A traffic incident detection method and device based on logical relationship

Abstract

The present invention discloses a traffic incident detection method and device based on logical relationship, including: obtaining target trajectory data and traffic incident detection data through millimeter wave radar; performing multi-dimensional association according to the target trajectory data and traffic incident detection data, generating associated event data corresponding to the traffic incident; constructing a traffic accident logic model based on the associated event data and the logical relationship between the associated events; inputting the detection data of the traffic incident to be analyzed into the traffic accident logic model, and outputting the alarm data corresponding to the traffic incident to be analyzed through the traffic accident logic model. The present invention can truly reflect the process of traffic accident occurrence based on logical relationship, improve the accuracy of alarm, and reduce the redundancy of event alarm.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a method and device for detecting traffic events based on logical relationships. Background Art

[0002] Millimeter wave radar is a high-precision measuring instrument in the field of transportation. It is widely used in the fields of vehicle trajectory detection, traffic statistics of road section cross sections, and event detection on highways. The types of event detection mainly include: spilled objects detection, vehicle speeding, vehicle slowing, vehicle reversing, occupying emergency lanes, vehicle illegal parking, road congestion, pedestrian intrusion, etc. However, the problems faced in the actual application of event detection are mainly the following two points:

[0003] (1) Accuracy issues in identifying various types of events, mainly including duplicate alarms and false alarms of an event.

[0004] (2) In practice, a traffic accident usually generates multiple types of event alarms. For example, a car accident can generally be decomposed into parking on the road + pedestrian intrusion, while road construction is composed of parking on the road + a large number of vehicles changing lanes at a distance + pedestrian intrusion, etc. This generates a large number of event alarms, resulting in heavy manual screening work. Summary of the invention

[0005] In order to solve at least one of the problems described in the above background technology, the present invention provides a traffic incident detection method based on logical relationship, comprising:

[0006] Through millimeter-wave radar, the trajectory data of the target and the detection data of traffic events are obtained;

[0007] Perform multi-dimensional association based on the trajectory data of the target and the detection data of the traffic event to generate associated event data corresponding to the traffic event;

[0008] Constructing a traffic accident logic model based on the associated event data and the logical relationship between the associated events;

[0009] The detection data of the traffic event to be analyzed is input into the traffic accident logic model, and the alarm data corresponding to the traffic event to be analyzed is output through the traffic accident logic model.

[0010] Preferably, the method includes: obtaining target trajectory data and traffic event detection data through millimeter wave radar, including:

[0011] Through millimeter-wave radar, the trajectory of the target on the road is detected to obtain the three-dimensional size data and location data;

[0012] Get the occurrence and end time of the traffic event.

[0013] Preferably, the method further comprises: after the step of obtaining the target trajectory data and the detection data of the traffic event, the method further comprises:

[0014] The trajectory data of the target and the detection data of the traffic event are solved, merged and filtered.

[0015] Preferably, the trajectory data of the target and the detection data of the traffic event are solved, merged and filtered, including:

[0016] According to the solving instructions, the acquired trajectory data of the target and the detection data of the traffic incident are solved;

[0017] Compare the detected data of traffic events after solving in time and space dimensions, and merge the events that meet the preset conditions into the same event;

[0018] Matching the solved target trajectory data with the detection data of the traffic event to obtain the detection data of the traffic event that matches the target trajectory data;

[0019] The interference data in the detection data of the solved traffic incidents is filtered through event rationality analysis and judgment.

[0020] Preferably, filtering is performed through event rationality analysis and judgment, including:

[0021] Obtaining the confidence of the detection data of the traffic event according to the continuity of the detection data of the traffic event, the trajectory information of the target and the dimension of the size information of the target;

[0022] The data whose confidence level is lower than the first preset threshold is filtered.

[0023] Preferably, multi-dimensional association is performed based on the trajectory data of the target and the detection data of the traffic event to generate associated event data corresponding to the traffic event, including:

[0024] Determine the spatial and temporal span of the event based on the trajectory data of the target;

[0025] Extracting feature values ​​of trajectory data of the target;

[0026] Determine the space and time of the event occurrence according to the change of the characteristic value;

[0027] Events that intersect in the two dimensions of space and time are associated to obtain associated event data corresponding to traffic events.

[0028] Preferably, based on the associated event data and the logical relationship between the associated events, a traffic accident logic model is constructed, including:

[0029] Sorting the associated event data according to the time of event occurrence to determine the logical relationship of the associated events;

[0030] The traffic accident logic model performs logical analysis and judgment on the associated event data, scores the rationality of the associated events, and obtains associated events whose rationality scores are greater than or equal to a second preset threshold;

[0031] The event with the highest score among the associated events is output as the associated event alarm result.

[0032] Preferably, it also includes:

[0033] Decompose related events into multiple basic events;

[0034] Among the multiple basic events, the basic event corresponding to the event with the highest score among the associated events is removed, and the alarm information of the remaining basic events is output.

[0035] Preferably, the rationality of the associated events is scored, including:

[0036] The conformity of the logical relationship between the basic events in the associated events is scored using quantifiable indicators.

[0037] The present invention also provides a traffic incident detection device based on logical relationship, comprising:

[0038] A data acquisition module is used to acquire target trajectory data and traffic event detection data through millimeter wave radar;

[0039] A correlation event generation module, used to perform multi-dimensional correlation based on the trajectory data of the target and the detection data of the traffic event, and generate correlation event data corresponding to the traffic event;

[0040] A logic model building module, used to build a traffic accident logic model based on the associated event data and the logical relationship between the associated events;

[0041] The alarm data acquisition module is used to input the detection data of the traffic event to be analyzed into the traffic accident logic model, and output the alarm data corresponding to the traffic event to be analyzed through the traffic accident logic model.

[0042] The present invention provides a traffic incident detection method and device based on logical relationships, which can more realistically reflect the accident process through the logical relationships between basic events, provide more accurate alarms, reduce the redundancy of incident alarms, and alleviate the workload of manual screening. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1It is a flow chart of a traffic incident detection method based on logical relationship provided by an exemplary embodiment of the present invention;

[0044] Figure 2 It is a schematic diagram of a flow chart of performing correlation analysis on traffic event data provided by an exemplary embodiment of the present invention;

[0045] Figure 3 It is a structural schematic diagram of a millimeter wave radar traffic incident detection device based on logical relationship provided by an exemplary embodiment of the present invention;

[0046] Figure 4 is a structural diagram of an electronic device provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0047] Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention, so the present invention is not limited to the specific implementation disclosed below.

[0048] Figure 1 A flow chart of a traffic incident detection method based on logical relationship provided by the present invention is shown, and the method comprises the following steps:

[0049] Step S101, obtaining target trajectory data and traffic event detection data through millimeter wave radar.

[0050] In an embodiment of the present invention, a millimeter wave radar is used to detect the trajectory of a target on the road, obtain the three-dimensional size data and the position data, and obtain the time when the traffic event occurs and the end time.

[0051] Furthermore, the target on the road may be a moving car. Then, the trajectory data of the target and the detection data of the traffic event are solved, merged and filtered. Specifically, it includes:

[0052] According to the solving instructions, the acquired trajectory data of the target and the detection data of the traffic event are solved; the solved detection data of the traffic event are compared in the time and space dimensions, and the events that meet the preset conditions are merged into the same event; the solved trajectory data of the target and the detection data of the traffic event are matched to obtain the detection data of the traffic event that matches the trajectory data of the target; the interference data in the solved detection data of the traffic event is filtered through event rationality analysis and judgment.

[0053] The analysis and judgment of the rationality of the event mainly includes obtaining the confidence of the detection data of the traffic event based on the continuity of the detection data of the traffic event, the trajectory information of the target and the size information dimension of the target; and filtering the data with the confidence lower than the first preset threshold.

[0054] The description of the confidence of the detection data of the traffic incident includes the ability of the same type of events to be confirmed continuously, the validity of the trajectory contained in each event data, and the validity of the detection of the three dimensions of the target. Among them, the ability of the same type of events to be confirmed continuously is described as the number of times the event is detected continuously / 10, and if the number of times the event is detected continuously is greater than 10, the value is 10; the validity of the trajectory contained in each event data is described as the number of location information contained in the trajectory data / 100, and if the number of location information is greater than 100, the value is 100; the validity of the detection of the three dimensions of the target is described as 1 when the event data carries three-dimensional data, otherwise it is 0.

[0055] Furthermore, according to the analysis dimension of the confidence of the detection data of the traffic incident, different difference weights are assigned to each dimension, and the confidence of the incident is calculated, including: the weight coefficient A is assigned to the ability of the same target and the same type of incident to be continuously confirmed, the weight coefficient B is assigned to the validity of the trajectory contained in the data of each incident, and the weight coefficient C is assigned to the validity of the detection of the three dimensions of the target, where A+B+C=100; the event confidence value is calculated by the following formula: event confidence value=A*the ability of the same target and the same type of incident to be continuously confirmed+B*the validity of the trajectory contained in the data of each incident+C*the validity of the detection of the three dimensions of the target.

[0056] Step S102 , performing multi-dimensional association based on the trajectory data of the target and the detection data of the traffic event to generate associated event data corresponding to the traffic event.

[0057] After obtaining the detection data of a traffic event with a high confidence level, determine the space and time span of the event based on the trajectory data of the target; extract the characteristic value of the trajectory data of the target; determine the space and time of the event based on the change of the characteristic value; associate the events that intersect in the two dimensions of space and time to obtain the associated event data corresponding to the traffic event. Furthermore, the space and time span before and after the event is improved based on the trajectory data of the target. First, push forward a period of time based on the start time of the event as the start time; secondly, extract characteristic values ​​from the target trajectory data based on this as the starting point, mainly including speed and acceleration information in the direction of motion and in the direction perpendicular to the motion; finally, determine the start time and final time of the event as well as the start position and end position based on the change of the characteristic value. Associate the events that intersect in the two dimensions of space and time and package them into associated data.

[0058] In the embodiment of the present invention, taking into account the actual application process of traffic event detection, multiple basic event alarms are often generated for a traffic accident or road construction. By performing spatiotemporal correlation analysis on all basic events and combining multiple basic events into one associated event, the alarm accuracy can be improved and the accident process can be fully reproduced.

[0059] For example: when a car accident occurs, the basic event of a vehicle parking on the road is detected first, and then one or more events such as pedestrian intrusion, spilled objects, vehicle congestion, and vehicle slowdown are detected successively. According to the spatiotemporal information of each event, the events with overlapping and intersecting spatiotemporal ranges are packaged into a related event.

[0060] Figure 2 The specific process of correlation analysis of traffic event data is shown in FIG. Figure 2 As shown in the figure, the specific process of data association and analysis includes the following steps:

[0061] 1) According to the starting time of the event, move forward a period of time as the starting time, taking 5 seconds as an example.

[0062] 2) Taking the time starting point determined in step 1 as the starting point, extract feature values ​​from the target trajectory data matched in the event, mainly including the speed and acceleration information in the direction of motion and perpendicular to the direction of motion.

[0063] 3) Determine the start time and final time of the event as well as the start position and end position based on the change of the characteristic value, thereby determining the time and space scope of the event.

[0064] 4) Pack events with overlapping and intersecting time and space ranges into one associated event.

[0065] Step S103: constructing a traffic accident logic model based on the associated event data and the logical relationship between the associated events.

[0066] The associated event data are sorted according to the time of event occurrence to determine the logical relationship of the associated events; the traffic accident logic model performs logical analysis and judgment on the associated event data, scores the rationality of the associated events, and obtains associated events with a rationality score greater than or equal to a second preset threshold; the event with the highest score among the associated events is output as the associated event alarm result. The associated events are decomposed into multiple basic events; among the multiple basic events, the basic events corresponding to the event with the highest score among the associated events are removed, and the alarm information of the remaining basic events is output. The basic event is the basic unit of the associated event, and the associated event is composed of basic events that intersect in the trajectory data of the target and the detection data of the traffic event in the two dimensions of space and time.

[0067] The conformity of the logical relationship between the basic events in the associated events is scored using quantifiable indicators.

[0068] Furthermore, it is assumed that the traffic accident logic model contains three basic events, A, B, and C. That is, when a related event occurs, the above basic events may be generated. A logic model is created based on the logical relationship between the above basic events. The logical relationship includes ABC, ACB, BAC, BCA, CBA, and CAB. The score of the possibility of each logical relationship is preset through data statistical analysis in the actual application process, and the value range is 0~100. The basic events in all related events are sorted by time to determine the logical relationship of the basic events in the related events, and then matched with the logical relationship in the logic model. Then, a rationality evaluation is given for each related event based on the matching result, and the related event with the highest rationality evaluation is selected, and the alarm result is output. The rationality evaluation of the related event mainly gives a quantifiable indicator for the conformity of the logical relationship of the basic events in the related event.

[0069] In the embodiment of the present invention, the present invention takes into account the actual application process of traffic incident detection, and more realistically reflects the accident process through the logical relationship between basic events, making the alarm more accurate, reducing the redundancy of event alarms, and alleviating the workload of manual screening.

[0070] Step S104, inputting the detection data of the traffic event to be analyzed into the traffic accident logic model, and outputting the alarm data corresponding to the traffic event to be analyzed through the traffic accident logic model.

[0071] Furthermore, based on the alarm data, an alarm message is issued, and the alarm message is tracked, processed, and eliminated.

[0072] Figure 3 A traffic event detection device 300 based on logical relationship is provided by an exemplary embodiment of the present invention, comprising:

[0073] The data acquisition module 310 is used to acquire the trajectory data of the target and the detection data of the traffic event through the millimeter wave radar;

[0074] The associated event generation module 320 is used to perform multi-dimensional association based on the trajectory data of the target and the detection data of the traffic event to generate associated event data corresponding to the traffic event;

[0075] A logic model building module 330, for building a traffic accident logic model based on the associated event data and the logical relationship between the associated events;

[0076] The alarm data acquisition module 340 is used to input the detection data of the traffic event to be analyzed into the traffic accident logic model, and output the alarm data corresponding to the traffic event to be analyzed through the traffic accident logic model.

[0077] Figure 4 is a structure of an electronic device provided by an exemplary embodiment of the present invention, such as Figure 4 As shown, the electronic device includes one or more processors 31 and a memory 42.

[0078] The processor 41 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0079] The memory 42 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 41 may run the program instructions to implement the methods of the software programs of the various embodiments of the present invention described above and / or other desired functions. In one example, the electronic device may also include: an input device 43 and an output device 44, which are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0080] In addition, the input device 43 may also include, for example, a keyboard, a mouse, etc.

[0081] The output device 44 can output various information to the outside, and can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto.

[0082] Of course, to simplify, Figure 4 Only some of the components related to the present invention in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application conditions, the electronic device may also include any other appropriate components.

[0083] In addition to the above-mentioned methods and devices, an embodiment of the present invention may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present invention described in the above-mentioned "Exemplary Method" section of this specification.

[0084] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the embodiments of the present invention, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0085] In addition, an embodiment of the present invention may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present invention described in the above “Exemplary Method” section of this specification.

[0086] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, system or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0087] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details disclosed above are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0088] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0089] The block diagrams of the devices, systems, equipment, and systems involved in the present invention are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, systems, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with it.

[0090] The method and apparatus of the present invention may be implemented in many ways. For example, the method and apparatus of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, firmware. The above order of steps for the method is only for illustration, and the steps of the method of the present invention are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present invention may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Thus, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0091] It should also be noted that in the system, device and method of the present invention, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. The above description of the disclosed aspects is provided to enable any technician in the field to make or use the present invention. Various modifications to these aspects are very obvious to those skilled in the art, and the general principles defined here can be applied to other aspects without departing from the scope of the present invention. Therefore, the present invention is not intended to be limited to the aspects shown here, but in accordance with the widest range consistent with the principles and novel features disclosed here.

[0092] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present invention to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A traffic incident detection method based on logical relationship, characterized in that: include: Through millimeter-wave radar, the trajectory data of the target and the detection data of traffic events are obtained; Solving, merging and filtering the trajectory data of the target and the detection data of the traffic event; Perform multi-dimensional association based on the trajectory data of the target and the detection data of the traffic event to generate associated event data corresponding to the traffic event; Constructing a traffic accident logic model based on the associated event data and the logical relationship between the associated events; Inputting the detection data of the traffic event to be analyzed into the traffic accident logic model, and outputting the alarm data corresponding to the traffic event to be analyzed through the traffic accident logic model; Solving, merging and filtering the target's trajectory data and the traffic event's detection data, including: According to the solving instructions, the acquired trajectory data of the target and the detection data of the traffic incident are solved; Compare the detected data of traffic events after solving in time and space dimensions, and merge the events that meet the preset conditions into the same event; Matching the solved target trajectory data with the detection data of the traffic event to obtain the detection data of the traffic event that matches the target trajectory data; The interference data in the detected data of the solved traffic incidents is filtered out through event rationality analysis and judgment; Through event rationality analysis and judgment, filtering processing is performed, including: Obtaining the confidence of the detection data of the traffic event according to the continuity of the detection data of the traffic event, the trajectory information of the target and the dimension of the size information of the target; The data whose confidence level is lower than the first preset threshold is filtered.

2. The method according to claim 1, characterized in that include: Through millimeter-wave radar, the target trajectory data and traffic event detection data are obtained, including: Through millimeter-wave radar, the trajectory of the target on the road is detected to obtain the three-dimensional size data and location data; Get the occurrence and end time of the traffic event.

3. The method according to claim 1, characterized in that According to the trajectory data of the target and the detection data of the traffic event, multi-dimensional association is performed to generate associated event data corresponding to the traffic event, including: Determine the spatial and temporal span of the event based on the trajectory data of the target; Extracting feature values ​​of trajectory data of the target; Determine the space and time of the event occurrence according to the change of the characteristic value; Events that intersect in the two dimensions of space and time are associated to obtain associated event data corresponding to traffic events.

4. The method according to claim 1, characterized in that Based on the associated event data and the logical relationship between the associated events, a traffic accident logic model is constructed, including: Sorting the associated event data according to the time of event occurrence to determine the logical relationship of the associated events; The traffic accident logic model performs logical analysis and judgment on the associated event data, scores the rationality of the associated events, and obtains associated events whose rationality scores are greater than or equal to a second preset threshold; The event with the highest score among the associated events is output as the associated event alarm result.

5. The method according to claim 4, characterized in that Also includes: Decompose related events into multiple basic events; Among the multiple basic events, the basic event corresponding to the event with the highest score among the associated events is removed, and the alarm information of the remaining basic events is output.

6. The method according to claim 4, characterized in that Score the plausibility of the associated events, including: The conformity of the logical relationship between the basic events in the associated events is scored using quantifiable indicators.

7. A traffic incident detection device based on logical relationship, characterized in that: include: A data acquisition module is used to acquire target trajectory data and traffic event detection data through millimeter wave radar; Solving, merging and filtering the trajectory data of the target and the detection data of the traffic event; A correlation event generation module, used to perform multi-dimensional correlation based on the trajectory data of the target and the detection data of the traffic event, and generate correlation event data corresponding to the traffic event; A logic model building module, used to build a traffic accident logic model based on the associated event data and the logical relationship between the associated events; An alarm data acquisition module, which inputs the detection data of the traffic event to be analyzed into the traffic accident logic model, and outputs the alarm data corresponding to the traffic event to be analyzed through the traffic accident logic model; Solving, merging and filtering the target's trajectory data and the traffic event's detection data, including: According to the solving instructions, the acquired trajectory data of the target and the detection data of the traffic incident are solved; Compare the detected data of traffic events after solving in time and space dimensions, and merge the events that meet the preset conditions into the same event; Matching the solved target trajectory data with the detection data of the traffic event to obtain the detection data of the traffic event that matches the target trajectory data; The interference data in the detected data of the solved traffic incidents is filtered out through event rationality analysis and judgment; Through event rationality analysis and judgment, filtering processing is performed, including: Obtaining the confidence of the detection data of the traffic event according to the continuity of the detection data of the traffic event, the trajectory information of the target and the dimension of the size information of the target; The data whose confidence level is lower than the first preset threshold is filtered.

Images