertical or horizontal motion speed; multiple times of adjustment are carried out, the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and equal risk threshold values in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are determined; the dynamic target sequences in each sequence group are sequentially detected according to the vertical movement speed. The application can effectively detect the running risk according to the dynamic target states of different levels of the vertical level, and has the function of reasonably monitoring and early warning the road traffic safety.
Description
Technical Field
The application relates to the technical field of road traffic, in particular to a traffic dynamic control method and system based on a deep learning network.
Background
At present, with the improvement of urban and motorized levels, the demands of various communities on intelligent traffic systems are also increasing. The development of artificial intelligence also makes the intelligent traffic system become the development direction of future traffic, and road monitoring is taken as a key in the intelligent traffic system, plays a very important role in the intelligent traffic system, and has great research and application value in realizing the segmentation of targets such as vehicles, pedestrians and the like in traffic monitoring videos.
Although the current methods for safe driving early warning are various, only one risk situation in traffic can be usually early warned, and a unified index is lacking to evaluate the risk degree of various situations possibly occurring in a road.
Disclosure of Invention
The application mainly aims to provide a traffic dynamic control method and system based on a deep learning network, which are used for solving the problem of inaccurate risk monitoring of a road traffic moving target in the prior art.
In order to achieve the above purpose, the present application provides the following technical solutions:
According to a first aspect of the present invention, the present invention claims a traffic dynamic control method based on a deep learning network, which is characterized in that the method comprises:
Collecting a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to a vertical motion speed or a horizontal motion speed, each risk threshold in any one of the dynamic target sequences in each sequence group is smaller than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the vertical motion speed, and each risk threshold in any one of the dynamic target sequences in each sequence group is larger than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the horizontal motion speed;
Performing a plurality of adjustments, each adjustment comprising:
The vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and the risk analysis end is used for determining the risk threshold value which is equal to that in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, and the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed, and the third risk threshold is larger than or equal to the fourth risk threshold, collecting the vertical motion dynamic target sequences positioned behind the vertical motion dynamic target sequences in the vertical motion sequence group as the vertical motion dynamic target sequences, wherein the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequences, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequences, and the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequences, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequences.
Further, the risk analysis end comprises an analysis grid, the analysis grid comprises k multiplied by k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each dynamic target sequence in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is less than or equal to k,
An mth vertical motion dynamic object in the vertical motion dynamic object sequence is sent to an nth parsing sub-grid along a horizontal motion direction in k parsing sub-grids positioned at a vertical motion object boundary line at an adjusted nth sending frequency, an xth horizontal motion dynamic object in the horizontal motion object sequence is sent to a yth parsing sub-grid along a vertical motion direction in k parsing sub-grids positioned at a horizontal motion object boundary line at an adjusted nth sending frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the parsing sub-grids sent by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is a direction pointing from the horizontal motion object boundary line to the inside of the parsing grid and perpendicular to the horizontal motion object boundary line, and the horizontal motion direction is a direction pointing from the vertical motion object to the inside of the parsing grid and perpendicular to the vertical motion object boundary line, m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
in a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
Further, the risk threshold value in the different dynamic targets in each sequence group is different, and each analysis sub-grid in the analysis grid is used for sending the vertical motion dynamic target to the next analysis sub-grid along the vertical motion direction and sending the horizontal motion dynamic target to the next analysis sub-grid along the horizontal motion direction in a scene that the risk threshold value in the vertical motion dynamic target is not equal to the risk threshold value in the horizontal motion dynamic target.
Further, the risk analysis end also comprises a screening grid, the screening grid comprises k screening sub-grids, the k screening sub-grids are respectively positioned behind the last analysis sub-grid of each row of the k rows of the analysis grid along the vertical movement direction along the horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to:
In a scene that the risk threshold in the vertical motion dynamic target is equal to the risk threshold in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction in a condition that the transmission frequency of the vertical motion dynamic target and the next transmission frequency of the horizontal motion dynamic target are received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold; or alternatively
Transmitting the analysis result to the next sub-grid along the horizontal movement direction at the next transmission frequency for receiving the analysis result;
The method further comprises the steps of: and under the scene that the vertical motion risk threshold is greater than or equal to the horizontal motion risk threshold, controlling the k screening sub-grids along the vertical motion direction to sequentially output the analysis results corresponding to the horizontal motion dynamic target sequence according to the transmission frequency.
According to a second aspect of the present invention, the present invention claims a traffic dynamic control system based on a deep learning network, comprising: the acquisition module and the analysis module are used for acquiring the data of the data,
The acquisition module is used for acquiring a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to a vertical motion speed or a horizontal motion speed, each risk threshold in any one of the dynamic target sequences in each sequence group is smaller than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the vertical motion speed, and each risk threshold in any one of the dynamic target sequences in each sequence group is larger than each risk threshold in the dynamic target sequences positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the horizontal motion speed;
the analysis module is used for carrying out multiple adjustment, and each adjustment comprises:
The vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and the risk analysis end is used for determining the risk threshold value which is equal to that in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, and the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed, and the third risk threshold is larger than or equal to the fourth risk threshold, collecting the vertical motion dynamic target sequences positioned behind the vertical motion dynamic target sequences in the vertical motion sequence group as the vertical motion dynamic target sequences, wherein the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequences, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequences, and the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequences, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequences.
Further, the risk analysis end comprises an analysis grid, the analysis grid comprises k multiplied by k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each dynamic target sequence in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is less than or equal to k,
An mth vertical motion dynamic object in the vertical motion dynamic object sequence is sent to an nth parsing sub-grid along a horizontal motion direction in k parsing sub-grids positioned at a vertical motion object boundary line at an adjusted nth sending frequency, an xth horizontal motion dynamic object in the horizontal motion object sequence is sent to a yth parsing sub-grid along a vertical motion direction in k parsing sub-grids positioned at a horizontal motion object boundary line at an adjusted nth sending frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the parsing sub-grids sent by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is a direction pointing from the horizontal motion object boundary line to the inside of the parsing grid and perpendicular to the horizontal motion object boundary line, and the horizontal motion direction is a direction pointing from the vertical motion object to the inside of the parsing grid and perpendicular to the vertical motion object boundary line, m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
in a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
Further, the risk threshold value in the different dynamic targets in each sequence group is different, and each analysis sub-grid in the analysis grid is used for sending the vertical motion dynamic target to the next analysis sub-grid along the vertical motion direction and sending the horizontal motion dynamic target to the next analysis sub-grid along the horizontal motion direction in a scene that the risk threshold value in the vertical motion dynamic target is not equal to the risk threshold value in the horizontal motion dynamic target.
Further, the risk analysis end also comprises a screening grid, the screening grid comprises k screening sub-grids, the k screening sub-grids are respectively positioned behind the last analysis sub-grid of each row of the k rows of the analysis grid along the vertical movement direction along the horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to: in a scene that the risk threshold in the vertical motion dynamic target is equal to the risk threshold in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction in a condition that the transmission frequency of the vertical motion dynamic target and the next transmission frequency of the horizontal motion dynamic target are received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold; or in the next sending frequency of the analysis result, sending the analysis result to the next sub-grid along the horizontal movement direction;
The analysis module is further configured to control the k screening sub-grids along the vertical motion direction to sequentially output the analysis result corresponding to the horizontal motion dynamic target sequence according to the transmission frequency in a scenario that the vertical motion risk threshold is greater than or equal to the horizontal motion risk threshold.
The application belongs to the technical field of road traffic, and particularly relates to a traffic dynamic control method and system based on a deep learning network, wherein a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence are acquired, each dynamic target sequence in each sequence group of a vertical motion sequence group and a horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, and the dynamic target sequences in each sequence group are sequentially detected according to vertical or horizontal motion speed; multiple times of adjustment are carried out, the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and equal risk threshold values in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are determined; the dynamic target sequences in each sequence group are sequentially detected according to the vertical movement speed. The application can effectively detect the running risk according to the dynamic target states of different levels of the vertical level, and has the function of reasonably monitoring and early warning the road traffic safety.
Drawings
FIG. 1 is a workflow diagram of a method for dynamically controlling traffic based on a deep learning network according to an embodiment of the present application;
Fig. 2 is a block diagram of a traffic dynamic control system based on a deep learning network according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application 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 application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms "vertical motion," "horizontal motion," "third" in this disclosure are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "vertical motion," "horizontal motion," and "third" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise. All directional indications (such as up, down, left, right, front, back … …) in the embodiments of the present application are merely used to explain the relative positional relationship between the components, the motion scene, etc. under a certain specific gesture (as shown in the drawings), and if the specific gesture changes, the directional indication changes accordingly. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or sub-grids is not limited to the listed steps or sub-grids, but may alternatively include steps or sub-grids not listed or may alternatively include other steps or sub-grids inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
According to a first embodiment of the present invention, the present invention claims a traffic dynamic control method based on a deep learning network, referring to fig. 1, the method includes:
Collecting a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed or the horizontal motion speed, each risk threshold in any one dynamic target sequence in each sequence group is smaller than each risk threshold in a dynamic target sequence positioned behind any one dynamic target sequence under a scene sequentially detected according to the vertical motion speed, and each risk threshold in any one dynamic target sequence in each sequence group is larger than each risk threshold in a dynamic target sequence positioned behind any one dynamic target sequence under a scene sequentially detected according to the horizontal motion speed;
Performing multiple times of adjustment, and transmitting the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence to a risk analysis end by each time of adjustment, wherein the risk analysis end is used for determining equal risk thresholds in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequence in each sequence group is sequentially detected according to the vertical motion speed, the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequence in each sequence group is sequentially detected according to the horizontal motion speed, the third risk threshold is larger than or equal to the fourth risk threshold, the vertical motion dynamic target sequence positioned behind the vertical motion dynamic target sequence in the vertical motion sequence group is collected and used as the vertical motion dynamic target sequence, the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequence, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequence, the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequence, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequence.
In this embodiment, the vertical moving dynamic target sequence and the horizontal moving dynamic target sequence are moving target sets of crisscross roads without traffic lights in the monitoring range in road traffic, and at least include moving targets such as buses, large trucks, small trucks, cars, electric vehicles, bicycles, pedestrians and the like in vertical and horizontal directions;
each dynamic target comprises a risk threshold, and a specific risk threshold indicates a risk grade value obtained by judging according to the current movement speed of the moving target and the situation of the moving target;
Each risk threshold value in any one of the dynamic target sequences in each sequence group is smaller than each risk threshold value in the dynamic target sequence located behind any one of the dynamic target sequences under the scene sequentially detected according to the vertical movement speed, which indicates that the accident risk of the front moving target of the traffic target located in the vertical direction is low and the accident risk (particularly rear-end collision risk) of the rear moving target is high.
Each risk threshold value in any one of the dynamic target sequences in each sequence group is larger than each risk threshold value in the dynamic target sequence positioned behind any one of the dynamic target sequences under the scene sequentially detected according to the horizontal movement speed, which indicates that the accident risk of the front moving target of the traffic target positioned in the horizontal direction is high and the accident risk (especially rear-end collision risk) of the rear moving target is low;
each risk threshold value in any one of the dynamic target sequences in each sequence group and each risk threshold value in the dynamic target sequence after any one of the dynamic target sequences can be set according to actual conditions.
Further, the risk analysis end comprises an analysis grid, the analysis grid comprises k multiplied by k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each dynamic target sequence in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is smaller than or equal to k,
The mth vertical motion dynamic object in the vertical motion dynamic object sequence is transmitted to the nth analytic sub-grid along the horizontal motion direction in the k analytic sub-grids positioned at the vertical motion object boundary line at the adjusted nth transmission frequency, the xth horizontal motion dynamic object in the horizontal motion object sequence is transmitted to the y analytic sub-grid along the vertical motion direction in the k analytic sub-grids positioned at the horizontal motion object boundary line at the adjusted nth transmission frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the analytic sub-grids transmitted by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is the direction from the horizontal motion object boundary line to the analytic grid interior and the vertical horizontal motion object boundary line, the horizontal motion direction is the direction from the vertical motion object boundary line to the analytic grid interior and the vertical motion object boundary line, and m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
In a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
Wherein, in this embodiment, the number of the at least one horizontal motion dynamic target sequence is a plurality,
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed and the vertical motion risk threshold value is larger than or equal to the horizontal motion risk threshold value, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed and the third risk threshold value is smaller than or equal to the fourth risk threshold value, collecting the horizontal motion dynamic target sequences positioned behind the horizontal motion dynamic target sequences in the horizontal motion sequence group as horizontal motion dynamic target sequences.
Further, the risk threshold value in different dynamic targets in each sequence group is different, and each analysis sub-grid in the analysis grid is used for sending the vertical motion dynamic target to the next analysis sub-grid along the vertical motion direction and sending the horizontal motion dynamic target to the next analysis sub-grid along the horizontal motion direction in the scene that the risk threshold value in the vertical motion dynamic target is not equal to the risk threshold value in the horizontal motion dynamic target.
Wherein in this embodiment, the risk analysis end further comprises an early warning triangle mesh, the early warning triangle mesh comprises k rows of early warning sub-meshes along the vertical movement direction, the number of the early warning sub-meshes along the vertical movement direction increases row by row,
Each of the plurality of alert sub-grids is for:
Receiving an analysis result output by a screening sub-grid before the pre-warning sub-grid along the horizontal movement direction, or receiving an analysis result output by a pre-warning sub-grid along a row in the vertical movement direction;
And at the next sending frequency of the analysis result, sending the analysis result to the early warning sub-grid of the next row along the vertical movement direction.
Further, the risk analysis end also comprises a screening grid, the screening grid comprises k screening sub-grids, the k screening sub-grids are respectively positioned behind the last analysis sub-grid of each of k rows of the analysis grid along the vertical movement direction along the horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to:
In a scene that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction under the condition that the next transmission frequency of the vertical motion dynamic target and the horizontal motion dynamic target is received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold value; or alternatively
At the next sending frequency of the received analysis result, sending the analysis result to the next sub-grid along the horizontal movement direction;
The method further comprises the steps of: and under the scene that the vertical motion risk threshold value is larger than or equal to the horizontal motion risk threshold value, controlling k screening sub-grids along the vertical motion direction to sequentially output analysis results corresponding to the horizontal motion dynamic target sequence according to the transmission frequency.
Wherein in this embodiment, the different risk thresholds correspond to different dynamic target sequences in the risk correlation table, the vertical motion dynamic target is used to indicate whether there is a relationship between the vertical motion dynamic target in the risk correlation table and at least one of the dynamic target sequences corresponding to the risk thresholds in the vertical motion dynamic target, the horizontal motion dynamic target is used to indicate whether there is a relationship between the horizontal motion dynamic target in the risk correlation table and at least one of the dynamic target sequences corresponding to the risk thresholds in the horizontal motion dynamic target,
Each analysis grid is further used for outputting analysis results in the scene that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, the analysis results are used for indicating query points in the risk association table, and the relationship scene between the query points and the two dynamic targets accords with the condition scene.
The vertical motion dynamic target further comprises a vertical motion accident probability value array of the vertical motion dynamic target corresponding to the risk threshold, the horizontal motion dynamic target further comprises a horizontal motion accident probability value array of the horizontal motion dynamic target corresponding to the risk threshold,
Each analysis sub-grid in the analysis grid is further used for respectively carrying out conditional analysis on each element of the vertical motion accident probability value array and the horizontal motion accident probability value array under the condition that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, the same element in the vertical motion accident probability value array and the horizontal motion accident probability value array corresponding to the equal risk threshold value corresponds to the same reference target in the dynamic target sequence corresponding to the equal risk threshold value, and the result of the conditional analysis of each element is used for indicating whether the relation scene between the reference target corresponding to the element and the two dynamic targets accords with the conditional scene.
In a scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, sequentially detecting the risk threshold value in each vertical motion dynamic target sequence from small to large; in the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal movement speed, the risk threshold value in each vertical movement dynamic target sequence is sequentially detected from large to small.
According to a second embodiment of the present invention, referring to fig. 2, the present invention claims a traffic dynamic control system based on a deep learning network, which is characterized by comprising: the acquisition module and the analysis module are used for acquiring the data of the data,
The acquisition module is used for acquiring a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed or the horizontal motion speed, each risk threshold in any one dynamic target sequence in each sequence group is smaller than each risk threshold in the dynamic target sequence behind any one dynamic target sequence under the scene sequentially detected according to the vertical motion speed, and each risk threshold in any one dynamic target sequence in each sequence group is larger than each risk threshold in the dynamic target sequence behind any dynamic target sequence under the scene sequentially detected according to the horizontal motion speed;
the analysis module is used for carrying out multiple adjustment, and each adjustment comprises:
the method comprises the steps that a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence are sent to a risk analysis end, and the risk analysis end is used for determining equal risk thresholds in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequence in each sequence group is sequentially detected according to the vertical motion speed, the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequence in each sequence group is sequentially detected according to the horizontal motion speed, the third risk threshold is larger than or equal to the fourth risk threshold, the vertical motion dynamic target sequence positioned behind the vertical motion dynamic target sequence in the vertical motion sequence group is collected and used as the vertical motion dynamic target sequence, the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequence, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequence, the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequence, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequence.
Further, the risk analysis end comprises an analysis grid, the analysis grid comprises k multiplied by k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each dynamic target sequence in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is smaller than or equal to k,
The mth vertical motion dynamic object in the vertical motion dynamic object sequence is transmitted to the nth analytic sub-grid along the horizontal motion direction in the k analytic sub-grids positioned at the vertical motion object boundary line at the adjusted nth transmission frequency, the xth horizontal motion dynamic object in the horizontal motion object sequence is transmitted to the y analytic sub-grid along the vertical motion direction in the k analytic sub-grids positioned at the horizontal motion object boundary line at the adjusted nth transmission frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the analytic sub-grids transmitted by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is the direction from the horizontal motion object boundary line to the analytic grid interior and the vertical horizontal motion object boundary line, the horizontal motion direction is the direction from the vertical motion object boundary line to the analytic grid interior and the vertical motion object boundary line, and m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
In a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
Wherein, in this embodiment, the number of the at least one horizontal motion dynamic target sequence is a plurality,
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed and the vertical motion risk threshold value is larger than or equal to the horizontal motion risk threshold value, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed and the third risk threshold value is smaller than or equal to the fourth risk threshold value, collecting the horizontal motion dynamic target sequences positioned behind the horizontal motion dynamic target sequences in the horizontal motion sequence group as horizontal motion dynamic target sequences.
Further, the risk threshold value in different dynamic targets in each sequence group is different, and each analysis sub-grid in the analysis grid is used for sending the vertical motion dynamic target to the next analysis sub-grid along the vertical motion direction and sending the horizontal motion dynamic target to the next analysis sub-grid along the horizontal motion direction in the scene that the risk threshold value in the vertical motion dynamic target is not equal to the risk threshold value in the horizontal motion dynamic target.
Wherein in this embodiment, the risk analysis end further comprises an early warning triangle mesh, the early warning triangle mesh comprises k rows of early warning sub-meshes along the vertical movement direction, the number of the early warning sub-meshes along the vertical movement direction increases row by row,
Each of the plurality of alert sub-grids is for:
Receiving an analysis result output by a screening sub-grid before the pre-warning sub-grid along the horizontal movement direction, or receiving an analysis result output by a pre-warning sub-grid along a row in the vertical movement direction;
And at the next sending frequency of the analysis result, sending the analysis result to the early warning sub-grid of the next row along the vertical movement direction.
Further, the risk analysis end also comprises a screening grid, the screening grid comprises k screening sub-grids, the k screening sub-grids are respectively positioned behind the last analysis sub-grid of each of k rows of the analysis grid along the vertical movement direction along the horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to: in a scene that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction under the condition that the next transmission frequency of the vertical motion dynamic target and the horizontal motion dynamic target is received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold value; or at the next sending frequency of the received analysis result, sending the analysis result to the next sub-grid along the horizontal movement direction;
The analysis module is further used for controlling the k screening sub-grids along the vertical movement direction to sequentially output analysis results corresponding to the horizontal movement dynamic target sequence according to the sending frequency in a scene that the vertical movement risk threshold is greater than or equal to the horizontal movement risk threshold.
Wherein in this embodiment, the different risk thresholds correspond to different dynamic target sequences in the risk correlation table, the vertical motion dynamic target is used to indicate whether there is a relationship between the vertical motion dynamic target in the risk correlation table and at least one of the dynamic target sequences corresponding to the risk thresholds in the vertical motion dynamic target, the horizontal motion dynamic target is used to indicate whether there is a relationship between the horizontal motion dynamic target in the risk correlation table and at least one of the dynamic target sequences corresponding to the risk thresholds in the horizontal motion dynamic target,
Each analysis grid is further used for outputting analysis results in a scene that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, the analysis results are used for indicating query points in the risk association table, and the relationship scene between the query points and the two dynamic targets accords with the condition scene; the vertical motion dynamic target further comprises a vertical motion accident probability value array of the vertical motion dynamic target corresponding to the risk threshold, the horizontal motion dynamic target further comprises a horizontal motion accident probability value array of the horizontal motion dynamic target corresponding to the risk threshold,
Each analysis sub-grid in the analysis grid is further used for respectively carrying out conditional analysis on each element of the vertical motion accident probability value array and the horizontal motion accident probability value array under the condition that the risk threshold value in the vertical motion dynamic target is equal to the risk threshold value in the horizontal motion dynamic target, the same element in the vertical motion accident probability value array and the horizontal motion accident probability value array corresponding to the equal risk threshold value corresponds to the same reference target in the dynamic target sequence corresponding to the equal risk threshold value, and the result of the conditional analysis of each element is used for indicating whether the relation scene between the reference target corresponding to the element and the two dynamic targets accords with the conditional scene.
In a scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, sequentially detecting the risk threshold value in each vertical motion dynamic target sequence from small to large; in the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal movement speed, the risk threshold value in each vertical movement dynamic target sequence is sequentially detected from large to small.
In the several embodiments provided in the present application, it should be understood that the disclosed system, system and method may be implemented in other manners. For example, the system embodiments described above are merely illustrative, e.g., the partitioning of sub-grids, merely a logical functional partitioning, and may be implemented in alternative ways, e.g., multiple sub-grids or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection through some interfaces, systems or sub-grids, which may be electrical, mechanical or otherwise.
In addition, each functional sub-grid in the embodiments of the present application may be integrated in one parsing sub-grid, or each sub-grid may exist physically separately, or two or more sub-grids may be integrated in one sub-grid. The integrated sub-grid can be realized in a hardware form or a software function sub-grid form. The foregoing is only the embodiments of the present application, and the patent scope of the application is not limited thereto, but is also covered by the patent protection scope of the application, as long as the equivalent structure or equivalent flow changes made by the description and the drawings of the application or the direct or indirect application in other related technical fields are adopted.
The embodiments of the application have been described in detail above, but they are merely examples, and the application is not limited to the above-described embodiments. It will be apparent to those skilled in the art that any equivalent modifications or substitutions to this application are within the scope of the application, and therefore, all equivalent changes and modifications, improvements, etc. that do not depart from the spirit and scope of the principles of the application are intended to be covered by this application.
Claims (8)
1. A traffic dynamic control method based on a deep learning network, the method comprising:
Collecting a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to a vertical motion speed or a horizontal motion speed, each risk threshold in any one of the dynamic target sequences in each sequence group is smaller than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the vertical motion speed, and each risk threshold in any one of the dynamic target sequences in each sequence group is larger than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the horizontal motion speed;
Performing a plurality of adjustments, each adjustment comprising:
The vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and the risk analysis end is used for determining the risk threshold value which is equal to that in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, and the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed, and the third risk threshold is larger than or equal to the fourth risk threshold, collecting the vertical motion dynamic target sequences positioned behind the vertical motion dynamic target sequences in the vertical motion sequence group as the vertical motion dynamic target sequences, wherein the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequences, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequences, and the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequences, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequences.
2. The traffic dynamic control method based on a deep learning network according to claim 1, wherein the risk analysis terminal comprises an analysis grid, the analysis grid comprises k×k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each of the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is less than or equal to k,
An mth vertical motion dynamic object in the vertical motion dynamic object sequence is sent to an nth parsing sub-grid along a horizontal motion direction in k parsing sub-grids positioned at a vertical motion object boundary line at an adjusted nth sending frequency, an xth horizontal motion dynamic object in the horizontal motion object sequence is sent to a yth parsing sub-grid along a vertical motion direction in k parsing sub-grids positioned at a horizontal motion object boundary line at an adjusted nth sending frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the parsing sub-grids sent by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is a direction pointing from the horizontal motion object boundary line to the inside of the parsing grid and perpendicular to the horizontal motion object boundary line, and the horizontal motion direction is a direction pointing from the vertical motion object to the inside of the parsing grid and perpendicular to the vertical motion object boundary line, m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
in a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
3. The traffic dynamic control method based on a deep learning network according to claim 2, wherein the risk threshold value in different dynamic targets in each sequence group is different, and each of the resolution sub-grids is configured to send the vertical motion dynamic target to the next resolution sub-grid in the vertical motion direction and send the horizontal motion dynamic target to the next resolution sub-grid in the horizontal motion direction in a scenario where the risk threshold value in the vertical motion dynamic target is not equal to the risk threshold value in the horizontal motion dynamic target.
4. The traffic dynamic control method based on deep learning network of claim 3, wherein said risk analysis terminal further comprises a screening grid comprising k screening sub-grids respectively located after the last analysis sub-grid of each of k rows of said analysis grid along said vertical movement direction along said horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to:
In a scene that the risk threshold in the vertical motion dynamic target is equal to the risk threshold in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction in a condition that the transmission frequency of the vertical motion dynamic target and the next transmission frequency of the horizontal motion dynamic target are received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold; or alternatively
Transmitting the analysis result to the next sub-grid along the horizontal movement direction at the next transmission frequency for receiving the analysis result;
The method further comprises the steps of: and under the scene that the vertical motion risk threshold is greater than or equal to the horizontal motion risk threshold, controlling the k screening sub-grids along the vertical motion direction to sequentially output the analysis results corresponding to the horizontal motion dynamic target sequence according to the transmission frequency.
5. A deep learning network-based traffic dynamic control system, comprising: the acquisition module and the analysis module are used for acquiring the data of the data,
The acquisition module is used for acquiring a vertical motion dynamic target sequence and a horizontal motion dynamic target sequence, wherein the vertical motion dynamic target sequence is an overspeed dynamic target sequence in a plurality of vertical motion dynamic target sequences of a vertical motion sequence group, the horizontal motion dynamic target sequence is an overspeed dynamic target sequence in at least one horizontal motion dynamic target sequence of a horizontal motion sequence group, each dynamic target sequence in each sequence group of the vertical motion sequence group and the horizontal motion sequence group comprises at least one dynamic target, each dynamic target comprises a risk threshold, the dynamic target sequences in each sequence group are sequentially detected according to a vertical motion speed or a horizontal motion speed, each risk threshold in any one of the dynamic target sequences in each sequence group is smaller than each risk threshold in a dynamic target sequence positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the vertical motion speed, and each risk threshold in any one of the dynamic target sequences in each sequence group is larger than each risk threshold in the dynamic target sequences positioned behind any one of the dynamic target sequences under a scene sequentially detected according to the horizontal motion speed;
the analysis module is used for carrying out multiple adjustment, and each adjustment comprises:
The vertical motion dynamic target sequence and the horizontal motion dynamic target sequence are sent to a risk analysis end, and the risk analysis end is used for determining the risk threshold value which is equal to that in the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence;
And under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the vertical motion speed, and the vertical motion risk threshold is smaller than or equal to the horizontal motion risk threshold, or under the scene that the dynamic target sequences in each sequence group are sequentially detected according to the horizontal motion speed, and the third risk threshold is larger than or equal to the fourth risk threshold, collecting the vertical motion dynamic target sequences positioned behind the vertical motion dynamic target sequences in the vertical motion sequence group as the vertical motion dynamic target sequences, wherein the vertical motion risk threshold is the largest risk threshold in the vertical motion dynamic target sequences, the horizontal motion risk threshold is the largest risk threshold in the horizontal motion dynamic target sequences, and the third risk threshold is the smallest risk threshold in the vertical motion dynamic target sequences, and the fourth risk threshold is the smallest risk threshold in the horizontal motion dynamic target sequences.
6. The traffic dynamic control system based on a deep learning network according to claim 5, wherein the risk analysis terminal comprises an analysis grid, the analysis grid comprises k×k analysis sub-grids, k is a positive integer, the number of at least one dynamic target in each of the vertical motion dynamic target sequence and the horizontal motion dynamic target sequence is less than or equal to k,
An mth vertical motion dynamic object in the vertical motion dynamic object sequence is sent to an nth parsing sub-grid along a horizontal motion direction in k parsing sub-grids positioned at a vertical motion object boundary line at an adjusted nth sending frequency, an xth horizontal motion dynamic object in the horizontal motion object sequence is sent to a yth parsing sub-grid along a vertical motion direction in k parsing sub-grids positioned at a horizontal motion object boundary line at an adjusted nth sending frequency, the vertical motion object boundary line is adjacent to the horizontal motion object boundary line, the parsing sub-grids sent by different dynamic objects in each dynamic object sequence are different, the vertical motion direction is a direction pointing from the horizontal motion object boundary line to the inside of the parsing grid and perpendicular to the horizontal motion object boundary line, and the horizontal motion direction is a direction pointing from the vertical motion object to the inside of the parsing grid and perpendicular to the vertical motion object boundary line, m, n, x, y is a positive integer;
Each analysis sub-grid in the analysis grid is used for determining whether the risk threshold value in the vertical motion dynamic target and the risk threshold value in the horizontal motion dynamic target which are transmitted to the analysis sub-grid at the same transmission frequency are equal;
in a scenario where k is greater than 1, each parsing sub-grid in the parsing grid is further configured to, upon receiving a next transmission frequency of the vertical motion dynamic object and the horizontal motion dynamic object, transmit the vertical motion dynamic object to a next parsing sub-grid along the vertical motion direction, and transmit the horizontal motion dynamic object to a next parsing sub-grid along the horizontal motion direction.
7. The deep learning network based traffic dynamics control system according to claim 6, wherein the risk threshold is different in different dynamic objects in each sequence group, each of the parsing sub-grids being configured to send the vertical motion dynamic object to a next one of the parsing sub-grids in the vertical motion direction and to send the horizontal motion dynamic object to a next one of the parsing sub-grids in the horizontal motion direction in a scenario where the risk threshold in the vertical motion dynamic object is not equal to the risk threshold in the horizontal motion dynamic object.
8. The traffic dynamic control system based on a deep learning network according to claim 7, wherein the risk analysis terminal further comprises a screening grid, the screening grid comprises k screening sub-grids, the k screening sub-grids are respectively located behind the last analysis sub-grid of each of k rows of the analysis grid along the vertical movement direction along the horizontal movement direction,
Each parsing sub-grid in the parsing grid is further configured to: in a scene that the risk threshold in the vertical motion dynamic target is equal to the risk threshold in the horizontal motion dynamic target, transmitting an analysis result of the analysis sub-grid to the next sub-grid along the horizontal motion direction in a condition that the transmission frequency of the vertical motion dynamic target and the next transmission frequency of the horizontal motion dynamic target are received, wherein the sub-grid is the analysis sub-grid or the screening sub-grid, and the analysis result comprises the equal risk threshold; or in the next sending frequency of the analysis result, sending the analysis result to the next sub-grid along the horizontal movement direction;
The analysis module is further configured to control the k screening sub-grids along the vertical motion direction to sequentially output the analysis result corresponding to the horizontal motion dynamic target sequence according to the transmission frequency in a scenario that the vertical motion risk threshold is greater than or equal to the horizontal motion risk threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410425775.7A CN118015844B (en) | 2024-04-10 | 2024-04-10 | Traffic dynamic control method and system based on deep learning network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410425775.7A CN118015844B (en) | 2024-04-10 | 2024-04-10 | Traffic dynamic control method and system based on deep learning network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118015844A true CN118015844A (en) | 2024-05-10 |
| CN118015844B CN118015844B (en) | 2024-06-11 |
Family
ID=90944980
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410425775.7A Active CN118015844B (en) | 2024-04-10 | 2024-04-10 | Traffic dynamic control method and system based on deep learning network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118015844B (en) |
Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190121348A1 (en) * | 2016-06-15 | 2019-04-25 | Strong Force Iot Portfolio 2016, Llc | Method and system of modifying a data collection trajectory for vehicles |
| CN111145545A (en) * | 2019-12-25 | 2020-05-12 | 西安交通大学 | Road traffic behavior unmanned aerial vehicle monitoring system and method based on deep learning |
| KR102122859B1 (en) * | 2020-03-03 | 2020-06-15 | (주)사라다 | Method for tracking multi target in traffic image-monitoring-system |
| KR102133913B1 (en) * | 2019-11-21 | 2020-07-15 | 문병국 | Digital display board system using object recognition based on deep learning for providing customized information of object and method for controlling thereof |
| CN111462488A (en) * | 2020-04-01 | 2020-07-28 | 北京工业大学 | Intersection safety risk assessment method based on deep convolutional neural network and intersection behavior characteristic model |
| CN112464546A (en) * | 2020-12-14 | 2021-03-09 | 上海交通大学设计研究总院有限公司 | Public space pedestrian flow motion risk discrimination method based on dynamic data analysis |
| KR20210072488A (en) * | 2019-12-09 | 2021-06-17 | 재단법인 자동차융합기술원 | Overspeed detection device that can output warning sound |
| CN114071015A (en) * | 2021-11-11 | 2022-02-18 | 浙江宇视科技有限公司 | Method, device, medium and equipment for determining linkage snapshot path |
| CN114724367A (en) * | 2022-03-29 | 2022-07-08 | 北京万集科技股份有限公司 | Intersection passage assisting method, apparatus, device, medium, and program product |
| CN115223107A (en) * | 2021-04-20 | 2022-10-21 | 南京大学 | Detection method for fit moving target detection frame |
| WO2022223952A1 (en) * | 2021-04-21 | 2022-10-27 | Zeta Specialist Lighting Limited | Traffic control at an intersection |
| CN116164760A (en) * | 2022-12-08 | 2023-05-26 | 北京航空航天大学 | Vehicle motion planning method based on risk field in multi-vehicle conflict scene |
| CN116386316A (en) * | 2022-12-08 | 2023-07-04 | 阿里云计算有限公司 | Traffic risk prediction method and device, electronic equipment and storage medium |
| CN116451862A (en) * | 2023-04-13 | 2023-07-18 | 腾讯科技（深圳）有限公司 | Model training method and device and traffic event occurrence probability evaluation method and device |
| KR102565227B1 (en) * | 2022-10-06 | 2023-08-28 | (주)지앤티솔루션 | Traffic safety device for providing risk prediction guidance and traffic safety system including the same |
| KR20230126927A (en) * | 2022-02-24 | 2023-08-31 | 정미정 | Traffic accident prevention system and its service method |
| WO2023179482A1 (en) * | 2022-03-25 | 2023-09-28 | 华为技术有限公司 | Image processing method, neural network training method and related device |
| CN116811854A (en) * | 2023-06-29 | 2023-09-29 | 北京航空航天大学杭州创新研究院 | Method and device for determining running track of automobile, automobile and storage medium |
| CN117037482A (en) * | 2023-08-08 | 2023-11-10 | 泰州视享智能科技有限公司 | Real-time road traffic running state sensing method, system, equipment and storage medium |
| CN117068134A (en) * | 2023-07-24 | 2023-11-17 | 东风汽车集团股份有限公司 | Crossroad steering automatic braking system and method based on vehicle speed |
| CN117141517A (en) * | 2023-09-13 | 2023-12-01 | 厦门大学 | Method for constructing vehicle track prediction model by combining data driving and knowledge guiding |
| CN117237418A (en) * | 2023-11-15 | 2023-12-15 | 成都航空职业技术学院 | Moving object detection method and system based on deep learning |
| KR102616571B1 (en) * | 2022-12-15 | 2023-12-21 | 한국건설기술연구원 | System and method for providing road traffic information based on image analysis using artificial intelligence |
| CN117373234A (en) * | 2023-09-28 | 2024-01-09 | 合肥工业大学 | Interactive prediction method for vehicle and vulnerable road user under safety theory |
| US20240112580A1 (en) * | 2022-09-30 | 2024-04-04 | Honda Motor Co., Ltd. | Traffic safety support system and storage medium |
-
2024
- 2024-04-10 CN CN202410425775.7A patent/CN118015844B/en active Active
Patent Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190121348A1 (en) * | 2016-06-15 | 2019-04-25 | Strong Force Iot Portfolio 2016, Llc | Method and system of modifying a data collection trajectory for vehicles |
| KR102133913B1 (en) * | 2019-11-21 | 2020-07-15 | 문병국 | Digital display board system using object recognition based on deep learning for providing customized information of object and method for controlling thereof |
| KR20210072488A (en) * | 2019-12-09 | 2021-06-17 | 재단법인 자동차융합기술원 | Overspeed detection device that can output warning sound |
| CN111145545A (en) * | 2019-12-25 | 2020-05-12 | 西安交通大学 | Road traffic behavior unmanned aerial vehicle monitoring system and method based on deep learning |
| KR102122859B1 (en) * | 2020-03-03 | 2020-06-15 | (주)사라다 | Method for tracking multi target in traffic image-monitoring-system |
| CN111462488A (en) * | 2020-04-01 | 2020-07-28 | 北京工业大学 | Intersection safety risk assessment method based on deep convolutional neural network and intersection behavior characteristic model |
| CN112464546A (en) * | 2020-12-14 | 2021-03-09 | 上海交通大学设计研究总院有限公司 | Public space pedestrian flow motion risk discrimination method based on dynamic data analysis |
| CN115223107A (en) * | 2021-04-20 | 2022-10-21 | 南京大学 | Detection method for fit moving target detection frame |
| WO2022223952A1 (en) * | 2021-04-21 | 2022-10-27 | Zeta Specialist Lighting Limited | Traffic control at an intersection |
| CN114071015A (en) * | 2021-11-11 | 2022-02-18 | 浙江宇视科技有限公司 | Method, device, medium and equipment for determining linkage snapshot path |
| KR20230126927A (en) * | 2022-02-24 | 2023-08-31 | 정미정 | Traffic accident prevention system and its service method |
| WO2023179482A1 (en) * | 2022-03-25 | 2023-09-28 | 华为技术有限公司 | Image processing method, neural network training method and related device |
| CN114724367A (en) * | 2022-03-29 | 2022-07-08 | 北京万集科技股份有限公司 | Intersection passage assisting method, apparatus, device, medium, and program product |
| US20240112580A1 (en) * | 2022-09-30 | 2024-04-04 | Honda Motor Co., Ltd. | Traffic safety support system and storage medium |
| KR102565227B1 (en) * | 2022-10-06 | 2023-08-28 | (주)지앤티솔루션 | Traffic safety device for providing risk prediction guidance and traffic safety system including the same |
| CN116386316A (en) * | 2022-12-08 | 2023-07-04 | 阿里云计算有限公司 | Traffic risk prediction method and device, electronic equipment and storage medium |
| CN116164760A (en) * | 2022-12-08 | 2023-05-26 | 北京航空航天大学 | Vehicle motion planning method based on risk field in multi-vehicle conflict scene |
| KR102616571B1 (en) * | 2022-12-15 | 2023-12-21 | 한국건설기술연구원 | System and method for providing road traffic information based on image analysis using artificial intelligence |
| CN116451862A (en) * | 2023-04-13 | 2023-07-18 | 腾讯科技（深圳）有限公司 | Model training method and device and traffic event occurrence probability evaluation method and device |
| CN116811854A (en) * | 2023-06-29 | 2023-09-29 | 北京航空航天大学杭州创新研究院 | Method and device for determining running track of automobile, automobile and storage medium |
| CN117068134A (en) * | 2023-07-24 | 2023-11-17 | 东风汽车集团股份有限公司 | Crossroad steering automatic braking system and method based on vehicle speed |
| CN117037482A (en) * | 2023-08-08 | 2023-11-10 | 泰州视享智能科技有限公司 | Real-time road traffic running state sensing method, system, equipment and storage medium |
| CN117141517A (en) * | 2023-09-13 | 2023-12-01 | 厦门大学 | Method for constructing vehicle track prediction model by combining data driving and knowledge guiding |
| CN117373234A (en) * | 2023-09-28 | 2024-01-09 | 合肥工业大学 | Interactive prediction method for vehicle and vulnerable road user under safety theory |
| CN117237418A (en) * | 2023-11-15 | 2023-12-15 | 成都航空职业技术学院 | Moving object detection method and system based on deep learning |
Non-Patent Citations (11)
| Title |
|---|
| "基于深度学习的高速公路交通事件检测研究", 中国优秀硕士学位论文全文数据库 工程科技II辑, 15 January 2023 (2023-01-15) * |
| CHEN YIFAN 等: "Conflict determination of without traffic control based on yolov3 and sort algorithm", 2023 IEEE INTERNATIONAL CONFERENCE ON SENSORS, ELECTRONICS AND COMPUTER ENGINEERING (ICSECE), 29 September 2023 (2023-09-29) * |
| 丁松涛;曲仕茹;: "基于深度学习的交通目标感兴趣区域检测", 中国公路学报, no. 09, 15 September 2018 (2018-09-15) * |
| 刘华平;李建民;胡晓林;孙富春;: "动态场景下的交通标识检测与识别研究进展", 中国图象图形学报, no. 05, 16 May 2013 (2013-05-16) * |
| 刘占文;王润民;赵祥模;: "基于车联网的交叉口行车事故预警方案及仿真", 计算机仿真, no. 06, 15 June 2016 (2016-06-15) * |
| 徐绍斌 等: "基于YOLOv5的车辆检测算法", 物联网技术, 19 April 2024 (2024-04-19) * |
| 朱彤;杨晓光;白玉;许建;马国胜;: "道路交叉口车辆碰撞动态风险估计方法", 中国公路学报, no. 04, 15 July 2010 (2010-07-15) * |
| 李娟: "城市交通***中行人交通视频检测的理论与方法", 万方学位论文库, 30 November 2011 (2011-11-30) * |
| 李昕蔚;丁正彦;尚岩峰;祝永新;汪辉;钟雪霞;田犁;黄尊恺;封松林;: "一种针对路口监控图像的区域分割方法", 计算机应用与软件, no. 03, 12 March 2020 (2020-03-12) * |
| 白思远: "基于深度学习的高速公路车辆识别与跟踪技术研究", 长安大学硕士学位论文, 14 April 2023 (2023-04-14) * |
| 袁琛: "基于交通冲突的高速公路实时安全分析方法", 中国优秀硕士学位论文全文数据库 工程科技I辑, 1 May 2023 (2023-05-01) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118015844B (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112462381B (en) | Multi-laser radar fusion method based on vehicle-road cooperation | |
| CN110971650B (en) | Collaborative sensing system and method based on V2X system and vehicle | |
| CN111695418A (en) | Method and system for safe driving based on road condition detection | |
| CN109696676B (en) | Effective obstacle target determination method and device and vehicle | |
| CN112633120B (en) | Model training method of intelligent roadside sensing system based on semi-supervised learning | |
| CN111308458A (en) | Vehicle speed estimation method based on vehicle-mounted millimeter wave radar | |
| CN105100218A (en) | Vehicle evaluation method, device and system based on automobile data recorder | |
| CN115691223A (en) | Cloud edge-end cooperation-based collision early warning method and system | |
| EP3467545A1 (en) | Object classification | |
| CN118015844B (en) | Traffic dynamic control method and system based on deep learning network | |
| CN112950929B (en) | All-weather real-time traffic information monitoring and QOS hierarchical control system and method | |
| CN112560546A (en) | Throwing behavior detection method and device and storage medium | |
| CN111580108B (en) | Vehicle-mounted millimeter wave radar occlusion detection method based on amplitude change analysis | |
| CN111406224B (en) | Target credibility determining method, target identifying method, system, vehicle and storage medium | |
| CN105684062A (en) | Method and apparatus for providing an event message with respect to an imminent event for a vehicle | |
| CN111582635A (en) | Multi-target processing method based on V2X | |
| CN115240470A (en) | NR-V2X-based weak traffic participant collision early warning system and method | |
| CN111522001B (en) | Automobile speed measurement method, automobile radar and automobile | |
| Schaudt et al. | Field evaluation of an enhanced rear signalling system for heavy trucks | |
| CN113960590A (en) | Racing fusion system and terminal | |
| CN115063980B (en) | Self-adaptive vehicle abnormal running detection method and device and terminal equipment | |
| CN114280609B (en) | 77GHz millimeter wave signal detection processing method and system | |
| CN112491814B (en) | Internet of vehicles networking intersection network attack detection method and system | |
| CN115914024B (en) | Method for detecting abnormal environment of intelligent network-connected automobile data | |
| CN114973663B (en) | Intelligent road side unit device based on edge calculation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |