KR20220150479A - Apparatus and method for providing traffic information using security CCTV images - Google Patents

Abstract

Provided are a device and method for providing road condition prediction information. The device for providing road condition prediction information comprises: an object recognition unit recognizing objects approaching an intersection in which traffic lights are not installed; a movement trajectory prediction unit predicting a movement trajectory of each of the recognized objects; a conflict possibility evaluation unit using the movement trajectory of each of the objects predicted by the movement trajectory prediction unit, for any one of the recognized objects (hereinafter referred to as 'target object'), and evaluating possibility of conflict with the remaining objects among the recognized objects for each road (hereinafter, referred to as a 'road that can be progressed') through which the target object can proceed at an intersection; and a road condition prediction information providing unit generating road condition prediction information which includes an evaluation result of the possibility of conflict evaluated by the road that can be progressed of the target object in the conflict possibility evaluation unit, and providing the same to a road condition information board installed at the intersection. Accordingly, traffic accidents can be prevented in advance.

Description

Apparatus and method for providing traffic information using security CCTV images}

The present invention relates to an apparatus and method for providing road condition prediction information, and more particularly, to a road capable of providing information for preventing traffic accidents with other vehicles by using a security CCTV near an intersection where traffic lights are not installed. It relates to a situation prediction information providing device and method.

One of the causes of traffic accidents that occur at alley intersections and side roads is that drivers cannot secure sight distance due to unpredictable intersections. In addition, when approaching an intersection without a traffic light with poor visibility, the driver often experiences the inconvenience of driving more slowly than necessary due to the lack of securing the visibility in the direction in which he is driving, or having to back up due to the impossibility of crossing with the vehicle he encounters after driving. .

On the other hand, CCTV (Closed-Circuit TeleVision) for crime prevention is installed not only on large roads but also around side roads operated without signals, and can take pictures of surrounding access roads to prevent crime. This seems to be because it is necessary to check the direction of the criminal's escape due to the nature of crime prevention. In addition, the controller of the CCTV for crime prevention is installed on the support and transmits the video taken and collected by the CCTV for crime prevention to the control center, and the control center analyzes the video to check the situation related to crime prevention (theft, robbery, assault, etc.).

However, since CCTVs for crime prevention installed around roads where it is difficult to secure visibility, such as back roads, are used only for crime prevention, it is necessary to develop technology that can provide road information to drivers who have difficulty securing visibility using CCTVs for crime prevention. do.

Domestic Patent Registration No. 10-2197199 (December 24, 2020)

In order to solve the above-described problems, the technical problem to be achieved by the present invention is to provide information on vehicles or pedestrians approaching from other access roads to drivers or pedestrians who have difficulty securing visibility in the vicinity of intersections where traffic lights are not installed. The purpose of this study is to propose an apparatus and method for providing road condition prediction information that can be analyzed and provided.

The problems of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

As a means for solving the above technical problem, according to an embodiment of the present invention, an apparatus for providing road condition prediction information includes: an object recognition unit recognizing objects approaching an intersection where traffic lights are not installed; a movement trajectory predictor for predicting a movement trajectory of each of the recognized objects; With respect to any one of the recognized objects (hereinafter referred to as 'target object'), a road (hereinafter referred to as 'target object') on which the target object can proceed at an intersection using the movement trajectory of each of the objects predicted by the movement trajectory prediction unit a conflict possibility evaluation unit that evaluates a possibility of conflict with the remaining objects among the recognized objects for each object; and a road condition prediction information providing unit that generates road situation prediction information including a conflict potential evaluation result evaluated for each progressability degree of the target object in the conflict possibility evaluation unit and provides it to a road situation information board installed at an intersection. do.

The conflictability evaluation unit may include: an object prediction unit that analyzes the movement trajectories of the remaining objects and predicts an object to be approached first (hereinafter, referred to as an 'approaching object') for each progressability of the target object; a conflict possibility determination unit for determining the possibility of collision between the approaching object predicted by the object prediction unit and the target object according to the progressability of the target object; And if it is determined that there is a possibility of conflict in the conflict possibility determination unit, a conflict direction weight check unit that checks a conflict possibility weight (hereinafter referred to as 'conflict direction weight) according to the driving direction of the target object and the approaching object; .

The conflict possibility evaluation unit, if it is determined that there is a possibility of conflict in the conflict possibility determination unit, checks the conflict area weight to check the conflict probability weight set in the area where conflict is predicted to occur (hereinafter referred to as 'conflict area weight'). part; further includes.

The road condition prediction information providing unit is based on the probability of conflict determined by the probability of conflict determination unit, the weight of the probability of conflict according to the driving direction determined by the weight determination unit for conflict direction, and the weight of the conflict area determined by the weight determination unit of conflict area. a blinking frequency calculation unit for calculating the blinking frequency for displaying the road condition prediction information according to the degree of progress of the target object; Information generation unit for generating road condition prediction information including at least one of the type of the approaching object, movement speed, movement trajectory, possibility of collision, and an area in which a collision with the approaching object is predicted for each degree of progress of the target object. ; and an information provision unit configured to provide the road condition information board with road condition prediction information generated for each degree of progress of the target object, wherein the road condition information plate provides information about the target object based on the calculated flashing frequency. Road condition prediction information is displayed for each possible road.

The direction of the possible road for providing the road condition prediction information is the intersection In the case of a +-shaped intersection, a right turn direction and a left turn direction are included based on the current driving direction of the target object, and when the intersection is a T-shaped intersection, among the right turn direction and the left turn direction based on the current driving direction of the target object contains at least one

Meanwhile, according to another embodiment of the present invention, a method for providing road condition prediction information includes: (A) recognizing, by an electronic device, objects approaching an intersection where traffic lights are not installed; (B) predicting, by the electronic device, a movement trajectory of each of the approaching objects; (C) With respect to any one of the objects recognized in step (A) (hereinafter referred to as 'target object'), using the movement trajectory of each of the objects predicted in step (B), the target object Evaluating a possibility of conflict with the remaining objects among the recognized objects for each road (hereinafter referred to as a 'progressability road') at the intersection; and (D) generating, by the electronic device, road condition prediction information including the conflictability evaluation result evaluated in step (C) and providing it to a road condition information board installed at an intersection.

In the step (C), (C1) the electronic device analyzes the movement trajectories of the remaining objects and predicts an object to be approached first (hereinafter referred to as an 'approaching object') for each degree of progress of the target object. step; (C2) determining the probability of conflict between the approach object predicted in step (C1) and the target object according to the progressability of the target object; and (C3) if it is determined that there is a possibility of conflict in the step (C2), checking a conflict possibility weight (hereinafter referred to as 'conflict direction weight) according to the driving direction of the target object and the approaching object.

In the step (C), if it is determined that there is a possibility of conflict in the step (C4) and the step (C2) after the step (C3), the weight set in the area in which conflict is expected to occur (hereinafter referred to as 'conflict area weight') It further includes; confirming that).

The step (D) is performed on the target object based on the possibility of conflict determined in the step (D1), the weight of the conflict direction determined in the step (C3), and the weight of the conflict area determined in the step (C4). Calculating a blinking frequency for displaying road condition prediction information for each progressable road; (D2) generating road situation prediction information including at least one of the type of the approaching object, movement speed, movement trajectory, possibility of collision, and an area in which a collision with the approaching object is predicted for each degree of progress of the target object step; and (D3) transmitting the road condition prediction information generated for each degree of progress of the target object in step (D2) and the blinking frequency calculated in step (D1) to the road condition information board.

On the other hand, according to another embodiment of the present invention, the road condition prediction information system analyzes images taken by closed-circuit televisions (CCTVs) installed on roads connected to intersections, and the roads on which vehicles can proceed at intersections (hereinafter a road condition prediction information providing device that evaluates a possibility of conflict with other objects for each degree of possibility of progression and generates road condition prediction information including a result of evaluating the possibility of conflict for each degree of possibility of progress; and a road condition information board installed at an intersection to display road condition prediction information generated by the road condition prediction information providing device for each possible road of the vehicle, and to allow the driver of the vehicle to view the road condition prediction information. do.

When the intersection is a +-shaped intersection, the road condition information board includes a right information board providing road condition prediction information corresponding to a right turn of the target object and a road condition corresponding to a left turn of the target object. Left information boards providing prediction information may be included as many as the number of roads connected to the intersection.

When the intersection is a T-shaped intersection, the road condition information board includes road condition prediction information corresponding to when the target object makes a right turn from the main road to a sub-road and when the target object makes a left turn from the main road to a sub-road. A double-sided information board providing road condition prediction information corresponding to both sides, a right information board providing road condition prediction information corresponding to the case where the target object makes a right turn from a secondary road to a main road, and the target object is a secondary road It may include a left information board that provides road condition prediction information corresponding to a left turn to the main road.

According to the present invention, road condition information (vehicle access, presence of pedestrians, two-wheeled vehicles, etc., approaching speed, location, etc.), conflict possibility, conflict expectation Traffic accidents can be prevented in advance by providing areas, etc.

In addition, according to the present invention, active efforts to prevent traffic accidents by providing road conditions for each access route to moving objects such as drivers, pedestrians, and two-wheeled vehicles at intersections with poor visibility, such as back roads, and predicting and displaying the possibility of conflict can induce

In addition, according to the present invention, by using the frequency of traffic accidents occurring near intersections where traffic lights are not installed or by traffic accidents in specific areas, road condition prediction information is generated and traffic accident risk is calculated to provide information considering regional characteristics. can

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a diagram showing a road information providing system for preventing traffic accidents according to an embodiment of the present invention;
2 is a diagram showing a +-shaped intersection;
3 is a diagram showing a T-shaped intersection;
4 is an exemplary view of a first road condition information board 110 installed at a +-shaped intersection;
5 is an exemplary view of a second road condition information board 120 installed at a T-shaped intersection;
6 is a block diagram showing the road condition prediction information providing device 200 shown in FIG. 1;
7 is a diagram showing conflicting areas of a +-shaped intersection;
8 is a diagram showing conflict areas of a T-shaped intersection;
9 is an exemplary view showing the movement trajectories of a vehicle and a two-wheeled vehicle running at a T-shaped intersection;
10 is a block diagram showing the conflict possibility evaluation unit 253 shown in FIG. 6 in detail;
11 is a block diagram showing in detail the road condition prediction information providing unit 254 shown in FIG. 6;
12 is a flowchart illustrating a method for providing road condition prediction information according to an embodiment of the present invention;
13 is a flowchart showing step S1240 of FIG. 12, and
FIG. 14 is a flowchart illustrating step S1250 of FIG. 12 .

The above objects, other objects, features and advantages of the present invention will be easily understood through the following preferred embodiments in conjunction with the accompanying drawings.

In some cases, it is mentioned in advance that parts that are commonly known in describing the invention and are not greatly related to the invention are not described in order to prevent confusion for no particular reason in explaining the present invention.

In this specification, when terms such as first and second are used to describe components, these components should not be limited by these terms. These terms are only used to distinguish one component from another. Embodiments described and illustrated herein also include complementary embodiments thereof.

Further, when a first component is referred to as being operated or executed on (ON) a second component, the first component is operated or executed in an environment in which the second component is operated or executed, or with the second component. It should be understood that it is operated or executed through direct or indirect interaction.

Where an element, component, device, or system is referred to as including a component consisting of a program or software, even if not explicitly stated otherwise, that element, component, device, or system means that the program or software executes or operates. It should be understood that it includes hardware (eg, memory, CPU, etc.) or other programs or software (eg, operating system or driver required to drive hardware) required to do so.

In addition, it should be understood that, unless otherwise specified, the component may be implemented in any form of software, hardware, or both software and hardware.

In addition, terms used in this specification are for describing embodiments and are not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. The terms 'comprises' and/or 'comprising' used in the specification do not exclude the presence or addition of one or more other elements.

In addition, in this specification, terms such as 'unit', 'server', 'system', 'device', etc. may be intended to refer to functional and structural combinations of hardware and software driven by or for driving the hardware. have. For example, the hardware herein may be a data processing device including a CPU or other processor. Also, software driven by hardware may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

Hereinafter, with reference to the accompanying drawings for the specific technical content to be carried out in the present invention will be described in detail.

Each component of the road information providing system for preventing traffic accidents shown in FIGS. 1, 6, 10, and 11 indicates that it may be functionally and/or logically separated, and each component must be separated physically. An average expert in the art of the present invention can easily infer that it does not mean that it is classified as a device or written as a separate code.

In addition, the apparatus 200 for providing road condition prediction information according to an embodiment of the present invention is, for example, a desktop PC (Personal Computer), a server, a laptop PC (Laptop PC), a netbook computer (Netbook Computer), etc. It can be any one of all electronic devices that can be executed.

1 is a diagram illustrating a road information providing system for preventing traffic accidents according to an embodiment of the present invention.

Referring to FIG. 1 , a road information providing system for preventing traffic accidents according to an embodiment of the present invention may include a road condition information board 100 and a road condition prediction information providing device 200 .

The 1st to 4th CCTVs (11 to 14) are installed for the purpose of crime prevention on roads where traffic lights are not installed or it is difficult to secure the driver's visibility (eg, back roads, alleys, etc.), or to grasp road conditions may be installed for this purpose. In addition, the first to fourth CCTVs 11 to 14 are connected to the intersection as shown in FIG. 2 or 3, for example, and are installed on multiple roads crossing at the intersection, that is, access roads approaching the intersection, There may be. Although three or four CCTVs 11 to 14 are shown in the drawing, this can be added or subtracted regardless of the shape of the intersection as an example.

The first to fourth CCTVs 11 to 14 may photograph surrounding roads (ie, access roads) in real time and transmit the captured video or still images to the road condition prediction information providing device 200 in real time. Hereinafter, the back road will be described as an example as a road on which the first to fourth CCTVs 11 to 14 are installed.

2 is a diagram showing a +-shaped intersection.

Referring to FIG. 2, first to fourth access roads are connected to the +-shaped intersection, and for example, the first vehicle 21 driving in the direction of the intersection on the first access road goes straight to the second access road. You can drive or turn right and drive on the third approach or turn left and drive on the fourth approach. In this way, the second to fourth access roads through which the first vehicle can travel starting from the intersection are referred to as roads where the first vehicle can proceed.

In the case of the second vehicle, the first, third, and fourth access roads that can be driven from the intersection are the first, third, and fourth access roads. In the case of the third vehicle, the first, second, and fourth access roads that can be driven from the intersection are the first, second, and fourth access roads. In the case of the fourth vehicle, the roads that can be driven from the intersection are the first to third access roads.

3 is a diagram illustrating a T-shaped intersection.

Referring to FIG. 3, in a T-shaped intersection, first to third access roads are connected. For example, the first vehicle 21 driving in the direction of the intersection on the first access road turns left and goes to the second access road. You can drive or go straight to the 3rd approach. In addition, the second vehicle 22 driving in the direction of the intersection on the second access road may turn right and drive on the first access road or turn left and drive on the third access road. The third vehicle driving on the third access road may go straight or turn left.

Referring back to FIG. 1 , the road condition information board 100 displays the road condition prediction information generated by the road condition prediction information providing device 200 and is installed at an intersection so that the driver of the vehicle can view the road condition prediction information. It can be. In an embodiment of the present invention, the first and second road condition information boards 110 and 120 installed at +-shaped intersections and T-shaped intersections will be described as examples.

The first and second road condition information plates 110 and 120 shown in FIGS. 4 and 5 individually drive or control each information plate 110 and 120 to display different road condition prediction information for each approach. can do.

4 is an exemplary view of a first road condition information board 110 installed at a +-shaped intersection.

Referring to FIG. 4, the first road condition information board 110 installed at a +-shaped intersection includes a right information board providing road condition prediction information corresponding to a vehicle making a right turn, and a right information board providing road condition prediction information corresponding to a left turn of a target object. The left information board providing road condition prediction information may be included as many as the number of access roads connected to the intersection.

In detail, for the first vehicle traveling in the direction of the intersection on the first access road, road condition prediction information predicted when turning right is displayed on the first right information board 111, and road condition prediction information predicted when turning left is displayed on the first left information plate 112.

In addition, for the second vehicle driving on the second access road, road condition prediction information predicted when making a right turn is displayed on the second right information board 113, and road condition prediction information predicted when making a left turn is displayed as second left information. Mark on board 114.

The third right information board 115 and the third left information board 116 are used to provide road condition prediction information to the third vehicle driving on the third access road, and to the fourth vehicle driving on the fourth access road. Road condition prediction information is provided using the fourth right information board 117 and the fourth left information board 118.

As a result, if the driver (or pedestrian) of the vehicle on each access road proceeds to the right and left access roads based on the current driving direction of the vehicle (ie, in the case of a turning direction in which the visibility is restricted), conflicts may occur for each. It is possible to check road condition prediction information including possibility (hereinafter referred to as 'conflict possibility'). Road condition prediction information in the straight direction is not provided because the driver can check it with the naked eye, but it can be selectively operated.

Here, conflicting means that spheres of influence between objects overlap at the same time. For example, while driving on the first access road while turning right at the intersection, the fourth vehicle makes contact with another object that has turned left on the third access road, or encounters the first vehicle traveling straight on the first access road so as to face each other. In this case, it can be said that there is a conflict.

5 is an exemplary view of a second road condition information board 120 installed at a T-shaped intersection.

Referring to FIG. 5, the second road condition information board 120 installed at the T-shaped intersection provides road condition prediction information corresponding to the case where the first vehicle driving on the first access road makes a left turn from the main road to a sub-road and A double-sided information board 121 providing road condition prediction information corresponding to the case where a third vehicle driving on the 3rd access road makes a right turn from the main road to a sub-road, and a second vehicle traveling on the 2nd access road A right information board 122 providing road condition prediction information corresponding to a right turn from a sub-road to a main road, and a road condition prediction information corresponding to a case where the second vehicle makes a left turn from a sub-road to a main road It includes a left information plate 123.

In addition, the same information may be displayed on the front and back sides of the double-sided information board 121 as the road condition prediction information on the first and third access routes.

In addition, on a T-shaped road, the second road condition information board 120 may be installed in front of the second access road as shown in FIG. It may be determined by the angle between the first to third approach paths. That is, the double-sided information board 121, the right information board 122, or the left information board 123 can be seen from the front at the maximum on each of the first to third access roads, which can be adjusted according to the driver's convenience. can

Referring back to FIG. 1, the road condition prediction information providing device 200 analyzes the images captured by the first to fourth CCTVs 11 to 14, and detects other objects for each possible road that the vehicle can proceed at the intersection. The possibility of conflict can be evaluated, and road condition prediction information including the result of the evaluation of the possibility of conflict can be generated for each degree of possibility of progress. The progressable road has been described with reference to FIGS. 2 and 3 .

FIG. 6 is a block diagram showing an apparatus 200 for providing road condition prediction information shown in FIG. 1 .

Referring to FIG. 6 , the road condition prediction information providing device 200 includes a communication interface unit 210, a user interface unit 220, a storage unit 230, a memory 240 and a processor 250.

The communication interface unit 210 may receive images from the first to fourth CCTVs 11 to 14 and transmit road condition prediction information generated by the processor 250 to the road condition information board 100 .

The user interface unit 220 provides an interfacing role between the device 200 and a user providing a service related to road condition prediction information.

The storage unit 230 may store images received from the first to fourth CCTVs 11 to 14 through the communication interface unit 210 .

Memory 240 may include volatile memory and/or non-volatile memory. In the memory (for example, in order to implement and/or provide operations, functions, etc. provided by the device for providing road condition prediction information 200, commands or data related to the components 210 to 250, one or more programs and /or software, operating system, etc. may be stored.

Programs stored in the memory 240 include an object recognition model and a conflictability evaluation program.

The object recognition model is a model created based on artificial intelligence learning to recognize an object traveling in the direction of an intersection from an approach road by analyzing an image. Object recognition related technology is a computer vision technology that identifies objects on images or videos, and is a technology calculated through deep learning or machine learning algorithms.

In the case of deep learning, object recognition can not only identify objects in an image, but also their location and type. This makes it possible to identify multiple objects in one image and determine the location of each. Object recognition using deep learning uses deep learning models, such as convolutional neural networks (CNNs), which are used to automatically learn features unique to objects to identify them.

The conflict possibility evaluation program is a program for evaluating the possibility of conflict between vehicles and other objects (two-wheeled vehicles, people, etc.) for each road on which vehicles can proceed at an intersection.

In addition, the memory 240 may store conflict direction weight tables of [Table 1] and [Table 2] and conflict area weight tables of [Table 3].

[Table 1] is an example of a conflict direction weight table that stores weights for the possibility of conflict according to the traveling direction of each access road of a +-shaped intersection, that is, conflict direction weights.

[Table 2] is an example of a conflict direction weight table that stores weights for the possibility of conflict according to the traveling direction of each access road of a T-shaped intersection.

Conflict direction weight tables in [Table 1] and [Table 2] can be determined based on the average value of the number of traffic accidents determined through frequency analysis by conflict type for the number of traffic accidents occurring at intersections. In the case of +-shaped intersections, the relative frequency (or relative frequency) of the number of accidents for each type of movement at +-shaped intersections can be calculated and determined, or the arithmetic average of the calculated relative frequencies can be calculated and set as the weight for the possibility of conflict. The same is true for T-shaped intersections.

Alternatively, the conflict direction weight tables in [Table 1] and [Table 2] may be determined by analyzing the frequency of conflict types of traffic accidents in the past (period designated by the manager) for each point where the road condition information board 100 is installed. This is in consideration of a point where the number of conflicts between characteristic movement flows is greater than the average of other regions due to the topographical characteristics of the location where the road condition information board 100 is installed.

The conflicting direction weight tables in [Table 1] and [Table 2] are values calculated by considering the accident frequency for each moving flow within the intersection, and can be updated by reflecting the accident characteristics for each intersection.

On the other hand, in the case of an object rotating at an intersection operated without traffic lights, such as a back road, poor visibility is experienced in the direction of travel. Therefore, it is an important factor in predicting whether or not an accident will occur whether or not an accident will occur. This is because the degree of securing the driver's sight distance is different according to the location of the conflicting area.

In the embodiment of the present invention, in consideration of these characteristics, based on the frequency of past traffic accident occurrence points at the intersection where the road condition information board 100 is installed, each area where a conflict may occur (hereinafter referred to as 'conflict area') You can set weights for potential conflicts. Conflict possibility means the possibility that conflict may occur for each conflict area.

7 is a diagram showing conflict areas of a +-shaped intersection, and FIG. 8 is a diagram showing conflict areas of a T-shaped intersection.

Referring to FIG. 7, there are 5 conflicting areas (0 to 4th conflicting areas) at the +-shaped intersection, and referring to FIG. 8, 4 conflicting areas (0 to 3rd conflicting areas) exist at the T-shaped intersection. areas) exist.

[Table 3] is an example of a conflict area weight table that stores the weights of the conflict potential set in the conflict area at the +-shaped intersection.

area of conflict Area 0 Area 1 Area 2 Area 3 Area 4 number of accidents 4 3 0 2 One area of conflict
weight 2 2 0.8 2 One

Referring to [Table 3], the maximum value of the conflict area weight is, for example, 2, and the minimum value may be 0.8 even in areas where no accident has occurred, which is adjustable. In addition, in the case of a newly constructed intersection, the weight of the conflicting area of area 0 may be set to 1.2, and the weight of the other areas may be set to 0.8. In addition, weights such as [Table 1], [Table 2], and [Table 3] may be periodically updated according to the frequency of traffic accidents or may be changed according to options.

In addition, a blinking color table in which display colors for each blinking frequency of road condition prediction information, which will be described later, is mapped may be stored in the memory 240 . For example, since the higher the blinking frequency, the higher the risk of an accident, red color may be mapped, and as the blinking frequency is 0 or lower, green or blue color may be mapped.

Meanwhile, the processor 250 executes one or more programs stored in the road condition prediction information providing device 200 to control the overall operation of the device 200. For example, the processor 250 may analyze an image using an object recognition model stored in the memory 240 and determine a moving object from the analysis result. In addition, the processor 250 executes a conflict possibility evaluation program to predict the movement trajectory of a moving object, evaluates the possibility of collision with other objects according to the movement trajectory, and obtains road condition prediction information from the evaluation result of the possibility of collision It can be created and provided for each object's progressability.

To this end, the processor 250 may include an object recognition unit 251, a movement trajectory prediction unit 252, a conflict possibility evaluation unit 253, a road condition prediction information providing unit 254, and a weight calculation unit 255. can

The object recognizing unit 251 may recognize objects approaching an intersection where traffic lights are not installed, that is, moving objects. The object recognition unit 251 may input images provided from the first to fourth CCTVs 11 to 14 into the object recognition model to recognize an object driving at an intersection and determine the type of the recognized object. Types of objects include moving objects capable of recognizing information, such as vehicles, two-wheeled vehicles, and people.

The movement trajectory predictor 252 may predict the movement trajectory of each of the objects recognized by the object recognizing unit 251 . That is, the movement trajectory predictor 252 predicts the movement trajectory for each approach where the moving object goes straight, right turn, and left turn, and generates road condition prediction information based on this.

For example, in FIG. 2 , the movement trajectory predictor 252 is a movement trajectory when the fourth vehicle makes a right turn to the first access road, a movement trajectory when the fourth vehicle makes a left turn to the second access road, and a movement trajectory when the fourth vehicle goes straight to the third access road. The movement trajectories of all cases can be predicted. This is because objects moving on back roads move at a relatively low speed compared to arterial roads or sub-arterial roads, so most of them must be close to the intersection to express their intention on whether to turn or go straight using turn signals, or to control turn indicators. This is because it often turns or goes straight without knowing or not knowing the information about the direction change.

To this end, the movement trajectory predictor 252 predicts the movement trajectory of the object by knowing the movement distance and movement time of the object from the image and calculating the movement speed and acceleration therefrom. [Equation 1] is an equation for obtaining the movement trajectory of the object.

In [Equation 1], S is the movement trajectory, V is the movement speed of the object, t is time, and a is acceleration.

Also, the movement trajectory predictor 252 sets the sphere of influence of the object in consideration of the size of the recognized object. The movement trajectory prediction unit 252 is expressed as a rectangle defined by the length (L) and width (w) for vehicles, two-wheeled vehicles, etc. In the case of humans, adults are defined as 1.0m in width and length, respectively, and children are defined in width and length respectively. The sphere of influence can be set as a rectangle defined by 0.75 m, which can be changed as an example. Therefore, conflicting in an embodiment of the present invention may mean overlapping spheres of influence between objects at the same time.

9 is an exemplary view showing the movement trajectories of a vehicle and a two-wheeled vehicle running at a T-shaped intersection.

Referring to FIG. 9 , as a result of predicting the movement trajectory of the two-wheeled vehicle 22 running on the second access road and the moving trajectory of the vehicle 23 running on the third access road, the sphere of influence of the two-wheeled vehicle 22 and the vehicle 23 It can be determined that there is a possibility of overlapping and conflicting in the 0th area.

On the other hand, the conflict possibility evaluation unit 253 moves each of the objects predicted by the movement trajectory prediction unit 252 with respect to an object to provide road condition information (hereinafter, referred to as a 'target object') among recognized objects. Using the trajectory, it is possible to evaluate the possibility of conflict with the remaining objects according to the possible progress of the target object at the intersection. The remaining objects refer to objects other than the target object among objects recognized by the object recognizing unit 251 .

That is, the possibility of collision evaluation unit 253 may predict all the possibilities of colliding with the other objects on the possibility of traveling by turning right or by the possibility of traveling by turning left.

FIG. 10 is a block diagram showing the conflict possibility evaluation unit 253 shown in FIG. 6 in detail.

Referring to FIG. 10 , the conflict possibility evaluation unit 253 may include an object prediction unit 253a, a conflict possibility determination unit 253b, a conflict direction weight verification unit 253c, and a conflict area weight verification unit 253d. have.

The object prediction unit 253a analyzes the movement trajectories of the remaining objects and predicts an object to be approached first (hereinafter, referred to as 'approaching object') for each degree of progression of the target object. The object prediction unit 253a analyzes all the movement trajectories of the remaining objects when they go straight, turn left, or turn right.

For example, when the fourth vehicle 24 is driving in the direction of area 0 at a +-shaped intersection and the first to third objects 21 to 23 are recognized on the first to third approach roads, the fourth vehicle 24 Vehicle 24 can turn right, go straight or turn left. The object prediction unit 253a selects an object to be approached first on the first access road, which is a possible road when the fourth vehicle makes a right turn, and on the second access road, which is a possible road when the fourth vehicle makes a left turn, among the first to third objects 21 ~23) can be predicted by analyzing their movement trajectories. Approach objects predicted for each of the first and second access routes may be the same or different.

In addition, at the same time, when the second object 22 is a vehicle, the object prediction unit 253a sets the second vehicle 22 as a target object, predicts an approaching object by the degree of progress of the second vehicle 22, and , this is the same for the third and fourth objects.

The conflict possibility determination unit 253b determines the possibility of conflict between the approaching object predicted by the object prediction unit 253a and the target object according to the progressability of the target object, and generates 1 if there is a possibility of conflict and 0 if there is no conflict direction weight. It can be transmitted to the verification unit 253c and the conflict area weight verification unit 253d. In addition, the conflict possibility determination unit 253b may further transfer identification information on areas in which there is a possibility of conflict to the conflict area weight verification unit 253d.

For example, in FIG. 2 , when the target object is the fourth vehicle 24 and the fourth vehicle 24 turns right and drives on the first access road, the approaching object is the first vehicle 21 and the fourth vehicle ( When 24) turns left and drives on the second access road and the approaching object is predicted to be the third vehicle 23, the collision possibility determination unit 253b determines the movement of the fourth vehicle 24 and the first vehicle 21 In the trajectory, if the spheres of influence of the fourth vehicle 24 and the first vehicle 21 overlap in the first area, it is determined that there is a possibility of conflict and 1 is output. If the spheres of influence do not overlap, it is determined that there is no possibility of conflict and 0 is output . In addition, if the spheres of influence of the fourth vehicle 24 and the third vehicle 23 overlap in the second area, the possibility of conflict determination unit 253b determines that there is a possibility of conflict and outputs 1, and if the spheres of influence do not overlap, the possibility of conflict It judges that there is none and outputs 0.

If the conflict direction weight check unit 253c determines that there is a possibility of conflict between the target object and the approaching object in the conflict possibility determination unit 253b, the conflict possibility weight according to the direction of movement of the target object and the approaching object (hereinafter referred to as 'conflict direction') weight) can be checked in the trade-off weight table.

For example, when the target object is the 4th vehicle, the approaching object is predicted as the 2nd object when the 4th vehicle turns right in the object prediction unit 253a, and the collision possibility determination unit 253b determines the 4th vehicle and the 4th vehicle. If there is a possibility of collision between the two objects and it is determined that they will collide in the first area, the collision direction weight check unit 253c determines the right turn item of the fourth approach and the second approach in the conflict direction weight table of [Table 1]. Check 1.4 as the weight of the point where the straight items meet.

In addition, when the fourth vehicle turns left in the object prediction unit 253a, the approaching object is predicted as a third object turning right, and the possibility of collision between the fourth vehicle and the third object is predicted in the collision possibility determination unit 253b in the second area. If it is determined that the conflicting direction weight check unit 253c checks 1.1 as the weight of the point where the left turn item of the 4th approach road and the right turn item of the 3rd access road meet in the conflict direction weight table of [Table 1]. do.

In addition, when the fourth vehicle makes a left turn in the object prediction unit 253a, the approaching object is predicted as a third object going straight, and the possibility of collision between the fourth vehicle and the third object in the collision possibility determination unit 253b is in the 0 area. If it is determined that the conflicting direction weight check unit 253c checks 1.4 as the weight of the point where the item turning left of the fourth approach road and the straight item of the third approach meet in the conflicting direction weight table of [Table 1]. do.

If the conflict probability determination unit 253b determines that there is a possibility of conflict, the conflict area weight check unit 253d sets the conflict probability weight (ie, conflict area weight) set in the area where conflict is predicted to occur in the conflict area weight table. You can check.

For example, if it is determined by the conflict possibility determination unit 253b that there will be a collision in the first area when the fourth vehicle makes a right turn, the conflict area weight check unit 253d determines the conflict area weight table in [Table 3]. Check 2 as the conflict area weight set in area 1. In addition, when it is determined that there will be a conflict in the second area when the fourth vehicle makes a left turn, the conflict area weight check unit 253d sets 0.8 as the conflict area weight set in the second area in the conflict area weight table of Table 3. Check.

Referring back to FIG. 6, the road condition prediction information providing unit 254 generates road condition prediction information including the conflict possibility evaluation result evaluated for each degree of progress of the target object in the conflict possibility evaluation unit 253, and then proceeds to the intersection. It is provided to the installed road condition information board 100.

FIG. 11 is a block diagram showing the road condition prediction information provider 254 shown in FIG. 6 in detail.

Referring to FIG. 11 , the road condition prediction information provider 254 may include a blinking frequency calculator 254a, an information generator 254b, and an information provider 254c.

The blink frequency calculation unit 254a determines the possibility of conflict determined by the conflict possibility determination unit 253b, the conflict direction weight determined by the conflict direction weight confirmation unit 253c, and the conflict area weight determined by the conflict area weight confirmation unit 253d. Based on this, the blinking frequency of the road condition prediction information can be calculated for each degree of progress of the target object. The flashing frequency of the road condition prediction information means the number of blinks per unit time when the road condition prediction information is displayed in a blinking form on the road condition information board 100 . The blinking frequency calculation unit 254a may calculate the blinking frequency using [Equation 2].

In [Equation 2], the possibility of conflict has a value of 0 or 1 determined by the possibility of conflict determination unit 253b. In addition, since the probability of occurrence of a dangerous situation increases as the conflicting direction weight or the conflicting area weight increases, the blinking frequency increases and the number of blinking per unit time increases. In [Equation 2], if the possibility of conflict is 0, it means that it is safe because there is no possibility of conflict. Accordingly, road condition prediction information may be continuously displayed in a color indicating safety (for example, green).

The information generating unit 254b includes the type of target object and approaching object recognized by the object recognition unit 251, the current driving speed and moving trajectory of the approaching object, the possibility of collision between the target object and the approaching object, and the predicted area where the collision will occur. Road condition prediction information including at least one may be generated for each degree of progress of the target object.

The information providing unit 254c may provide the road condition information board 100 with the road condition prediction information generated for each degree of progress of the target object and the blinking frequency calculated by the blinking frequency calculating unit 254a.

Therefore, for example, when the first to fourth vehicles are simultaneously driving at a +-shaped intersection, each of the first to fourth vehicles is a target object and an approaching object at the same time, and can be provided as road condition prediction information. Up to 8 road condition prediction information and flashing frequency can be generated.

The road condition information board 100 displays road condition prediction information for each progressable degree of the target object based on the calculated flashing frequency. As a result, it is possible to provide pedestrians, vehicle drivers, and motorcycle drivers with road condition prediction information in different colors and blinking cycles according to the degree of risk to prevent traffic accidents at intersections on back roads operated without traffic lights.

Meanwhile, the weight calculation unit 255 shown in FIG. 6 may calculate the weight of the conflicting area using [Equation 3] when the number of traffic accidents for each area of the intersection is updated.

In [Equation 3], the number of traffic accidents in the corresponding area in the past is the number of traffic accidents that occurred during a predetermined period in the past for each area existing at the intersection, as shown in [Table 3], and the minimum number of traffic accidents that is not zero is It is the minimum number excluding 0 among the number of traffic accidents that occurred during a fixed period in the past for each area.

Then, the weight calculation unit 255 sets the conflict area weight to the maximum value M if the calculated conflict area weight is M or more, and sets the conflict area weight to the minimum value N if the calculated conflict area weight is 0. If the area weight is between M and N, the calculated value is set as it is. M may be 2 as an example, and N may be 0.8, which is variable. In this way, if the maximum and minimum values of the conflicting area weight are not set, the blinking frequency calculated in [Equation 2] may be unrealistically fast or may be calculated as 0, so the present invention determines the maximum and minimum values of the conflicting area weight . Taking [Table 3] as an example, the weight calculation unit 255 calculates the weight of the conflicting area of the 0th area as follows.

, 이 때 산출된 4는 2보다 크므로 가중치 산출부(255)는 제0영역의 가중치를 2로 정한다. 또한, 가중치 산출부(255)는 교통사고가 발생수가 0인 제2영역의 경우 최소값인 0.8을 상충영역 가중치로 정한다.

, Since the calculated 4 at this time is greater than 2, the weight calculation unit 255 sets the weight of the 0th area to 2. In addition, the weight calculation unit 255 sets the minimum value of 0.8 as the conflict area weight in the case of the second area where the number of traffic accidents is zero.

12 is a flowchart illustrating a method for providing road condition prediction information according to an embodiment of the present invention.

Since the method for providing road condition prediction information shown in FIG. 12 is operated by the road condition prediction information providing apparatus 200 described with reference to FIGS. 1 to 11 , detailed description thereof will be omitted.

Referring to FIG. 12, the road condition prediction information providing device 200 captures images of roads (eg, back roads) connected to intersections where traffic lights are not installed from the first to fourth CCTVs 11 to 14. Receive (S1210).

The road condition prediction information providing device 200 analyzes the received image and recognizes an object approaching the intersection as each approach road (ie, each road connected to the intersection) (S1220).

The road condition prediction information providing device 200 may recognize the movement trajectory of each moving object among the recognized objects by analyzing the image (S1230).

The road condition prediction information providing device 200 uses the movement trajectory of each of the objects predicted in step S1230 for any one of the objects recognized in step S1220 (hereinafter, referred to as 'target object') to target the target object. It is possible to evaluate whether there is a possibility of conflict with the rest of the objects for each possible progress at the intersection (S1240). For example, when the target object is the first object, the possible paths may be second to fourth approaches, and at least one of the second to third objects may be the remaining objects.

The road condition prediction information providing device 200 generates road condition prediction information including the conflict potential evaluation result evaluated in step S1240 and provides it to the road condition information board 100 installed at the intersection (S1250).

FIG. 13 is a flowchart illustrating step S1240 of FIG. 12 .

Referring to FIG. 13 , the road condition prediction information providing apparatus 200 analyzes the movement trajectories of the remaining objects and predicts an object to be approached first (hereinafter, referred to as 'approaching object') for each degree of progress of the target object. (S1241).

The road condition prediction information providing apparatus 200 may determine the possibility of conflict between the approaching object predicted in step S1241 and the target object according to the progressability of the target object (S1242). In step S1242, it is possible to determine the possibility of conflict by predicting whether the spheres of influence overlap when the approaching object and the target object move along the moving trajectories using the moving trajectories of the approaching object and the target object.

If it is determined that there is a possibility of conflict (S1243-Yes), the road condition prediction information providing device 200 generates 1 (S1244).

The road condition prediction information providing device 200 checks the weight of the possibility of collision according to the direction of collision between the target object and the approaching object, that is, the weight of the collision direction in the collision direction weight table of [Table 1] (S1245).

The apparatus 200 for providing road condition prediction information checks the weight set in the area in which conflict is predicted to occur, that is, the weight of the conflict area in the conflict area weight table of [Table 2] (S1246).

In addition, if it is determined that there is no possibility of conflict in step S1242 (S1243-No), the road condition prediction information providing device 200 generates 0 (S1247).

FIG. 14 is a flowchart illustrating step S1250 of FIG. 12 .

Referring to FIG. 14, the road condition prediction information providing device 200 substitutes the conflict possibility 1 generated in step S1244, the conflict direction weight determined in step S1245, and the conflict area weight determined in step S1246 into [Equation 2]. Thus, it is possible to calculate the blinking frequency for displaying road condition prediction information for each degree of progress of the target object (S1251).

The road condition prediction information providing device 200 includes the type of approaching object predicted in step S1241, the moving speed of the approaching object and the target object, the moving trajectory of the approaching object and the target object, the possibility of conflict between the approaching object and the target object, and the approaching object and Road condition prediction information including at least one of regions in which a conflict between the target object is predicted to occur may be generated for each degree of progress of the target object (S1252).

The road condition prediction information providing device 200 provides the road condition information board 100 with the road condition prediction information generated for each progress possibility of the target object in step S1252 and the flashing frequency calculated for each progress possibility of the target object in step S1251. Transmit (S1253).

The road condition information board 100 displays the road condition prediction information received for each degree of progress of the target object at a set flickering cycle in a corresponding area. As a result, a vehicle or other moving object driving at an intersection can check the road condition forecast information displayed on the front to recognize in advance that there is a possibility of a collision (or collision) with an object approaching from another approach and prepare in advance, such as driving at a reduced speed. can

For example, when the objects recognized in step S1220 are the first to third objects, the road condition prediction information providing apparatus 200 sets each of the first to third objects as a target object and performs the above-described operation, Road condition prediction information is generated for each drivable road of the first to third objects, and thus, the first or second road condition information plates 110 and 120 may display the road condition prediction information in the corresponding area. .

In the above, even though all the components constituting the embodiment of the present invention have been described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent hardware, some or all of the components are selectively combined to perform some or all of the combined functions in one or a plurality of hardware. It may be implemented as a computer program having. Codes and code segments constituting the computer program can be easily inferred by a person skilled in the art. Such a computer program may implement an embodiment of the present invention by being stored in a computer readable storage medium, read and executed by a computer.

On the other hand, although the above has been described and illustrated in relation to preferred embodiments for illustrating the technical idea of the present invention, the present invention is not limited to the configuration and operation as shown and described in this way, and departs from the scope of the technical idea. It will be apparent to those skilled in the art that many changes and modifications can be made to the present invention without modification. Accordingly, all such appropriate alterations and modifications and equivalents are to be regarded as falling within the scope of the present invention. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached claims.

11 to 14: 1st to 4th CCTVs 21 to 24: 1st to 4th objects
100: road condition information board 110, 120: first and second road condition information boards
200: road condition prediction information providing device
210: communication interface unit 220: user interface unit
130: storage unit 140: memory
150: processor 251: object recognition unit
252: movement trajectory prediction unit 253: conflict possibility evaluation unit
254: road condition prediction information providing unit 255: weight calculation unit
253a: object prediction unit 253b: conflict possibility determination unit
253c: Conflict direction weight confirmation unit 253d: Conflict area weight confirmation unit
254a: blinking frequency calculation unit 254b: information generation unit
254c: information provision unit

Claims (13)

an object recognition unit recognizing objects approaching an intersection where traffic lights are not installed;
a movement trajectory predictor for predicting a movement trajectory of each of the recognized objects;
With respect to any one of the recognized objects (hereinafter referred to as 'target object'), a road (hereinafter referred to as 'target object') on which the target object can proceed at an intersection using the movement trajectory of each of the objects predicted by the movement trajectory prediction unit a conflict possibility evaluation unit that evaluates a possibility of conflict with the remaining objects among the recognized objects for each object; and
A road situation prediction information providing unit that generates road situation prediction information including the conflictability evaluation result evaluated for each progressability of the target object in the conflictability evaluation unit and provides it to a road situation information board installed at an intersection; Road condition prediction information providing device.
According to claim 1,
The conflict potential evaluation unit,
an object prediction unit that analyzes the movement trajectories of the remaining objects and predicts an object to be approached first (hereinafter, referred to as 'approaching object') for each degree of progress of the target object;
a conflict possibility determination unit for determining the possibility of collision between the approaching object predicted by the object prediction unit and the target object according to the progressability of the target object; and
If it is determined that there is a possibility of conflict in the conflict possibility determination unit, a conflict direction weight check unit for checking a conflict possibility weight (hereinafter referred to as 'conflict direction weight) according to the driving direction of the target object and the approaching object; Road condition prediction information providing device characterized by.
According to claim 2,
The conflict potential evaluation unit,
If it is determined that there is a possibility of conflict in the conflict possibility determination unit, a conflict area weight check unit that checks a conflict possibility weight (hereinafter referred to as 'conflict area weight') set in an area in which a conflict is predicted to occur; further comprising Road condition prediction information providing device characterized in that.
According to claim 3,
The road condition prediction information providing unit,
The road condition prediction information is displayed based on the possibility of conflict determined by the conflict possibility determination unit, the conflict possibility weight according to the driving direction determined by the conflict direction weight verification unit, and the conflict area weight determined by the conflict area weight verification unit. a blinking frequency calculation unit for calculating a blinking frequency for each degree of progress of the target object;
Information generation unit for generating road condition prediction information including at least one of the type of the approaching object, movement speed, movement trajectory, possibility of collision, and an area in which a collision with the approaching object is predicted for each degree of progress of the target object. ; and
An information providing unit providing road condition prediction information generated for each possible progress of the target object to the road condition information board;
The road condition prediction information providing device, characterized in that the road condition information board displays road condition prediction information for each possible progress of the target object based on the calculated flashing frequency.
According to claim 1,
The direction of the progressable road for providing the road condition prediction information is
the intersection In the case of a +-shaped intersection, a right turn direction and a left turn direction are included based on the current driving direction of the target object,
When the intersection is a T-shaped intersection, the apparatus for providing road condition prediction information, characterized in that it includes at least one of a right turn direction and a left turn direction based on the current driving direction of the target object.
(A) recognizing, by an electronic device, objects approaching an intersection where traffic lights are not installed;
(B) predicting, by the electronic device, a movement trajectory of each of the approaching objects;
(C) With respect to any one of the objects recognized in step (A) (hereinafter referred to as 'target object'), using the movement trajectory of each of the objects predicted in step (B), the target object Evaluating a possibility of conflict with the remaining objects among the recognized objects for each road (hereinafter referred to as a 'progressability road') at the intersection; and
(D) generating, by the electronic device, road condition prediction information including the conflictability evaluation result evaluated in step (C) and providing it to a road condition information board installed at an intersection; providing road condition prediction information including; Way.
According to claim 6,
In step (C),
(C1) predicting, by the electronic device, an object to be approached first (hereinafter, referred to as an 'approaching object') for each progressability of the target object by analyzing the movement trajectories of the remaining objects;
(C2) determining the probability of conflict between the approach object predicted in step (C1) and the target object according to the progressability of the target object; and
(C3) if it is determined that there is a possibility of conflict in the step (C2), checking a conflict possibility weight (hereinafter referred to as 'conflict direction weight) according to the driving direction of the target object and the approaching object; A method for providing road condition prediction information.
According to claim 7,
The step (C), after the step (C3),
(C4) if it is determined that there is a possibility of conflict in step (C2), checking the weight set in the area in which conflict is predicted (hereinafter referred to as 'conflict area weight'); How to provide road condition forecast information.
According to claim 8,
In the step (D),
(D1) Based on the possibility of conflict determined in the step (C2), the weight of the conflict direction determined in the step (C3), and the weight of the conflict area determined in the step (C4), the road situation for each possible progress of the target object. Calculating a blinking frequency for displaying prediction information;
(D2) generating road situation prediction information including at least one of the type of the approaching object, movement speed, movement trajectory, possibility of collision, and an area in which a collision with the approaching object is predicted for each degree of progress of the target object step; and
(D3) transmitting the road condition prediction information generated for each degree of progress of the target object in the step (D2) and the blinking frequency calculated in the step (D1) to the road condition information board. How to provide road condition forecast information.
According to claim 6,
The direction of the progressable road for providing the road condition prediction information is
the intersection In the case of a +-shaped intersection, a right turn direction and a left turn direction are included based on the current driving direction of the target object,
and when the intersection is a T-shaped intersection, at least one of a right turn direction and a left turn direction based on the current driving direction of the target object.
By analyzing images captured by closed-circuit televisions (CCTVs) installed on roads connected to intersections, the possibility of conflict with other objects for each road (hereinafter referred to as 'progressable road') on which vehicles can proceed at the intersection is analyzed. a road condition prediction information providing device that evaluates and generates road condition prediction information including a result of conflict possibility evaluation for each of the possible roads; and
A road condition information plate installed at an intersection to display road condition prediction information generated for each possible road of the vehicle in the road condition prediction information providing device, and to allow the driver of the vehicle to view the road condition prediction information; Road condition prediction information provision system.
According to claim 11,
If the intersection is a +-shaped intersection,
The road condition information board includes a right information board providing road condition prediction information corresponding to a right turn of the target object and a left information plate providing road condition prediction information corresponding to a left turn of the target object. A road condition prediction information providing system characterized in that it includes as many as the number of roads connected to the intersection.
According to claim 11,
If the intersection is a T-shaped intersection,
The road condition information board includes road condition prediction information corresponding to when the target object makes a right turn from a main road to a sub-road and road condition prediction information corresponding to a case where the target object makes a left turn from a main road to a sub-road on both sides. A double-sided information board provided as a right side information board, a right side information board providing road condition prediction information corresponding to when the target object makes a right turn from a sub-road to a main road, and a corresponding case when the target object makes a left turn from a sub-road to a main road. A system for providing road condition prediction information, characterized in that it includes a left information board for providing road condition prediction information.

Images