KR102585032B1 - Object recognition safety system on the road - Google Patents

Abstract

The road traffic object recognition safety system includes a plurality of cameras installed toward each road at a residential intersection on the back side of a street, an object recognition unit that recognizes moving objects on the road, and a plurality of cameras installed toward each road at a residential intersection. A display unit including a plurality of LED electric signs adjacent to the cameras, and a control unit that displays risk information related to the moving object on each of the plurality of LED electric signs when a moving object is recognized on a specific road among the plurality of roads at an intersection. Includes.
The control unit differently displays risk information related to the moving object based on at least one of the speed, type, and approach distance of the recognized moving object.

Description

{Object recognition safety system on the road}

The embodiment relates to a road traffic object recognition safety system that can quickly and efficiently notify blind spot risks.

Typically, at intersections on or adjacent to major streets, the sidewalks adjacent to the major streets are wide, so there are no blind spots. Even if there are blind spots around intersections, crosswalks are installed or warning devices that issue voice or text warnings of danger are installed to inform pedestrians or moving objects of danger and prevent accidents from occurring.

On the other hand, there are no crosswalks or warning devices installed on back roads off the main street, that is, roads or intersections in residential areas, so there is always a very high possibility of pedestrian accidents occurring. Each road adjacent to a residential intersection on this back road has one lane in width, and even if it has two lanes in one way, cars are often parked on both sides of the road, so the road width on which moving objects can actually move is very narrow. do.

In particular, in countries where the population density is concentrated in cities, such as Korea, not only are there no sidewalks adjacent to back roads, but high-rise apartments, multi-family houses, and villas are densely packed around the back roads. Additionally, in Korea, the majority of cars that cannot be parked in houses are frequently parked on one or both sides of side roads or at the corners of intersections.

In this way, because the actual or realistic width of the road adjacent to the residential intersection on the back side of the street is narrow, only when a pedestrian enters the intersection can the driving status of a moving object on a road other than the pedestrian's entry road be ascertained. If a pedestrian enters an intersection in a hurry, the driving state of a moving object on a road other than the pedestrian's entry road cannot be recognized, resulting in serious injury or death due to a collision with a moving object on another road.

In particular, in the case of children, when they focus on play due to poor judgment, they often run into intersections without checking the situation around the intersection, which is a problem that causes more and more accidents among children.

The embodiments aim to solve the above-described problems and other problems.

Another object of the embodiment is to provide a road traffic object recognition safety system capable of displaying optimal risk information related to an object to suit each of the various situations of objects located on the road corresponding to a blind spot at a residential intersection on the back side of a boulevard. It is done.

Another purpose of the embodiment is to track the movement of an object located on the road corresponding to a blind spot at a residential intersection on the back side of the street and determine whether to display risk information related to the object, thereby preventing information display errors. The goal is to provide a road traffic object recognition safety system that can

Another purpose of the embodiment is that when it is raining and dark or there are no street lights in the evening, the camera operates in infrared mode or color mode to recognize objects, thereby delivering accurate dangerous situations not only during the day but also in situations with poor visibility, thereby reducing the accident rate. The goal is to provide a road traffic object recognition safety system that can minimize

The technical problems of the embodiments are not limited to those described in this item and include those that can be understood through the description of the invention.

According to an aspect of the embodiment to achieve the above or other purposes, the road traffic object recognition safety system includes a plurality of cameras installed toward each road of a residential intersection on the back of a main street, and recognizes moving objects on the road. an object recognition unit; A display unit installed toward each road of the residential intersection and including a plurality of LED electronic signs adjacent to the plurality of cameras, respectively; and a control unit that, when a moving object is recognized on a specific road among the plurality of roads at the intersection, displays risk information related to the moving object on each of the plurality of LED signboards, wherein the control unit displays the recognized moving object. Risk information related to the moving object is displayed differently based on at least one of the moving object's speed, type, and entry distance.

Risk information related to the moving object may include icons, indicators, and speed information corresponding to the moving object.

The speed of the moving object may be determined based on the speed limit set at a residential intersection. In this case, the control unit may display the icon or the indicator differently depending on the degree to which the speed of the moving object is greater than the specified speed.

Types of the moving objects may include pedestrians, bicycles, motorcycles, electric kickboards, wheel chairs, and cars. In this case, the control unit may display the icon or the indicator differently depending on the type of the moving object.

The control unit 130 for recognizing moving objects in an embodiment may include an AI Box loaded with image analysis firmware, and may have an object recognition algorithm and a circuit for driving the algorithm built-in. In addition, the control unit 130 for recognition of moving objects in the embodiment may include an additional object recognition device to increase the recognition rate of moving objects in addition to the AI Box in the control unit 130 in order to increase the recognition rate of electric kickboards among moving objects.

Additionally, the embodiment may recognize a wheelchair among moving objects as a person (pedestrian).

The entry distance of the moving object may be the separation distance between the center of the residential intersection and the moving object. In this case, the control unit may display the icon or the indicator differently as the entry distance of the moving object decreases.

The control unit may display the icon to blink or the dots within the arrow to blink along the direction indicated by the arrow, depending on the speed, type, or entry distance of the recognized moving object.

The control unit may display the expanded size or brightness of the icon differently depending on the speed, type, or entry distance of the recognized moving object.

The control unit may display the expansion size of the arrow or the sequential blinking speed of the dots within the arrow differently depending on the speed, type, or entry distance of the recognized moving object.

When a plurality of moving objects are recognized on the specific road, the control unit sends information related to the moving object with the highest risk in case of an accident based on a preset risk order for the plurality of moving objects to each of the plurality of LED electronic signs. can be displayed.

The control unit maps a plurality of virtual areas set for each road in the residential area to images obtained from each of the plurality of cameras, and tracks the movement of a moving object on the mapped virtual areas to determine the risk associated with the moving object. It is determined whether to display information on each of the plurality of LED electronic signs, and when it is determined that the moving object is entering the intersection, risk information related to the moving object may be displayed on each of the plurality of LED electric signs. there is.

When it is raining and dark or there are no street lights in the evening, the camera can operate in infrared or color mode. In this case, the object recognition unit may recognize a moving object on the road based on the infrared or color image acquired by the camera.

The effects of the road traffic object recognition safety system according to the embodiment will be described as follows.

According to at least one of the embodiments, a moving object on each road at the intersection is recognized through a plurality of cameras installed at a residential intersection on the back side of the street, and information on at least one of the speed, type, and approach distance of the moving object is provided. Based on this, risk information related to the moving object may be displayed differently. Accordingly, the optimal risk information suitable for each of the various situations of the moving object is displayed, so that pedestrians and drivers can move according to the optimal risk information displayed. By making it possible to more clearly determine whether an object is dangerous to oneself, it is possible to solve the problem that the risk of an accident is still not reduced by displaying only simple risk information that does not help pedestrians or drivers determine the risk information of the moving object.

According to at least one of the embodiments, the size of the icon included in the risk information related to the moving object or the size of the brightness indicator is changed, or the number of dots in the indicator is changed, so that the pedestrian can determine the type of moving object or the degree of risk of the moving object. Depending on this, you can easily deal with it in advance.

According to at least one of the embodiments, a plurality of virtual areas are set, and the movement of a moving object recognized on the plurality of virtual areas is tracked through images acquired in frame units, thereby generating risk information related to the moving object. Product reliability can be improved by preventing incorrect display.

According to at least one of the embodiments, when the size of the LED signboard of the display unit is small or the area for displaying the corresponding risk information is small, the highest risk is detected based on a preset risk order among a plurality of moving objects recognized on a specific road. By displaying risk information related to moving objects, the display area of the LED signboard can be used efficiently.

According to at least one of the embodiments, in an environment where visibility is dark, each of the plurality of cameras of the object recognition unit is changed to an infrared or color mode to obtain an infrared or color image, so that object recognition is possible safely and accurately even in an environment with poor visibility, Accidents at intersections, where accidents are more common at night, can be prevented.

Additional scope of applicability of the embodiments will become apparent from the detailed description that follows. However, since various changes and modifications within the spirit and scope of the embodiments may be clearly understood by those skilled in the art, the detailed description and specific embodiments, such as preferred embodiments, should be understood as being given by way of example only.

Figure 1 shows a road traffic object recognition safety system according to an embodiment installed at a residential intersection on the back side of a boulevard.
FIG. 2 is a plan view showing an object recognition unit and a display unit mounted on the pole of FIG. 1.
Figure 3 is a diagram illustrating a road traffic object recognition safety system according to an embodiment.
Figure 4 shows that when a moving object enters each road of a residential intersection on the back side of a main street, risk information related to the moving object is displayed on each of a plurality of LED electronic signs.
Figure 5 is a flowchart explaining the operation method of a road traffic object recognition safety system according to an embodiment.
Figure 6 shows recognition of a moving object using a preset virtual area.
Figure 7 is a diagram explaining the entry distance of a moving object.
Figure 8 shows how risk information related to a moving object is displayed.
Figure 9 is a flowchart explaining a method of differently displaying risk information related to a moving object depending on the speed of the moving object.
Figure 10 shows a moving object exceeding the speed limit set at an intersection.
Figure 11 shows how risk information related to an object is displayed differently depending on the speed of the moving object.
Figure 12 is a flowchart explaining a method of differently displaying risk information related to a moving object depending on the type of the moving object.
Figure 13 shows how risk information related to an object is displayed differently depending on the type of moving object.
Figure 14 shows a plurality of moving objects entering a specific road.
Figure 15 shows information related to the highest risk moving object among a plurality of moving objects entering a specific road being displayed.
Figure 16 is a flowchart explaining a method of determining whether to display risk information related to a moving object through tracking the movement of the moving object.
Figure 17 shows a moving object detouring before entering an intersection or entering an intersection.
18 to 23 show risk information related to various moving objects displayed on the display unit.
Figure 24 is an example diagram of a display unit that displays risk information.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of reference numerals, and duplicate descriptions thereof will be omitted. The suffixes 'module' and 'part' for components used in the following description are given or used interchangeably in consideration of ease of specification preparation, and do not have distinct meanings or roles in themselves. Additionally, the attached drawings are intended to facilitate easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed in this specification is not limited by the attached drawings. Additionally, when an element such as a layer, region or substrate is referred to as being 'on' another component, this includes either directly on the other element or there may be other intermediate elements in between. do.

In the following embodiments, a moving object may include a person riding, pushing, or pulling the moving object. For example, if the moving object is a motorcycle, it may include a person riding the motorcycle. For example, when the moving object is a wheelchair, it may include a person pulling the wheelchair.

Figure 1 shows a road traffic object recognition safety system according to an embodiment installed at a residential intersection on the back side of a boulevard. FIG. 2 is a plan view showing an object recognition unit and a display unit mounted on the pole of FIG. 1. Figure 3 is a diagram illustrating a road traffic object recognition safety system according to an embodiment.

Referring to FIGS. 1 to 3 , the road traffic object recognition safety system 100 according to an embodiment may include an object recognition unit 110, a display unit 120, and a control unit 130. The control unit 130 in the embodiment may be an artificial intelligence box (AI box) as a device loaded with image analysis firmware.

Hereinafter, for convenience of explanation, the road traffic object recognition safety system 100 according to the embodiment is limited to being installed on the pole 10, but instead of the pole 10, another member is used to detect the road traffic object according to the embodiment. A cognitive safety system 100 may be installed. That is, the pole 10 may or may not be included in the road traffic object recognition safety system 100 according to the embodiment.

The object recognition unit 110 and the display unit 120 may be installed together on the support 101, but this is not limited. For example, the display unit 120 may be installed on the support 101, and the object recognition unit 110 may be installed on another member. The support 101 may be called a frame or bracket.

The pole 10 may include a pillar 10a and an extended support 10b. The pillar 10a is buried in the ground on one side of the intersection 20 and may serve to support the road traffic object recognition safety system 100 according to the embodiment, that is, the object recognition unit 110 and the display unit 120. . The extension support 10b may extend toward the center CP of the intersection 20 from the upper side of the maneuver.

In the embodiment, the intersection 20 may not be located on the main street, but may refer to a residential intersection 20 on the back side of the main street. That is, the intersection 20 of the embodiment may be a place where there is no crosswalk and no blind spot warning device. As previously described,

Although the drawing shows an intersection 20 where four roads intersect, this embodiment can be equally applied to an intersection 20 where three or five or more roads intersect.

The support 101 on which the object recognition unit 110 and the display unit 120 are mounted may be fastened to the extension support 10b. At this time, the support body 101 may be located at the center (CP) of the intersection 20, and the object recognition unit 110 and the display unit 120 may also be located at the center (CP) of the intersection 20. It is not limited. The center (CP) may be located on an imaginary straight line perpendicular to the traveling direction of the moving object or may be a center point where two diagonals of an intersection intersect, but is not limited thereto.

The object recognition unit 110 may include a plurality of cameras 110-1 to 110-4. Each of these cameras 110-1 to 110-4 may be installed toward each road adjacent to the intersection 20. Accordingly, each of the plurality of cameras 110-1 to 110-4 may obtain an image of a moving object on the road toward which the corresponding camera 110-1 to 110-4 is directed. Each of the plurality of cameras 110-1 to 110-4 may recognize a moving object on the road toward which the corresponding cameras 110-1 to 110-4 are directed from the corresponding image.

As shown in FIG. 1, a first road 21, a second road 22, a third road 23, and a fourth road 24 may be provided adjacent to the intersection 20. For example, the first camera 110-1 may be installed toward the first road 21 to obtain an image of a moving object, for example, a pedestrian 11, on the first road 21. For example, the second camera 110-2 may be installed toward the second road 22 to obtain an image of a moving object, for example, a car 12, on the second road 22. For example, the third camera 110-3 may be installed toward the third road 23 to obtain an image of a moving object, for example, a bicycle 13, on the third road 23. For example, the fourth camera 110-4 may be installed toward the fourth road 24 to obtain an image of a moving object, for example, a motorcycle 14, on the fourth road 24. In addition, the first camera 110-1, the second camera 110-2, the third camera 110-3, and the fourth camera 110-4 each detect a moving object, that is, a pedestrian 11, from the corresponding image. , cars (12), bicycles (13), and cars (14) can be recognized. The moving object may further include an electric kickboard or a wheel chair (see FIGS. 21A and 22B).

Although the drawing shows one camera being installed toward each road 21 to 24, two or more cameras may be installed.

Each of the plurality of cameras 110-1 to 110-4 may be a CCTV, IP camera, network camera, webcam, body cam, etc. Each type of camera may be a fixed camera, a rotating camera, a dome camera, a box camera, or a PTZ camera.

In an embodiment, the plurality of cameras 110-1 to 110-4 may have an object recognition algorithm and a circuit for driving the algorithm. Alternatively, an object recognition algorithm and a circuit that drives the algorithm may be built into the control unit 130.

The object recognition unit 110 including a plurality of cameras 110-1 to 110-4 may recognize moving objects on each road 21 to 24 adjacent to the intersection 20. For example, feature points may be extracted from images obtained from cameras 110-1 to 110-4, and a moving object may be recognized based on the extracted feature points. For example, a moving object can be recognized by comparing images obtained from cameras 110-1 to 110-4 with a comparison target image on a frame-by-frame basis. For example, the moving object may be recognized by detecting boundaries within the images obtained from the cameras 110-1 to 110-4 using an edge detection algorithm.

The display unit 120 may include a plurality of LED electronic signs (120-1 to 120-4). Each of these plurality of LED electronic signs (120-1 to 120-4) may be installed toward each road (21 to 24) adjacent to the intersection 20, but this is not limited. In an embodiment, each of the LED electronic signs (120-1 to 120-4) displays an image based on a plurality of LEDs, but other self-luminous devices other than LEDs, such as OLED, quantum dot, HUD ( Images can be displayed based on Head Up Display or holograms.

For example, the first LED display board 120-1 is installed toward the first road 21 and moves on roads other than the first road 21, that is, the second road 22 and the fourth road 24, respectively. By displaying risk information related to the object, risk information related to the moving object of each of the second road 22 and the fourth road 24 can be informed to pedestrians on the first road 21. For example, the second LED display board 120-2 is installed toward the second road 22 and moves on roads other than the second road 22, that is, the first road 21 and the third road 23, respectively. By displaying risk information related to the object, risk information related to the moving object of each of the first road 21 and the third road 23 can be informed to pedestrians on the second road 22.

For example, the third LED display board 120-3 is installed toward the third road 23 and moves on roads other than the third road 23, that is, the second road 22 and the fourth road 24, respectively. By displaying risk information related to the object, risk information related to the moving object of each of the second road 22 and the fourth road 24 can be informed to pedestrians on the third road 23. For example, the fourth LED display board 120-4 is installed toward the fourth road 24 and moves on roads other than the fourth road 24, that is, the first road 21 and the third road 23, respectively. By displaying risk information related to the object, risk information related to the moving object of each of the first road 21 and the third road 23 can be informed to pedestrians on the fourth road 24.

In an embodiment, the risk information may include an icon 31, an indicator 32, and speed information 33, as shown in FIG. 8. The icon 31 may include text, images, videos, etc. that symbolize or abstract the corresponding moving object. Although an arrow is shown as the indicator 32 in the drawing, it may include text, images, videos, etc. in a form that can indicate danger or the degree of danger to pedestrians. The speed information 33 refers to the actual driving speed of a moving object, but may also refer to a speed exceeding the speed limit specified on the plurality of roads 21 to 24 adjacent to the intersection 20.

The control unit 130 may be installed on the pillar 10a of the pole 10, but this is not limited. The control unit 130 may generally manage and/or control the road traffic object recognition safety system 100 according to the embodiment. The control unit 130 can control communication with a central control center (140 in FIG. 3) or other external devices, or transmission and reception of data and/information. The control unit 130 can control the object recognition unit 110 to recognize a moving object. The control unit 130 may control the display unit 120 to display risk information related to a moving object.

Meanwhile, risk information (FIGS. 18 to 23) related to a moving object may be displayed through the display unit 120.

18A and 18B are risk information related to pedestrians, and icons and arrows related to pedestrians may be displayed.

19A and 19B show risk information related to a car, and icons and arrows related to the car may be displayed.

20A and 20B are risk information related to bicycles, and icons and arrows related to bicycles may be displayed.

21A and 21B show risk information related to the electric kickboard, and icons and arrows related to the electric kickboard may be displayed.

22A and 22B are risk information related to a wheelchair, and icons and arrows related to the wheelchair may be displayed.

23A and 23B are risk information related to a motorcycle, and icons and arrows related to the motorcycle may be displayed. Figure 24 is an example diagram of a display unit that displays risk information.

As shown in FIG. 24, an icon for a car (including an arrow), an icon for an electric kickboard (including an arrow), and the text “Caution, Drive Slowly” may be displayed on the display unit 120. Pedestrians and drivers can easily recognize through the display unit 120 that a car and an electric kickboard on another road will enter the intersection, and can take actions such as waiting without entering the intersection.

Although speed information is not displayed in FIGS. 18A to 24, speed information may be displayed.

On the other hand, as shown in FIG. 1, at a residential intersection 20 on the back side of a street, the plurality of roads 21 to 24 adjoining the intersection 20 are narrow, and houses are concentrated around the intersection 20. It is difficult to secure visibility. In addition, there is no sidewalk adjacent to the roads (21 to 24), so it is a space to secure visibility.

In this case, the pedestrian must stop adjacent to the intersection 20, look left and right to check the presence or absence of dangerous moving objects on roads other than the pedestrian's entry road, and then pass through the intersection 20. If a child, a mentally challenged person, or an elderly person with poor sense of reason passes through the intersection 20 without stopping at the intersection 20 and checking the surrounding situation, the moving object may result in serious injury or, in severe cases, death. In particular, due to the rapid expansion of delivery services in recent years, various types of mobile objects such as motorcycles, electric kickboards, and electric scooters on the roads 21 to 24 are on each road 21 to 24 at intersections on residential streets on the main street. As cars are driven at high speeds, the risk of accidents is increasing.

In order to solve this problem, the embodiment uses a plurality of cameras (110-1 to 110-4) installed on the residential intersection (20) on the back side of the street to capture images on each road (21 to 24) of the intersection (20). A moving object can be recognized, and risk information related to the moving object can be displayed differently based on at least one of the moving object's speed, type, and entry distance. Accordingly, optimal information for each of the various situations of the moving object can be displayed. By displaying risk information, pedestrians can more clearly determine whether the moving object is dangerous to them based on the displayed optimal risk information, which does not help pedestrians determine the risk information of the moving object. By displaying only simple risk information, the problem that the risk of an accident is still not reduced can be solved.

Specifically, when a moving object is recognized on a specific road among the plurality of roads 21 to 24 of the intersection 20, the control unit 130 displays risk information related to the moving object through a plurality of LEDs of the display unit 120. It can be displayed on each of the electronic signs 120-1 to 120-4.

As shown in FIG. 4, when a pedestrian enters the intersection 20 through the first road 21, the pedestrian displays the first LED signboard 120-1 of the display unit 120 installed on the intersection 20. ), the risk information of a moving object on another road, that is, the second road 22 to the fourth road 24, may be displayed. The first LED signboard 120-1 may be installed toward the first road 21. The third road 23 is in direct contact with the first road 21 where pedestrians enter through the intersection 20, so that pedestrians can directly identify moving objects on the third road 23 with the naked eye. 3 Risk information related to a moving object on the road 23 may not be displayed on the first LED signboard 120-1.

Risk information related to moving objects on each of the second road 22 and the fourth road 24 may be displayed on the first LED signboard 120-1. For example, risk information related to cars on the second road 22 and risk information related to motorcycles on the fourth road 24 may be displayed on the first LED signboard 120-1.

In the drawing, the roads 21 to 24 that are in contact with the intersection 20 and correspond to each other are shown to be in contact with each other in a straight line through the intersection 20, so that the plurality of LED electronic signs 120-1 to 120 of the display unit 120 120-4) Risk information related to moving objects on each of the left and right roads may be displayed based on the road toward which the plurality of LED electronic signs (120-1 to 120-4) are heading.

Risk information related to moving objects on each of the first road 21 and the third road 23 may be displayed on the second LED signboard 120-2. For example, risk information related to pedestrians on the first road 21 and risk information related to bicycles on the third road 23 may be displayed on the second LED signboard 120-2.

Risk information related to moving objects on each of the second road 22 and the fourth road 24 may be displayed on the third LED signboard 120-3. The third LED display board 120-3 may display risk information related to cars on the second road 22 and risk information related to motorcycles on the fourth road 24.

Risk information related to moving objects on each of the first road 21 and the third road 23 may be displayed on the fourth LED signboard 120-4. The fourth LED display board 120-4 may display risk information related to pedestrians on the first road 21 and risk information related to bicycles on the third road 23.

Although not shown, the third road 23 is in curved contact with the first road 21 through which pedestrians enter and the intersection 20, so that pedestrians can directly identify moving objects on the third road 23 with the naked eye. If not present, risk information related to the moving object on the third road 23 may be displayed on the first LED signboard 120-1.

The non-illustrated reference numeral 112 shown in FIG. 3 is a speed dome camera, which can rotate 360 degrees and zoom in and out, allowing clear pictures of people or objects in the distance depending on the magnification of the lens. These speed dome cameras have the advantage of being able to quickly change the angle of view and having a wide operating range for the user. In the embodiment, the speed dome camera must rotate 360 degrees to obtain images for each of the plurality of roads 21 to 24 adjoining the intersection 20, so it is positioned lower than the plurality of cameras 110-1 to 110-4. can be installed in

The central control center 140 may be a site that comprehensively manages traffic facilities or traffic conditions in a specific area. The road traffic object recognition safety system 100 according to the embodiment may be controlled or managed in real time by the central control center 140, and communication or data may be performed with the road traffic object recognition safety system 100 according to the embodiment. can be sent and received.

Meanwhile, the control unit 130 may display risk information related to the moving object based on the speed of the moving object. The speed of the moving object may be the speed at which the moving object actually travels or may be a speed that exceeds the speed specified on the plurality of roads 21 to 24 that border the intersection 20.

The control unit 130 may display risk information related to the moving object based on the speed and type of the moving object. Types of moving objects may include pedestrians (11), bicycles (12), motorcycles (13), and cars (14). Types of moving objects may also include members that cannot move on their own, such as wheelchairs and carriers. In addition, all objects that can move may be included in the moving objects of the embodiment.

The control unit 130 may display risk information related to the moving object based on the speed and entry distance of the moving object. As shown in FIG. 7, the entry distance (L) is the separation distance between the center (CP) of the intersection 20 and the moving object, and may decrease over time when the moving object is traveling toward the intersection 20. there is.

The control unit 130 may display risk information related to the moving object based on the type and entry distance (L) of the moving object. The control unit 130 may display risk information related to the moving object based on the speed, type, and entry distance (L) of the moving object.

Hereinafter, the embodiment will be described in more detail through various operation methods.

1 to 5, the control unit 130 uses a plurality of cameras 110-1 to 110-4 of the object recognition unit 110 installed toward each road 21 to 24 of the intersection 20. ) can be used to recognize moving objects on the corresponding roads 21 to 24 (S210).

As shown in FIG. 1, a plurality of cameras 110-1 to 110-4 constituting the object recognition unit 110 are fastened to the support 101 and extend from the upper side of the pole 10. 10b) It can be installed on top. Each of the plurality of cameras 110-1 to 110-4 may acquire images of moving objects on each of the plurality of roads 21 to 24 bordering the intersection 20, and recognize the moving object from the corresponding images. .

As shown in FIG. 6, the cameras 110-1 to 110-4 are focused to include specific areas on the roads 21 to 24, thereby producing images including the specific areas on the roads 21 to 24. It can be obtained. At this time, the control unit 130 may set a virtual area 50 that is mapped to a specific area in the image including the specific area on the corresponding roads 21 to 24. That is, the virtual area 50 may be set to correspond to a specific area on the roads 21 to 24. Accordingly, when an image is acquired on the roads 21 to 24, the virtual area 50 is mapped to the image, so that a moving object in the image corresponding to the virtual area 50 can be recognized.

The control unit 130 may obtain at least one information among the speed, type, and entry distance (L) of the moving object (S220).

The control unit 130 receives images of moving objects on each road 21 to 24 obtained from each of the plurality of cameras 110-1 to 110-4, analyzes the images, and determines the speed and type of the moving object. and at least one information of the approach distance (L) can be obtained.

In an embodiment, through image analysis, at least one of the speed, type, and entry distance (L) of the moving object may be obtained through more precise analysis using an artificial intelligence function. The speed of the moving object can be obtained by tracking changes in the moving object by comparing it with the image input in frame units based on the comparison target image, but the speed of the moving object is not limited to this. Types of moving objects include pedestrians, cars, bicycles, electric kickstands, wheel chairs, motorcycles, etc., and the type of the object can be determined through object recognition. The entry distance (L) is calculated by mapping preset coordinates to the images acquired by the cameras 110-1 to 110-4 and tracking the moving object on the coordinates, thereby determining the entry distance (L) for the moving object. can be obtained.

The control unit 130 may display different risk information related to the moving object based on the at least one piece of information obtained (S230).

For example, the control unit 130 may display the expanded size or brightness of the icon 31 included in the risk information related to the moving object differently depending on the speed, type, and/or entry distance (L) of the moving object. The greater the speed of the moving object, the more dangerous the type of moving object is, or the closer the entry distance (L) of the moving object is to the intersection 20, the more the icon 31 expands or its brightness increases. You can.

For example, the control unit 130 may display the expanded size of the arrow included in the risk information related to the moving object differently depending on the speed, type, and/or entry distance (L) of the moving object. The greater the speed of the moving object, the more dangerous the type of moving object is, or the closer the entry distance (L) of the moving object is to the intersection 20, the larger the arrow may expand or the length of the arrow may become longer. there is.

For example, the control unit 130 may display the sequential blinking speed of the dots included in the risk information related to the moving object at different rates depending on the speed, type, and/or entry distance (L) of the moving object. A plurality of dots are included within the arrow, and these dots may be spaced apart from each other. These dots may blink sequentially, and the blinking speed may vary depending on the speed, type, and/or entry distance (L) of the moving object. For example, the greater the speed of the moving object, the more dangerous the moving object is, or the closer the entry distance (L) of the moving object is to the intersection 20, the faster the blinking speed at which these dots blink sequentially. However, there is no limitation to this.

Figure 9 is a flowchart explaining a method of differently displaying risk information related to a moving object depending on the speed of the moving object.

As shown in FIG. 9, the control unit 130 can recognize a moving object (S311). The control unit 130 may recognize a moving object from images obtained from each camera 110-1 to 110-4 of the object recognition unit 110, but this has been described above and will therefore be omitted.

The control unit 130 may obtain whether the speed of the recognized moving object is greater than the set specified speed (S312).

As described above, the speed of the moving object can be obtained by comparing the image to be compared with the image obtained in frames from the corresponding cameras 110-1 to 110-4.

The prescribed speed is the speed prescribed on the plurality of roads 21 to 24 in contact with the intersection 20. In FIG. 1, the prescribed speed for each of the plurality of roads 21 to 24 is shown as the same, but the plurality of roads ( 21 to 24) Each specified speed may be different.

The control unit 130 may display the icon 31 or indicator 32 differently depending on the degree to which the speed of the moving object becomes greater than the set standard speed (S313). Here, the icon 31 or indicator 32 may be included in the risk factors associated with the moving object.

As shown in FIG. 10, the regulation speed prescribed on the second road 22 is 30 km/h, while the actual driving speed of the motorcycle on the second road 22 is 50 km/h, and the regulation speed is 20 km/h. h may be exceeded. As the motorcycle on the second road 22 travels at a speed of 50 km/h or more, the motorcycle may exceed the prescribed speed. In this case, the icon 31 or indicator 32 may be displayed differently as the actual driving speed of the motorcycle exceeds the prescribed speed by 20 km/h, 40 km/h, 70 km/m, etc.

As shown in FIG. 11(a), when the motorcycle travels at a specified speed, the icon 31 and/or the arrow may be displayed at a default size. As shown in FIG. 11(b), when the driving speed of the motorcycle is less than the specified speed, the icon 31a and/or arrow 32a are displayed smaller than the size of the icon 31 and/or arrow set by default. It can be. As shown in FIG. 11(c), when the driving speed of the motorcycle is greater than the specified speed, the icon 31b and/or arrow 32b are displayed larger than the size of the icon 31a and/or arrow set by default. It can be.

Although not shown, the size of the icon 31a and/or the arrow may increase in proportion to the extent to which the motorcycle's driving speed exceeds the specified speed.

Meanwhile, when the speed of the recognized moving object is equal to or smaller than the set standard speed, the control unit 130 may constantly display the icon 31 or indicator 32 at a preset size (S314).

Figure 12 is a flowchart explaining a method of differently displaying risk information related to a moving object depending on the type of the moving object.

As shown in FIG. 12, the control unit 130 can recognize a moving object (S321).

The control unit 130 may obtain whether the moving object is a pedestrian through object recognition (S322).

When the moving object is a pedestrian, the control unit 130 may display an arrow including one dot 32-1, as shown in FIG. 13(a) (S323).

When the moving object is a bicycle (S324), the control unit 130 may display an arrow including two dots 32-1 and 32-2, as shown in FIG. 13(b) (S325). .

When the moving object is a motorcycle (S326), the control unit 130 may display an arrow including three dots 32-1 to 32-3, as shown in FIG. 13(c) (327). .

When the moving object is a car (S328), the control unit 130 may display an arrow including four dots 32-1 to 32-4, as shown in FIG. 13(d) (S329). .

In FIG. 13 , displaying arrows differently depending on the type of moving object is just one example, and can be displayed in various ways using the icon 31a, arrows, or other members.

Meanwhile, although not shown, the control unit 130 may display the icon 31a or indicator 32 differently as the entry distance (L) of the moving object decreases.

As shown in FIG. 7 , when the motorcycle on the second road 22 is traveling toward the intersection 20, the approach distance L may be reduced. The control unit 130 may display the icon 31a or arrow corresponding to the motorcycle differently as the approach distance L between the motorcycle and the center CP of the intersection 20 decreases.

Meanwhile, although not shown, the control unit 130 may display an icon to blink or dots within an arrow to blink along the direction indicated by the arrow, depending on the speed, type, or entry distance of the recognized moving object.

Figure 14 shows a plurality of moving objects entering a specific road.

As shown in FIG. 14 , a plurality of moving objects 11 to 14 may be traveling toward the intersection 20 on a specific road, for example, the fourth road 24 . When a plurality of moving objects 11 to 14 are recognized on the fourth road 24, the control unit 130 determines the moving object associated with the highest risk based on a preset risk order for the plurality of moving objects. Information can be displayed.

The risk order can be set based on various factors such as the average running speed of the moving object, the shape of the moving object, and the material of the moving object. For example, the risk order may be set in the order of cars, motorcycles, bicycles, and pedestrians, but this is not limited. In this case, as shown in FIG. 15 , risk information related to automobiles among automobiles, motorcycles, bicycles, and pedestrians traveling on the fourth road 24 may be displayed.

Alternatively, risk information related to motorcycles, which are next in risk order, may be displayed together with cars. Alternatively, risk information related to each of cars, motorcycles, bicycles, and pedestrians in the order of risk, such as cars, motorcycles, bicycles, and pedestrians, may be displayed in order from one side on each of the plurality of LED electronic signs (120-1 to 120-4). there is. For example, risk information related to cars is located on the left screen of each of the plurality of LED signs 120-1 to 120-4, and risk information related to pedestrians is located on the plurality of LED signs 120-1 to 120-4. ) are located on each right screen, and between these risk information, risk information related to motorcycles and risk information related to bicycles can be located in order from left to right.

Figure 16 is a flowchart explaining a method of determining whether to display risk information related to a moving object through tracking the movement of the moving object.

As shown in FIG. 16, the control unit 130 may set a plurality of virtual areas 51 to 53 on each road 21 to 24 of the residential intersection 20 (S331).

As shown in FIG. 17, recognition areas capable of recognizing moving objects are allocated on the fourth road 24, and the control unit 130 is configured to correspond to the recognition areas allocated on the fourth road 24. Virtual areas 51 to 53 may be set. As another example, the virtual areas 51 to 53 set in the control unit 130 may be directly allocated to the fourth road 24.

The control unit 130 may check whether an image has been acquired from each of the plurality of cameras 110-1 to 110-4 of the object recognition unit 110 (S332).

When an image is acquired from each of the plurality of cameras 110-1 to 110-4, the control unit 130 may map a plurality of preset virtual areas 51 to 53 to the corresponding image (S333) ).

The control unit 130 can track the movement of the moving object by recognizing the moving object that inputs and outputs from each of the plurality of virtual areas 51 to 53 (S334). The control unit 130 may obtain whether to display risk information related to the moving object (S335).

As shown in FIG. 17, a motorcycle on the fourth road 24 may drive toward the intersection 20 or may take a detour before entering the intersection 20.

The control unit 130 can track the movement of the motorcycle by checking whether the motorcycle inputs or outputs to each of the plurality of virtual areas 51 to 53. For this purpose, an image of a motorcycle on the fourth road 24 may be acquired in frame units by the fourth camera 110-4. A plurality of virtual areas 51 to 53 are mapped to each of the images acquired in frame units, so that the input and output of the motorcycle on the corresponding virtual areas 51 to 53 can be confirmed.

For example, if a motorcycle is recognized temporally at the starting point of the first virtual area 51 in an image acquired in frame units and crosses the first virtual area 51 and passes the end point of the first virtual area 51, the motorcycle It may be determined that it is outside the first virtual area 51. The starting point of the first virtual area 51 may be a position that is encountered first in the driving direction of the motorcycle, and the end point of the first virtual area 51 may be a position that is encountered later than the starting point of the first virtual area 51 . Thereafter, when the motorcycle crosses the second virtual area 52 and leaves the end point of the second virtual area 52, it may be determined that the motorcycle has left the second virtual area 52.

Afterwards, if the motorcycle is not recognized as crossing the starting point of the third virtual area 53 and the third virtual area 53, the motorcycle detours between the second virtual area 52 and the third virtual area 53. As such, it can be determined that the vehicle no longer enters the intersection 20. If the motorcycle crosses the third virtual area 53 and leaves the end point of the third virtual area 53, it may be determined that the motorcycle leaves the third virtual area 53 and enters the intersection 20.

If it is determined that risk information related to the moving object is displayed, the control unit 130 displays the risk information related to the moving object on the plurality of LED signs 120-1 to 120-4 of the display unit 120. (S336).

As shown in FIG. 17, when the motorcycle detours between the second virtual area 52 and the third virtual area 53 and is not recognized in the third virtual area 53, the control unit 130 detects the motorcycle. It is determined that the vehicle is entering the intersection 20, and risk information related to the motorcycle may be displayed on the plurality of LED signs 120-1 to 120-4 of the display unit 120. As described above, the second road 22 is in direct contact with the fourth road 24 on which the motorcycle travels through the intersection 20, so that a pedestrian on the second road 22 can directly see the fourth road 22 with the naked eye. 24) If the motorcycle and its movement can be identified, risk information related to the motorcycle may not be displayed on the second LED signboard 120-2 of the display unit 120.

In FIG. 17, the third virtual area 53 is located after the last detour that a moving object on the fourth road 24, for example, a motorcycle, can detour, and after the third virtual area 53, the moving object can detour. It may be a position where it cannot be done. If there is still a detour that the moving object can bypass after the third virtual area 53, the fourth virtual area or additional areas after the fourth area may be further set.

Meanwhile, due to weather conditions, it may be difficult to secure visibility even during the day. For example, when it rains with thunder and lightning, when it is dark in the early evening, or when there are no street lights in the evening. In this case, when an image is acquired from each of the plurality of cameras 110-1 to 110-4, the clarity of the image deteriorates, making it difficult not only to recognize the object but also to obtain the moving speed of the object, the type of the object, and the entry distance (L). This may be difficult or impossible.

To solve this problem, in an embodiment, each of the plurality of cameras 110-1 to 110-4 of the object recognition unit 110 may be an infrared or color camera as well as a CCTV. That is, each of the plurality of cameras 110-1 to 110-4 may be an infrared or color camera combined CCTV. Accordingly, when it is difficult to secure visibility, each of the plurality of cameras 110-1 to 110-4 is changed to infrared or color mode, so that infrared or color images are displayed on each road 21 to 24 of the intersection 20. A moving object on the corresponding roads 21 to 24 may be recognized based on the obtained infrared or color image.

Meanwhile, when object recognition is not easy with each of the plurality of cameras 110-1 to 110-4 constituting the object recognition unit 110 described above, an additional device capable of supporting object recognition, such as a radar, An ida sensor, a distance measurement sensor, a motion detection sensor, an infrared or color camera, a thermal imaging camera, a vision camera, a telephoto camera, etc. may be provided. Accordingly, object recognition can be performed more clearly and accurately by these additional devices in addition to the plurality of cameras.

The above detailed description should not be construed as restrictive in any respect and should be considered illustrative. The scope of the embodiments should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the embodiments are included in the scope of the embodiments.

Claims (10)

an object recognition unit that includes a plurality of cameras installed toward each road at a residential intersection and recognizes moving objects on the road;
A display unit installed toward each road of the residential intersection and including a plurality of LED electronic signs adjacent to the plurality of cameras, respectively; and
When a moving object is recognized on a specific road among the plurality of roads at the intersection, a control unit that displays risk information related to the moving object on each of the plurality of LED signs,
When a plurality of moving objects are recognized as driving on one specific road, the control unit displays information related to the moving object with the highest risk in case of an accident based on a preset risk order for the plurality of moving objects, and ,
The plurality of cameras are installed together with the plurality of LED electronic signs on a support,
Road traffic object recognition safety system. According to paragraph 1,
The control unit,
A road traffic object recognition safety system that differently displays risk information related to the moving object based on information on the type of the recognized moving object. According to paragraph 2,
Risk information associated with the moving object includes icons and indicators corresponding to the moving object,
Road traffic object recognition safety system. delete According to paragraph 3,
Types of the moving objects include pedestrians, bicycles, electric kickboards, motorcycles, and cars,
The control unit,
Displaying the icon or the indicator differently depending on the type of the moving object,
Road traffic object recognition safety system. delete delete delete According to any one of claims 1 to 3,
The control unit,
Map a plurality of virtual areas set for each road in the residential area to images obtained from each of the plurality of cameras,
Tracking the movement of a moving object on the mapped virtual areas to determine whether to display risk information related to the moving object on each of the plurality of LED electronic signs,
When it is determined that the moving object is entering the intersection, displaying risk information related to the moving object on each of the plurality of LED electronic signs,
Road traffic object recognition safety system. According to any one of claims 1 to 3,
When it is raining and dark or there are no street lights in the evening, the camera operates in infrared or color mode,
The object recognition unit,
Recognizing moving objects on the road based on infrared or color images acquired by the camera,
Road traffic object recognition safety system.

Images