KR101117071B1 - Energy saving apparatus for crosswalk safty lighting adaptively according to various conditions - Google Patents

Abstract

PURPOSE: A situation adaptive lighting type energy saving crosswalk safety lighting apparatus is provided to project light near a pedestrian by rapidly detecting the pedestrian, thereby preventing a car accident in beforehand by alerting the pedestrian and drivers. CONSTITUTION: A crossing situation sensing device(10) detects a crossing permission state of a pedestrian. A lighting lamp(30) uniformly projects light on a crosswalk region by controlling light projecting directions of a plurality of lamps. The lighting lamp additionally projects light near the crosswalk region. A control apparatus(40) transfers local lighting within the crosswalk region towards a crossing direction of the pedestrian. A camera(20) records the crosswalk region and the region near the crosswalk region.

Description

ENERGY SAVING APPARATUS FOR CROSSWALK SAFTY LIGHTING ADAPTIVELY ACCORDING TO VARIOUS CONDITIONS}

According to the present invention, when the green light of a pedestrian traffic light is turned on, the local lighting position on the crosswalk is adjusted according to the remaining time to induce the pedestrian crossing safely and achieve a power saving effect. The present invention relates to a situational lighting-type energy-saving crosswalk safety lighting device that gives everyone an alarm to prevent collision accidents.

In recent years, by installing lighting lights that focus on pedestrian crossings, the location of pedestrian crossings can be confirmed, and the driver can correctly recognize pedestrians passing through the pedestrian crossing or near pedestrian crossings. In this way, the accident rate at the crosswalk could be lowered. In addition, when the pedestrians walked the crosswalk, the surroundings of the pedestrians were brightened to recognize the pedestrians more accurately, and only when there were pedestrians, the power was reduced by the lighting of the lamps.

However, the conventional pedestrian crossing lighting was able to be interlocked with the pedestrian traffic light, but still serves only as a light and was not configured to allow the driver to recognize the remaining time of the pedestrian traffic light. In other words, when the pedestrian traffic light is recently installed, a device for informing the remaining time is also provided. However, since the remaining time notification device is for pedestrians, it is difficult for the driver to recognize it. It was also difficult to recognize due to low visibility and carelessness. From the perspective of the pedestrian, it is difficult to accurately determine the speed at which the crossing time is to be seen, but in this case, the driver could not see the pedestrians walking late, but only the traffic lights, leading to an accident.

In addition, in order to change the lighting of the crosswalk lighting according to the presence or absence of pedestrians, pedestrians must be detected without omission. To this end, a background image is accurately acquired from an image taken by a camera, and the image is changed every frame of the image. Acquire the background image by adapting to the change of background.

As a method of acquiring a background image, there is an adaptive Gaussian Mixture model method that can cope with a change in brightness such as shadows and highlights. The adaptive Gaussian mixture model models a pixel value of each pixel by successively input image frames as a Gaussian distribution, and determines whether an object appears according to the mean and covariance of the modeled Gaussian distribution, wherein the current frame The degree of deviation from the mean value of the Gaussian distribution is determined based on the covariance value to detect whether the pixel is caused by an object or a pixel due to a change in the background. The adaptive Gaussian mixture model then updates the mean and covariance of the Gaussian distribution with each frame input. Accordingly, the adaptive Gaussian mixture model method appropriately copes with the change in the brightness of the background and can detect the appearance of an object probabilisticly and accurately.

However, this adaptive Gaussian mixture model effectively adapts to the passage of a moving object because the moving object passes through the specified area because the influence of the pixel value by the moving object is small and its effect disappears after the moving object disappears. For example, when a moving object stops in a designated area for a long time, the pixel value caused by the moving object greatly affects the Gaussian distribution, and thus it is not accurately modeled as a background image. In addition, when the object is moved again, the Gaussian distribution of the pixel corresponding to the stop position is greatly influenced by the pixel value of the object, so that the object can be modeled as an original background only after the object is updated for a long time. . Therefore, modeling a background image with an adaptive Gaussian mixture model requires a method of rapidly updating the real background even if the object is stopped and moved for a long time in the surveillance area.

In particular, the pedestrian waiting on the sidewalk to cross the crosswalk waits until it turns into a walking signal, so the image of the waiting pedestrian remains in the adaptive Gaussian mixture model as an afterimage, and the afterimage gradually disappears, so that the adaptive Gaussian mixing There is a concern that the pedestrian may not be detected accurately with the background image modeled as a model.

Accordingly, an object of the present invention is to serve as a lighting lamp, and also to visually inform the pedestrian and the driver of the remaining time allowed for the crossing of the pedestrian traffic light, so that the driver can safely cross the pedestrian while accurately indicating the remaining time. In this regard, according to the present invention, it is to provide a situation-adaptive lighting energy-saving pedestrian crossing safety lighting device that can also save power.

Another object of the present invention is to detect a pedestrian by modeling a background image with an adaptive Gaussian mixture model, to quickly remove the effect of the pedestrian in the crossover atmosphere and to model it as an accurate background image. It is to provide a situational adaptive lighting energy-saving crosswalk safety lighting device.

The present invention for achieving the above object, in the lighting device is installed on the crosswalk, the cross-scenario sensing device for sensing whether or not the pedestrian crossing allowed; Illumination 30 to uniformly illuminate the crosswalk area (A) by varying the irradiation direction of the plurality of lamps 31, and to control the lighting of each of the lamps 31 individually; When the crossing permit signal is received from the crossing situation sensor 10, the lighting position on the crosswalk area A is moved toward the crossing direction of the pedestrian, but at a predetermined speed according to a preset crossing allowance time. And a control device 40 for controlling the lamp 30 to illuminate.

The crossing situation sensing device 10 is installed on the pedestrian traffic light 5, characterized in that the photosensor for detecting the green lighting.

The crossing situation sensing device 10 is characterized in that the operation switch that the pedestrian can operate.

The control device 40 is equipped with a day and night sensing device 200 for detecting the night, and when the day and night sensing device 200 detects that the night is turned on the lamps of the lamp 30 in the taillight lighting state , The lamps corresponding to the lighting position to be moved at the time of the green light of the pedestrian traffic light (5) is characterized in that the lighting in a fully lit state.

The lamp 30 is configured to include a cross-waiting area (B) on the sidewalk in contact with the crosswalk, the control device 40, the cross-waiting area (B) in a fully lit state at night It characterized in that to control the lamp 30 to be maintained.

The situational lighting energy saving pedestrian crossing safety lighting device, characterized in that further comprises a camera 20 for photographing the crosswalk area (A) and the cross-waiting area (B).

The control device 40 is provided with an image captured by the camera 20 is provided with a modeling unit 300 for detecting whether the situation is a pedestrian in the crosswalk area (A) and the cross-waiting area (B), the control device The situation-aware modeling unit 300 of 40 calculates an average and covariance according to the Gaussian distribution of each pixel in the initial continuous frame of the image obtained when the situation-adaptive lighting energy saving pedestrian crossing safety lighting device is installed. An initialization module 310 for initializing the adaptive Gaussian mixture model per pixel for the background image; A foreground region extraction module 320 for extracting a foreground region by a pedestrian from an image frame after initialization based on the pixel-specific adaptive Gaussian mixture model; A history management module 340 for managing the location and time history of the extracted foreground area; If the foreground area is not extracted and the extracted foreground area is moving, the Gaussian model for each pixel of the background is updated, and if the extracted foreground area stops for more than a preset time and then moves, And a background update module (350) for initializing the average value of the adaptive Gaussian mixture model to the current pixel value. The control device (40) includes a history of the foreground area managed by the history management module (140). By detecting the position and movement of the pedestrian on the basis, it is characterized in that to control the lamp 30 so as to illuminate the surroundings of the pedestrian fully lit.

The situational modeling unit 300 of the control device 40, when modeling the image of the crosswalk area (A) as an adaptive Gaussian mixture model, in accordance with the time that only the area that is fully lit, fully lit Characterized by updating.

Therefore, the present invention constituted as described above, the pedestrians to cross to match the brightly lit area on the crosswalk, and also visually and accurately recognize the remaining allowable time for the pedestrians, to ensure the safe crossing of pedestrians and smooth communication of the vehicle In addition, since only necessary parts are brightened by local lighting, power saving effects can be obtained.

In addition, the present invention accurately detects pedestrians in a state where the cross-waiting area is brightly illuminated, and accurately removes afterimages caused by pedestrians waiting in the cross-waiting area when modeling a background image with an adaptive Gaussian mixture model. Pedestrians are detected without omission based on the modeled background image, and the surroundings of the detected pedestrians are brightened, so there is no accident due to omission of pedestrian detection.

1 is a block diagram of a situation-adaptive lighting energy-saving crosswalk safety lighting device according to a first embodiment of the present invention.
Figure 2 is an exemplary view showing a situation-adaptive lighting energy-saving crosswalk safety lighting device according to the first embodiment of the present invention installed on the crosswalk.
3 is a view for explaining the lighting method of the lamp 40 in the situation-adaptive lighting energy-saving pedestrian crossing safety lighting apparatus according to the first embodiment of the present invention.
4 is a view showing another form of the lamp 40 in the situational lighting-type energy-saving pedestrian crossing safety lighting apparatus according to the first embodiment of the present invention.
Figure 5 is a perspective view of a pedestrian traffic light showing the installation form of the cross-situation sensing device 10 consisting of a photosensor 11 in the context-sensitive lighting energy-saving crosswalk safety lighting device according to the first embodiment of the present invention.
Figure 6 is a flow diagram of a crosswalk illumination method made by a situation-adaptive lighting energy-saving crosswalk safety lighting device according to a first embodiment of the present invention.
7 is a block diagram of a situational awareness modeling unit 300 in the situational lighting energy-saving crosswalk safety lighting apparatus according to the first embodiment of the present invention.
8 is a flow chart of the adaptive background image modeling step performed by the situational awareness modeling unit 300 in the context-sensitive lighting energy-saving pedestrian crossing safety lighting apparatus according to the first embodiment of the present invention.
Figure 9 is a block diagram of a situation-adaptive lighting energy-saving crosswalk safety lighting device according to a second embodiment of the present invention.
10 is an exemplary view showing a situation-adaptive lighting energy-saving pedestrian crossing safety lighting device installed in a pedestrian crossing according to a second embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. It should be noted that, in the drawings, the same reference numerals are used to denote the same or similar components in other drawings. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the following description of the present invention, detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

1 is a block diagram of a situation-adaptive lighting energy-saving crosswalk safety lighting apparatus according to a first embodiment of the present invention, Figure 2 is an exemplary view showing a state in which the first embodiment of the present invention is installed in a crosswalk, 3 and 4 are views for explaining the lighting method of the lamp 40, Figure 5 is a perspective view of a pedestrian traffic light showing the installation state of the cross-sensor 10 consisting of the photosensor 11.

1 to 5, the situation-adaptive lighting energy-saving pedestrian crossing safety lighting device according to an embodiment of the present invention, the prop erected near the road boundary stone (3) between the road (1) and the sidewalk (2) Installed in the pedestrian traffic light (A) and the cross-waiting area (B) on the sidewalk (2) connected to the crosswalk (4) designated by partitioning the pedestrian crossing (4) displayed on the road (1). 5) A device for lighting at night by adjusting the position of the brightly illuminated according to the lighting state or the presence of pedestrians, for this purpose, the cross-situation sensor 10, the camera 20, the lamp 30 and the control device 40 It is configured to include.

In the embodiment of the present invention, because the width of the road (1) is large, the posts are installed on both sides of the crosswalk 4, respectively, the lamps 30, 30a and cameras 20, 20a are installed on each post, The crossing situation sensor 10 was installed in any one of the pedestrian traffic lights 5 and 5a provided at both sides of (4). Then, the control device 40 is installed on any one of the two posts, so that the control device 40 is turned on and the cameras 20 and 20a of the pedestrian traffic light detected by the crossing situation sensor 10. The lighting of the both side lights (30, 30a) is controlled based on the image photographed at. On the other hand, when the width of the road (1) is narrow, the supporter may be installed only on either side of both sides of the crosswalk 4, the camera 20, the lamp 30 and the control device 40 may be provided.

The crossing situation sensor 10 detects a crossing permit signal indicating that a pedestrian may cross the crosswalk and transmits the signal to the control device 40. In the first embodiment of the present invention, the pedestrian traffic light 5 It consists of a photosensor 11 for detecting a green lighting state that is a crossing permit signal. That is, the transverse situation sensor 10 composed of the photosensor 11 is a red light emitting portion 5-1, the green light emitting portion (5-2) and the screen covering the light covering the upper side as shown in FIG. In the pedestrian traffic light (5) provided, it is attached to the lower portion of the light shade shade (5-3) provided in the green light emitting portion (5-2) to detect the light emitted from the green light emitting portion (5-2). . In addition, the photosensor 11 is electrically connected to transmit the detection signal to the control device 40 to be described later, or is connected by short-range wireless communication. As a method of detecting the lighting state of the pedestrian traffic light 5, the photosensor 11 is attached to the lower part of the light shade shade provided in the red light emitting unit 5-1 to turn off the red light emitting unit 5-1. You can also detect the green lighting status by detecting. In addition, the lighting state information of the pedestrian traffic light detected by the crossing situation sensor 10 is transmitted to the control device 40, the information transmission method may be connected by a signal line or wireless communication method.

Meanwhile, the crossing situation sensor 10 detects a green lighting state in an image captured by a camera (not shown) and a camera (not shown) photographing a pedestrian traffic light 5 instead of the photo sensor 11 by image processing. The image processing unit (not shown) may be configured as a combination. When the transverse situation sensing device 10 is configured using a camera (not shown), the camera (not shown) described above is not installed separately, but instead installed to detect a pedestrian. By including the pedestrian traffic light (5) in the imaging range of the, the image corresponding to the pedestrian traffic light (5) when processing the image captured by the camera 30 to be described later in the situation modeling unit 300 of the control device 40 It may be configured to detect the lighting state in the partial image.

The camera 20 photographs the crosswalk areas A and Aa partitioned by the crosswalk 4 and the cross-waiting areas B and Ba partitioned on the sidewalk 2 that is in contact with the crosswalk 4. To be installed. In the embodiment of the present invention shown in FIG. 2, two cameras are provided on both shores, and one of the two cameras adjacent to the partitioned area divided by the center dividing line on the road 1 represents a crosswalk area. The other one was taken with the cross-waiting area, which was used with two cameras in consideration of the camera's imaging angle. However, if the wide-angle camera can take the crosswalk area and the cross-waiting area, only one camera is installed on each prop. You may also do it. In addition, the image photographed by the cameras 20 and 20a is transmitted to the control device 40.

The lamp 30 is provided to illuminate the pedestrian crossing areas A and Aa and the cross-waiting areas B and Ba, and is configured to control the lighting according to a control signal transmitted from the control device 40. In the embodiment of the present invention shown in Figure 2, by dividing to both sides based on the center dividing line on the road (1) to illuminate the crosswalk area and the cross-waiting area in the partitioned direction to illuminate the lamp 30, 30a for each strut Installed. The lighting lamp 30 will be described in detail with reference to FIG. 3. The pedestrian crossing area A and the cross-waiting area B are configured to be within a range of irradiation angle. That is, the lamp 30 has a structure in which the plurality of lamps 31 are evenly attached to the lamp attachment surface 32 curved downward, so that the irradiation directions of the lamps 31 are different from each other. The cross-sectional area A and the cross-waiting area B are illuminated by different lamps 31 by the irradiation angles of the beams. In this case, when all the lamps 31 provided in the lamp 30 are completely turned on, it is preferable to illuminate the pedestrian crossing areas A and Aa and the cross-waiting areas B and Ba with uniform illuminance. For example, referring to FIG. 3, the lamp illuminating the crosswalk area A and the lamp illuminating the crosswalk area B are separated, and the crosswalk area A is the crosswait area. When divided into four subdivisions (A1, A2, A3, A4) according to the distance from (B), the lamps which concentrate illumination of each subdivision (A1, A2, A3, A4) are also separated.

In addition, the lamps 31 provided in the lamp 30 may be individually lit, and the lamps illuminating the crosswalk areas A and Aa may change the brightness. That is, the lamp 30 allows only the lamps indicated by the control device 40 among the lamps for illuminating the pedestrian crossing areas A and Aa to be completely turned on and the remaining lamps to be turned off or turned off. For example, in FIG. 3, only the small areas indicated by the control device 40 among the small areas A1, A2, A3, and A4 may be intensively illuminated by full illumination, and the remaining small areas may be illuminated by the tail light lighting. Here, the full lighting of the lamp is to turn on the lamp at the rated value, the taillight lighting is to light at a lower brightness than the full lighting and the brightness of the taillight lighting is set in advance.

On the other hand, the lamp 31 provided in the lamp 30 may be composed of, for example, an LED lamp to ensure low power and long life. In addition, a technique of adjusting the brightness of the lamp 31 may adopt, for example, a method of controlling power supplied to the lamp, and since such technique is a known technique, a detailed description thereof will be omitted.

In addition, in the above description, the brightness of each lamp is controlled by lighting the lamp in full lighting or taillight lighting, but as another method, the lighting lamp 30 may be configured to be turned on in the taillight state by reducing the number of lamps that are fully lit. have.

In addition, in FIG. 3, the surface on which the lamps are mounted is shown in a curved shape, but the shape of the lamp 30 is not limited to the exemplary view of FIG. 3. If the structure can illuminate evenly, it is satisfied.

4 shows that the lamp 30 may be configured in other forms. The lamp 30 according to FIG. 4 has a curved shape in a downward direction and has a support 34 installed on the arm of the support, and a plurality of partial lighting lamps sequentially fixed along the curved bottom of the support 34 ( 33). At this time, the plurality of partial lamps 33 illuminate the crosswalk area (A) and the cross-waiting area (B) while illuminating differently concentrated areas. In the embodiment shown in FIG. 4, four sub-zones A1 and A2 partitioned in the crosswalk area A according to the distance from the cross-wait area B, a partial lamp for intensively illuminating the cross-wait area B. FIG. , A3, A4) consists of four partial lamps for intensive illumination respectively. Therefore, when the lamp of the type shown in FIG. 4 is used, it is possible to control the partial lamps 33 individually to select a position for intensive illumination, and to control the brightness of each of the partial lamps 33. What is necessary is just to control in the way which only one part of the lamp 31 provided in 33 is made to light. That is, the method controlled by the control device 40 to be described later to turn on the partial lamp to intensively illuminate the cross-waiting area (B) with full lighting, and partial lamps to intensively illuminate the crosswalk area (A) In the state where the tail light is turned on, when the partial lamps intensively illuminating the pedestrian crossing area A are sequentially turned on, the light is sequentially illuminated from the small area A1 to the small area A4. Meanwhile, the lamp described with reference to FIG. 4 is a modified form of the lamp described with reference to FIG. 3, but is configured to function substantially the same in terms of localized illumination and adjustment of illumination brightness.

In addition, in FIG. 2, the lamps 30 and 30a are provided on both sides of the crosswalk, respectively, but only one lamp may be installed when the width of the road is narrow.

The control device 40 is a crossing situation determination unit 100 to determine whether the green light indicating the crossing permission by checking the lighting state of the pedestrian traffic light 5 sensed by the crossing situation sensing device 10, night. Day and night sensing device 200 for detecting, the situation modeling unit 300 to determine the presence and location of the pedestrian from the image taken by the camera 20, the data storage unit for storing and storing the image taken when there is a pedestrian 500, and an illumination control unit 400 for controlling the lamp 30 according to the lighting state of the pedestrian traffic light 5 and whether the pedestrian is present or not at night.

Here, the day and night sensing device 200, in order to control the lighting lamp 40 to be turned on only at night, it is satisfied if configured as a general day and night detection sensor for detecting the night. Here, the day and night sensing device 20 may be configured as a timer that sets the day and night time zones for each season, but is not limited to the day and night time zone to illuminate the pedestrian crossing with the lights 40, cross the crosswalk in the heavy clouds Since it is preferable to illuminate, it is preferable to configure an optical sensor that senses the surrounding brightness.

In storing the image, the data storage unit 500 may store the shooting time and lighting state information of the pedestrian signal.

The situation-aware modeling unit 300 is a configuration for detecting a pedestrian by reading an image photographed by the camera 20, which will be described in detail with reference to FIGS. 7 and 8 below.

First, with reference to FIG. 3, the lighting method of the lighting lamp 40 made by the control apparatus 40 provided with the said lighting control part 400 is demonstrated in detail.

Figure 6 is a flow chart of a crosswalk illumination method made by a situation-adaptive lighting energy-saving crosswalk safety lighting device according to an embodiment of the present invention.

First, the control device 40 determines the day and night by the signal detected by the day and night sensor 200 (S10) during the day to continuously check the signal detected by the day and night sensor 200, the night comes When the surrounding becomes dark (S11), the lighting control operation of the lamp 30 starts (S20).

The control device 40 starts to check whether the pedestrian traffic light 5 is in the green light state when the lighting control operation is started (S30, S31), and simultaneously, to the camera 20. The pedestrian is detected in the captured image and the position of the detected pedestrian is started to check with the situation modeling unit 300 (S70, S80, S90).

In addition, the control device 40 causes the lamps for illuminating the cross-waiting area B among the lamps of the lamp 30 to be turned on in a fully lit state when the lighting control operation is started, and to illuminate the crosswalk area A. FIG. The lamps for the lamps are turned on in the tail light lighting state. Accordingly, since the street lamp 30 illuminates the pedestrian crossing area (A) by the lighting of the taillight in a situation where the pedestrian is not in the pedestrian crossing area (A), it is possible to reduce power consumption than to unnecessarily brighten, and also bright light to the car driver The crosswalk area A can be perceived even by low roughness without giving the fatigue caused by. In addition, the street lamp 30 serves as a street lamp by brightly illuminating the cross-waiting area B, which is an area on the sidewalk 2, with full lighting, thereby accurately detecting a pedestrian in the bright-crossing standby area B. To be able.

Here, the monitoring of the pedestrians (S70, S80, S90), the background image of the crosswalk area (A) and the cross-waiting area (B) taken by the camera 20 is adaptive Gaussian mixture model (AGMM: Adaptive) Adaptive background image modeling step (S70) to check the presence, location and movement of the pedestrian based on the position and time history of the foreground area displayed on the image by the pedestrian in the process of modeling and Gaussian Mixture Model If the pedestrian is detected by the adaptive background image modeling step (S70) (S80), including a pedestrian position determination step (S90) for determining the pedestrian position on the crosswalk area (A) and the cross-waiting area (B) Is done. The situation-aware modeling step S70 will be described in detail with reference to FIG. 8.

In addition, the control device 40 determines the lighting form of the lamp 30 according to the green light of the pedestrian traffic light 5 or the detection and position of the pedestrian, and controls the lamp 30 to light up according to the determined lighting form.

When explaining a specific lighting control method, when the lighting state of the pedestrian traffic light 5 is a green light indicating crossing permission, the control device 40 determines the lighting position on the crosswalk area A to be brightly lit with full lighting. The lamp 30 is controlled to move in a direction crossing the crosswalk at a predetermined speed. That is, in a state where the lamps illuminating the crosswalk area (A) is illuminated by the tail light lighting, starting from the lamp illuminating the area close to the cross-waiting area (B) to the lamp illuminating the other area, sequentially turn on at full intervals After the predetermined time, the lamp which has been completely turned on is returned to the tail light lighting state.

Referring to FIG. 3, for example, all lamps illuminating the crosswalk area A are lit in the tail light state, starting from the subregion A1 and then fully illuminated in the order of subregions A2, A3, and A4. . As a result, the crosswalk area (A) moves from the A1 subregion to the A4 subregion as a whole, in the taillights, so that the pedestrian crossing the location of the bright subregion can see the remaining time of the crossing allowance. In this way, the pedestrian can be guided to traverse according to the position of the brightly lit subregion. That is, the situation-adaptive lighting energy-saving pedestrian crossing safety lighting device according to the present invention not only completely illuminates all the lamps of the lamp 30 but also illuminates with local lights, thereby achieving a power saving effect, and remaining time rather than crossing the area ( By using the brightly illuminated location in A) to visually inform and induce crossings, accident prevention effects can be obtained that can prevent accidents due to lack of awareness of the remaining time. Here, the movement speed of the position to be illuminated by full lighting can be determined by dividing the crossing distance of the pedestrian crossing 4 at time intervals lit by the green lighting in the pedestrian traffic light 5.

In addition, the control device 40 may determine an area to brighten according to the pedestrian traffic light 5, but the adaptive background image modeling step (S70) and the pedestrian position determination step (S90) including the situation-aware modeling unit 300. By detecting the position of pedestrians), the surroundings of pedestrians who cannot follow the brightly lit areas according to the pedestrian traffic lights (5) are also brightly lit to facilitate the crossing of pedestrians and to prevent accidents caused by cars. In other words, irrespective of the lighting state of the pedestrian traffic light 5, since the cross-waiting area B is brightly lit with full lighting, the camera 20 photographs and processes the image. In the waiting area (B) can be detected without omission, it is to brighten the surroundings of the pedestrian according to the position and movement of the pedestrian. In addition, since the surroundings of the pedestrians are brightly illuminated according to the movement of the pedestrians, the pedestrians can be detected more accurately than the tail lights when the pedestrians are detected by image processing. At this time, the brightly illuminated area is biased in the direction of movement of the pedestrian, so that the front portion is wider than the back of the pedestrian.

As described above, the process of lighting the lighting 30 according to the presence or absence of a pedestrian traffic light and a pedestrian is continuously performed only at night (S50), and ends until the next night time arrives (S60).

On the other hand, since pedestrians can enter from both sides of the crosswalk and cross in opposite directions, as shown in FIG. 2, when the lights are respectively installed on both sides of the crosswalk, the pedestrian is illuminated with full lighting. The location will return to the sidewalk after moving to the center line of the road. And, if the lighting is installed only on one side of the crosswalk to illuminate the entire crosswalk with a single lamp, the positions of the lights are fully lit, respectively, on each side of the crosswalk, facing each other, close to each other, and after crossing on the center dividing line, the opposite sides. It will take the form of moving to.

Next, the adaptive background image modeling step S70 including the situation awareness modeling unit 300 and the situation awareness modeling unit 300 will be described in detail.

6 is a block diagram of a situation awareness modeling unit 300 in the situational lighting energy-saving crosswalk safety lighting apparatus according to an embodiment of the present invention, Figure 7 is made by the situation awareness modeling unit 300 It is a flowchart of the adaptive background image modeling step (S70).

Prior to explaining the adaptive background image modeling method performed by the context aware modeling unit 300, an adaptive Gaussian Mixture Model will be described.

The adaptive Gaussian mixture model models the distribution of pixel values as Gaussian probability distributions for each pixel in a frame of a continuously captured image, wherein the average and covariance parameters of the Gaussian model for each pixel are calculated every frame. Update every time. Here, although one Gaussian distribution may be used for each pixel, K Gaussian distributions, which are usually designated as 3 to 5, are used to approximate each pixel value representing a recently observed distribution. This will be described in detail.

First, the statistical information on the pixel value of the pixel {x, y} at a certain time t is assumed to be [Equation 1] below.

[Equation 1]

는 K 가우시안 모델이며, 시간 t에서

의 값을 갖는 입력영상의 화소에 대한 확률은 아래의 [수학식 2]로 추정된다.Here, I is an input image,

Is the K Gaussian model and at time t

The probability of the pixel of the input image having the value of is estimated by Equation 2 below.

[Equation 2]

와

로 표시하였으며,

는 가우시안 확률밀도함수(Gaussian probability density function)로서 아래의 [수학식 3]으로 구한다.Where K is the number of Gaussian models for each pixel, and the mean and covariance matrix of the i th Gaussian mixture model at time t

Wow

,

Is a Gaussian probability density function obtained from Equation 3 below.

&Quot; (3) "

는 시간 t에서 i번째 가우시안 모델의 가중치(weight)로서, 아래의 [수학식 4]로 구한다.here,

Is the weight of the i th Gaussian model at time t, and is obtained from Equation 4 below.

&Quot; (4) "

는 학습율로서 이 값이 크면 기존에 구축된 모델이 새로 입력되는 영상에 의해서 빠르게 변경되고 이 값이 작으면 천천히 변경되므로, 사용자가 적절한 값으로 지정한다. 또한,

는 화소값이 가우시안 모델에 정합되면 1, 그렇지 아니한 다른 가우시안 모델들은 0의 값을 갖는다.here,

Is a learning rate. If this value is large, the existing model is quickly changed by the newly input image. If this value is small, it is changed slowly. Also,

Is 1 if the pixel value matches the Gaussian model, and other Gaussian models otherwise have a value of zero.

)과 공분산(

)은 아래의 [수학식 4]로 구한다. And mean (

) And covariance (

) Is obtained from Equation 4 below.

[Equation 5]

는 단위 행렬 I를 이용하여

와 같은 형태를 갖는다. 그리고, K개의 가우시안 분포들은

의 값들을 통하여 정렬되며, 첫 번째 분포 B가 배경 모델로 사용된다. 이때의 B는 아래의 [수학식 5]로 결정된다.On the other hand, the equation of [Equation 5] is made under the premise that the RGB pixel values (red, green and blue pixels) have independent variances from each other.

Using the identity matrix I

Has the same form as And the K Gaussian distributions

Sorted by values, the first distribution B is used as the background model. B at this time is determined by Equation 5 below.

&Quot; (6) "

Here, the threshold T (threshold) is the minimum of all weights obtained from the background model.

)도 학습시키면서, 각 가우시안 분포의 평균 및 공분산을 학습하므로, 단일 가우시안 분포로 모델링할 때보다 감시대상 영역의 실제 배경 영상 변화에 유연하게 대처한다.The adaptive Gaussian mixture model described above may determine whether the current pixel is the background or the foreground area pixel due to the appearance of the moving object according to how similar the background pixel corresponding to the current pixel is by a statistical method. To be able. Accordingly, if the adaptive Gaussian mixture model is used as the background image, the object can be detected by adapting to the brightness change well while being less affected by noise, and the influence of the moving object is temporary even when the moving object passes the monitoring area. As a result, the effect remains on the background image but the effect is less. Moreover, after the moving object passes, the effect remaining by the actual background image gradually disappears. In addition, the weights for the K Gaussian distributions (

), While learning the mean and covariance of each Gaussian distribution, it responds more flexibly to changes in the actual background image of the monitored region than when modeling with a single Gaussian distribution.

However, the adaptive Gaussian mixture model increases the effect when the moving object stops in the monitored area for more than a certain time. Moreover, even if the stopped object is moved again and moves out of the monitored area, the adaptive Gaussian mixture model is applied to the pixel where the object stops. Since the remaining influence is gradually reduced, there is a problem in that it takes a long time to update the real background image disappeared. This phenomenon makes it difficult to detect later appearing objects. In particular, in the pedestrian crossing to which the present invention is applied, since the pedestrian waits in the cross-waiting area B until the pedestrian traffic light 5 turns green, the remaining influence of the pixels corresponding to the waiting pedestrian increases. Will have

The adaptive background image modeling step, which is applied to the present invention, is designed to solve the above-mentioned problems, and the background image modeled even when a pedestrian appears and passes through or stops and disappears in a region of interest that is a region photographed by a camera. Adapt quickly to the actual background. This will be described in detail with reference to FIGS. 6 and 7.

Referring to FIG. 6, the situational cognition modeling unit 300 receives a continuous image frame photographed by a camera and obtains a background image using a Gaussian model. For this purpose, the situational lighting type energy saving pedestrian crossing safety lighting device An initialization module 310 for initializing the Gaussian model by calculating an average and a covariance according to the Gaussian distribution for each pixel with a specified number of initial frames in the image obtained by photographing the image; A foreground region extraction module 320 for extracting a foreground region generated by a pedestrian using a Gaussian model as a background image from an input image frame after initializing the Gaussian model; A module 330 for recognizing a situation in which the foreground region moves or stops when the foreground region is extracted; A background update module 350 for initializing the Gaussian model and updating the Gaussian model according to a recognized situation, and extracting the foreground region from the foreground region extraction module 320 using the Gaussian model as a background image; And a history management module 340 for managing the history of the extracted location of the foreground area, the history of temporal movement and stop of the foreground area so as to be utilized when the situation is recognized by the situation recognition module 330. . Here, the foreground area means a local area where a pedestrian is detected on an image frame.

The background update module 350 includes: a global background update module 351 for updating a Gaussian model for the pixels of the entire background image; And a local background update module 352 for updating only the Gaussian model of the pixel corresponding to the extracted foreground region. That is, according to the present invention, the entire background image is updated according to the recognized situation, or only the extracted foreground area is updated.

Next, the adaptive background image modeling step S70 performed by the situational awareness modeling unit 100 will be described with reference to FIG. 8.

First, in a state in which an image captured by the camera is input as a continuous frame (S110), it is determined whether to initialize the Gaussian mixed model (S120). In general, when a background image is modeled using an adaptive Gaussian model, a Gaussian mixture model for each pixel is trained using 15 to 30 frames in the initial input image. That is, when the situational modeling unit 300 configured as shown in FIG. 7 is mounted on the image processing apparatus, the background image to be initially modeled as a Gaussian model is learned, and after the initial learning, the image to receive the learned Gaussian model is input. To upgrade.

In this case, when it is determined to initialize, the initialization module 310 causes the background update model 350 to train the Gaussian mixture model until a predetermined number of frames are received from the initial input image frame (S130). When initialized as described above, the weight, average, and covariance for each pixel of the Gaussian mixture model are calculated according to the actual background image, thereby initializing the Gaussian mixture model for each pixel.

After learning the Gaussian mixture model with the initial image frame, the foreground region extraction module 320 extracts the foreground region by the object from the input image frame based on the Gaussian mixture model learned as the background image (S140). The foreground area extraction process applied to the embodiment of the present invention is as follows. The foreground area appearing in the image by the object appearing in the area photographed by the camera is extracted based on the Gaussian mixture model, and the shadowed area is removed because the extracted foreground area includes the shadow. After blocking each pixel of the foreground region from which the shadow is removed according to the association, the portion determined to be noise (noise) is removed, the merge between adjacent blobs and the blob by motion are performed. An object of interest (OOI) is obtained by extracting a minimum enclosing rectangle (MER) that includes the boundary of. As described above, the process of acquiring the foreground area based on the background image is not limited to the method described in the embodiment of the present invention.

After performing the process of extracting the image of the foreground area as described above (S140), the situation awareness module 330 recognizes whether the foreground area is extracted and whether the extracted foreground area moves or stops (S150). In operation S160, the global background update module 351 and the local background update module 352 of the background update module 350 are selectively operated according to the recognized situation.

Here, the situation recognition module 330 transmits and manages the temporal and spatial history of the extracted foreground area to the history management module 340, and is continuously input based on the history of the foreground area managed by the history management module 340. We can recognize temporal and spatial changes of the foreground area in the frame. That is, the situation awareness module 130 checks whether the foreground area OOI is extracted (S151), and when the foreground area OOI is not extracted, the Gaussian mixture model for all pixels as the currently input image frame. When the foreground area OOI is extracted (S161) and the foreground area OOI is extracted, the history according to the temporal and spatial change of the foreground area OOI is managed (S152). On the other hand, the step (S151) of checking whether the foreground area is extracted, even if the foreground area is extracted, it is preferable to determine whether the extracted foreground area is of interest and adopt it as the foreground area only if it is of interest. That is, according to the present invention, since the pedestrian is detected using a camera, if the size of the extracted foreground area is within a predetermined range, it is determined as the foreground area for the pedestrian.

Next, the situation awareness module 330 checks whether the foreground area OOI is stopped based on the history management (S153). Here, when a plurality of foreground areas OOI are detected, each of the foreground areas OOI is checked to be stopped and separated and processed separately in the following steps. Through the history management of the foreground area, it is possible to determine the position and movement of the pedestrian and select a lamp to be turned on by the full lighting in the lamp 30 according to the determination result.

At this time, the situation awareness module 330 starts counting the stopped time for the foreground area (OOI) determined to be stopped in the step S153 of checking whether the foreground area is stopped, so that the stopped time is a preset time. It is checked whether or not the stop is abnormal (S154).

)으로 하는 것을 의미한다. 아울러, 초기화할 경우에 가우시안 모델의 공분산은 초기화 전에 사용하였던 공분산(

)을 그대로 사용할 수 있으며, 공분산(

)을 그대로 사용하는 것은 과거 이력에 의해서 공분산(

)이 결정되는 것인데 현재 화소값으로 평균값(

)을 중심으로 공분산을 다시 학습 또는 업데이트하기엔 시간이 경과하기 때문이다. 여기서, 정지한 전경영역의 화소만 현재 촬영된 전경영역의 화소값으로 초기화하므로, 초기화한 전경영역 위치의 화소에 대해 이전에 업데이트하였던 가우시안 분포에 대한 정보는 삭제되므로, 이후에, 정지한 전경영역이 가우시안 모델의 배경영상으로 된다. 이에 따라, 초기화 후에는 정지한 전경영역을 제외하고 새로 진입하는 물체(보행자)만 전경영역으로 추출되며, 그만큼 물체에 따른 전경영역의 추출 시간을 단축할 수 있다. 한편, 정지한 전경영역이 가우시안 모델의 배경영상으로 된 상태이므로 상기한 전경영역 추출단계(S140)에서 추출되지 않지만, 정지한 전경영역에 대한 이력이 이력관리모듈(340)에 의해서 관리되므로 정지한 전경영역에 대해서는 상기한 전경영역 추출단계(S140)에서 감지되는 것이 아니라 전경영역(OOI)의 이력관리 단계(S152)에서 감지된다.In addition, the situation awareness module 330 is a pixel of the foreground region from which the Gaussian model of the pixels of the foreground region OI which have been stopped is currently extracted when the foreground region OOI enters the photographing area and stops for more than a preset time. Initialize to a value (S163). In this case, the initialization is performed by selecting only pixels belonging to the stationary foreground area among all pixels, and locally updating the background. The pixel value of the currently extracted foreground area is averaged as a parameter of the Gaussian model (

Means). In addition, when initializing, the covariance of the Gaussian model is used.

) Can be used as is, and covariance (

) As it is, the covariance (

) Is determined, and the average value (

This is because it takes time to relearn or update the covariance. Here, since only the pixels of the stopped foreground region are initialized to the pixel values of the currently photographed foreground region, the information on the Gaussian distribution previously updated for the pixel of the initialized foreground region position is deleted. It becomes the background image of this Gaussian model. Accordingly, after initialization, only a newly entering object (pedestrian) is extracted to the foreground area except for the stopped foreground area, and the extraction time of the foreground area according to the object can be shortened by that amount. On the other hand, because the stopped foreground area is a background image of the Gaussian model, it is not extracted in the foreground area extraction step (S140), but because the history of the stopped foreground area is managed by the history management module 340, The foreground area is not detected in the foreground area extraction step (S140) but is detected in the history management step (S152) of the foreground area (OOI).

In addition, if the situation awareness module 330 stops the foreground area OOI but does not reach the preset time, the process proceeds to step S110 of receiving an input of the next frame, and the stopped foreground area OOI is the next frame. If you continue to stop at, increase the time counter. Accordingly, when the global area OOI stops within a preset time, the next frame is processed without updating the Gaussian model, thereby eliminating the effect of the afterimage caused by the foreground area OOI.

The situation awareness module 330 checks whether the foreground area OOI is stopped or not in the step S153, with respect to the foreground area OOI which is found to be moving without stopping, and the foreground area OOI is the photographing area of the camera. In step S155, it is checked whether the vehicle stops after the preset time or moves.

Then, the situation awareness module 130 globally updates the Gaussian mixture model for all pixels with the currently input image frame, if the foreground area OI has not stopped for more than a preset time (S161). The Gaussian mixture model for the extracted foreground region OOI pixel is also updated, but the influence of the current value of the moving foreground region OOI on the Gaussian mixture model is small. In this case, since it is determined that the foreground area moving after stopping within a preset time also moves, the Gaussian mixture model is updated globally. In addition, if the user enters the shooting area and appears as the foreground area (OOI) in the image frame, it is a new foreground area (OOI), so it is not stopped at the time of appearance and does not stop for a preset time, so the Gaussian mixture model is updated globally. Although not shown in FIG. 7, if a new foreground area is registered in the history management of the foreground area, the process may be performed to move to the global background update step (S161).

On the other hand, when the situation awareness model 330 confirms that the foreground area OOI has been stopped for more than a predetermined time and started to move, the current foreground area OOI leaves and the foreground area OOI previously stopped. Since the pixels of the position become the background image, the pixels of the position where the foreground area OI was previously stopped are initialized with the current frame pixel values (S162). In this case, the initialization is to locally initialize the Gaussian model to the current frame with respect to the position where the stopped foreground area (OOI) leaves, which means that the pixel value of the current frame is the average value of the Gaussian model. Here too, the previously used covariance value can be used.

In addition, when updating the Gaussian mixture model corresponding to the pedestrian crossing area A, the situation awareness module 330 updates only the area illuminated with full lighting during the time of full lighting. This is because the area illuminated by the tail light is low illumination and no pedestrian to detect pedestrians, and when it is used as a background image, when there is a pedestrian, the image of the illuminated area is displayed only when it is brightly lit with full lighting. To update. In addition, this is because the viewpoint used as the actual background image is the viewpoint illuminated brightly with full lighting. In order to operate in such an update mode, the Gaussian mixture model is updated only for pixels having a predetermined brightness or more among the partial images of the pedestrian crossing area A in which the camera is photographed, or alternatively, The position information may be previously stored in a memory (not shown), and a method of selecting a pixel to be modeled as a Gaussian mixture model based on the position information corresponding to a lamp that is fully lit may be adopted.

The present invention defines a crosswalk area (A) and a cross-waiting area (B) which are areas of interest in the area photographed by the camera, and in order to detect only pedestrians within the limited area, the limited area is previously designated on the image frame. In addition, the background image may be modeled using a Gaussian model for each pixel only in a specified region among all regions of the image frame. This is because there is a case where it is not necessary to generate a background image for an area outside of the interest, thereby improving the processing speed. In this case, the initialization background update (S130) and the global background update (S161) are performed only for the designated area, and in the foreground area extraction process (S140), only the foreground area corresponding to the pedestrian is extracted only within the designated area.

As described above, in the adaptive background image modeling step according to an embodiment of the present invention, the background image is modeled by using an adaptive Gaussian Mixture Model, by detecting a change in the situation according to the pedestrian movement type. In response to changes in the situation, the Gaussian model is updated quickly.If no pedestrians are detected and pedestrians are detected, the background image is updated globally. In the case of stopping after moving for more than a time, only the region of the stopped position updates the background image locally. Accordingly, the present invention can quickly remove the influence of the pixel value of the object remaining in the Gaussian model as the afterimage of the pedestrian, thereby accurately modeling the background image with the pixel value corresponding to the actual background. Pedestrian detection errors that can occur can be reduced. In addition, since the background image is updated locally, the time required for obtaining the background image may be shortened.

9 and 10 are diagrams for explaining a situation-adaptive lighting energy-saving crosswalk safety lighting device according to a second embodiment of the present invention, Figure 9 is a block diagram, Figure 10 is installed in a crosswalk It is an illustration.

9 and 10, in the second embodiment of the present invention, the transverse situation sensor 10 is configured as an operation switch 12. The operation switch 12 is a switch which is installed on the support and is provided to push the pedestrian by hand to cross the crosswalk, the configuration of the camera 20, the lamp 30 and the control device 40 is the present invention described above Since it is configured in the same manner as in the first embodiment, the description thereof will be simplified. That is, the crosswalk safety lighting device according to the second embodiment of the present invention detects the pressing signal of the operation switch 12 as a crossing permission signal, and when the pressing operation of the operation switch 12 is made, the crossing crossing time is set in advance. In order to induce the pedestrian crossing in line with each other, the focused spot is moved in the direction crossing the pedestrian crossing starting from the area close to the sidewalk.

The second embodiment of the present invention is configured to be installed in a crosswalk where no pedestrian traffic light 5 is installed.

The control switch installed in the conventional pedestrian crossing is configured to traverse after operating the operation switch only when the pedestrian wants to cross, and is linked to the pedestrian traffic light so that the pedestrian traffic light for the preset crossing allowance time at the moment the operation switch is operated. Lights green. On the contrary, since the crosswalk safety lighting apparatus according to the second embodiment of the present invention can inform the crossing allowable remaining time to the focused area, the crossing allowable remaining time can be notified even if a separate pedestrian traffic light is not installed.

Although illustrated and described in the specific embodiments to illustrate the technical spirit of the present invention, the present invention is not limited to the same configuration and operation as the specific embodiment as described above, within the limits that various modifications do not depart from the scope of the invention It can be carried out in. Therefore, such modifications should also be regarded as belonging to the scope of the present invention, and the scope of the present invention should be determined by the claims below.

1: Road 2: India 3: Road boundary stone
4: pedestrian crossing 5: pedestrian traffic light
A: crosswalk area B: crosswalk area
10: crossing situation sensor 20: camera 30: lighting
40: controller
100: crossing situation judgment 200: day and night sensing machine
300: situational modeling unit 310: initialization module
320: foreground area extraction module 330: context awareness module
340: history management module 350: background update module
351: global background update module 352: local background update module
400: lighting control unit 500: data storage unit

Claims (7)

delete delete delete delete delete In the lighting device installed on the crosswalk,
Crossing situation sensing device 10 for detecting whether or not to allow the pedestrian crossing;
Illumination 30 to uniformly illuminate the crosswalk area (A) by varying the irradiation direction of the plurality of lamps 31, and to control the lighting of each of the lamps 31 individually;
When the crossing permit signal is received from the crossing situation sensor 10, the local lighting position on the crosswalk area A is moved toward the crossing direction of the pedestrian, but at a predetermined speed according to a preset crossing allowance time. A control device (40) for controlling the lamp (30) to go and illuminate;
Consists of including
The control device 40,
Equipped with a day and night sensing device 200 for detecting that the night,
When the day and night sensing device 200 detects the night, the lamps of the lamp 30 are turned on in the tail light lighting state, and the lamps corresponding to the lighting positions to move at the green light of the pedestrian traffic light 5 are completely turned on. Lights up in a state,
The cross situation sensing device 10,
Installed in the pedestrian traffic light (5) consists of a photo sensor for detecting the green lighting, or the operation switch that the pedestrian can press,
The lamp 30,
It is also configured to illuminate a crosswalk zone (B) on sidewalks in conjunction with the crosswalk,
The control device 40,
The lamp 30 is controlled to keep the cross-waiting area B in a fully lit state at night,
Further comprising a camera 20 for photographing the crosswalk area (A) and the cross-waiting area (B),
The control device 40,
It is provided with a modeling unit 300 for detecting a pedestrian in the crosswalk area (A) and the cross-waiting area (B) as the image taken by the camera 20,
The situation recognition modeling unit 300 of the control device 40,
Initialized adaptive Gaussian mixture model for each background image is calculated by calculating the average and covariance according to the Gaussian distribution of each pixel in the initial continuous frame of the image obtained when the situational lighting energy saving pedestrian crossing safety lighting device is installed. An initialization module 310;
A foreground region extraction module 320 for extracting a foreground region by a pedestrian from an image frame after initialization based on the pixel-specific adaptive Gaussian mixture model;
A history management module 340 for managing the location and time history of the extracted foreground area;
If the foreground area is not extracted and the extracted foreground area is moving, the Gaussian model for each pixel of the background is updated, and if the extracted foreground area stops for more than a preset time and then moves, A background update module 350 for initializing the mean value of the adaptive Gaussian mixture model to the current pixel value;
And,
The control device 40 detects the position and movement of the pedestrian based on the history of the foreground area managed by the history management module 140, and controls the lamp 30 to illuminate the pedestrian's surroundings with full lighting. Situational adaptive lighting energy-saving crosswalk safety lighting device characterized in that. The method of claim 6,
The situation recognition modeling unit 300 of the control device 40,
When modeling the image of the crosswalk area (A) as an adaptive Gaussian mixture model, only the area illuminated by full lighting is updated according to the time of full lighting. Device.

Images