KR101840761B1 - sensing device and sensing method using image recognition - Google Patents

Abstract

The present invention relates to an image recognition sensing device for determining a view state in the air, and a method thereof. The image recognition sensing device according to an embodiment of the present invention includes a light emitting part for emitting light while sequentially flickering at a plurality of positions; a first camera for photographing the light emitting part at a distance from the light emitting part; and a control part which receives image data photographed by the first camera, analyzes at least two among the luminous intensity of each of the positions, a luminous position shift by the sequential flickering, and a luminous object size for each of the positions, from the received image data, and determines the view state in the air.

Description

[0001] The present invention relates to a sensing device and a sensing method using the same,

[0001] The present invention relates to an image recognition apparatus and a method thereof capable of determining an in-the-view state.

Fog, etc., the visibility of the object and the clock are remarkably reduced, thereby increasing the incidence of traffic accidents such as vehicle collision and lane departure. This leads to serious vehicle damage and personal injury. In particular, it is difficult to predict the occurrence of an atmospheric phenomenon such as a fog, and it is difficult to predict the occurrence time and occurrence of the fog.

The road weather information system (RWIS) uses weather sensors such as rainfall / snowfall detectors, barometers, underground thermometers, road surface detectors, freezing sensors, temperature / humidity sensors, wind direction anemometers, solar radiation meters, Monitor the status.

In the conventional mist detector, a visibility sensor for measuring the forward scattering of light using an infrared laser is generally used.

However, such a corrective sensor is relatively expensive and has a disadvantage in that a clear error occurs when the measurement principle is raining.

BACKGROUND OF THE INVENTION Field of the Invention [0001] The present invention relates to an image sensing apparatus and method for determining a visual state of an atmospheric phenomenon such as a fog using image data photographed through a camera and using the image sensing apparatus for traffic safety.

According to an embodiment of the present invention, there is provided an image sensing apparatus including: a light emitting unit emitting light while sequentially flashing at a plurality of positions; A first camera for photographing the light emitting portion at a distance from the light emitting portion; At least two or more of the luminous intensity of each of the plurality of positions, the luminous position shift by the sequential blinking, and the luminous object size of each of the plurality of positions are analyzed from the received image data, And a control unit for determining the visual field state.

The control unit compares the light emission luminances for each of a plurality of positions, the light emission movement by the sequential blinking, and the size of the light emission image for each of the plurality of positions, with a reference value according to a predetermined in- , It is possible to stepwise determine the state of the in-the-view field. For example, each of the reference values according to the preset in-the-view visual field state is set to have a level-specific reference value based on the distance between the first camera and the light-emitting portion, the luminance of the light-emitting portion, and the light- And has a level score corresponding to a level-specific reference value according to the visual field state.

Here, the stepwise determination of the in-the-view field state can be determined as the attention step, the alarm step, and the emergency step according to the summed score.

The light emitting portion includes a plurality of LED portions of different colors arranged at regular intervals. In one example, the plurality of LED portions include a red LED portion, a green LED portion, and a blue LED portion.

The image recognition sensing apparatus according to the present invention may further include a second light emitting unit that is photographed at a second predetermined distance from the first camera as a position different from the light emitting unit and sequentially emits light at a plurality of positions. In this case, the first camera alternately captures the light emitting unit and the second light emitting unit at predetermined time intervals.

The image recognition sensing apparatus according to the present invention may further include an induction lamp disposed at predetermined intervals along the roadway. In this case, the controller can control the operation of the guide light step by step in accordance with the visual field state determined step by step.

The image recognition apparatus according to the present invention may further include a display board disposed at a predetermined position of the road map. In this case, the control unit may control to display at least one of the step-by-step view state information and the vehicle safe speed information according to the view state determined stepwise.

In addition, the image recognition sensing apparatus according to the present invention may further include a second camera for photographing a predetermined section of the road map. In this case, the control unit receives the image data photographed by the second camera, performs image analysis of the vehicle speed within a predetermined interval from the received image data, determines the traffic condition according to the vehicle speed step by step, As shown in Fig.

For example, the control unit can transmit the determined waiting state information to the control server managing the road traffic situation in real time.

According to another embodiment of the present invention, there is provided a method of detecting an image, comprising: emitting light while sequentially emitting light at a plurality of positions; Photographing the light emitting portion at a distance from the light emitting portion; And at least two of the luminous intensity of each of the plurality of positions, the luminous movement of the plurality of positions, the luminous movement of the luminous flux by the blinking and the luminous image sizes of the plurality of positions are analyzed from the received image data, .

According to the present invention, by analyzing a plurality of elements by using image data of a light emitting unit photographed through a camera, it is possible to accurately determine a visual state of an atmospheric phenomenon such as a fog.

Other effects according to the present invention will be further described with reference to the following embodiments.

1 is a block diagram illustrating a configuration of an image recognition sensing apparatus according to an embodiment of the present invention.
FIG. 2 is an exemplary view of the light emitting portion of FIG. 1. FIG.
3 is an exemplary view for explaining shooting of a light emitting portion using a camera.
4 (a) to 4 (b) are diagrams for explaining an image analysis process of photographed image data.
Fig. 5 (a) and Fig. 5 (b) are illustrations for explaining an induction lamp.
6 is a flowchart illustrating an image recognition detection method according to an embodiment of the present invention.
7 is a flowchart illustrating an image recognition detection method according to another embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the following description of the present invention, detailed description of known related arts will be omitted when it is determined that the gist of the present invention may be blurred.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Referring to the accompanying drawings, the same or corresponding components are denoted by the same reference numerals, .

1 is a block diagram illustrating a configuration of an image recognition sensing apparatus according to an embodiment of the present invention.

As shown in FIG. 1, the image recognition sensing apparatus according to the present invention includes at least a light emitting unit 10, a first camera 20, and a control unit 30.

First, the light emitting portion 10 is configured to emit light while sequentially flashing at a plurality of positions. For example, the light emitting portion 10 may include a plurality of LED portions 11 of different colors arranged at regular intervals as shown in FIG.

The plurality of LED portions 11 may include a red LED portion R, a green LED portion G, and a blue LED portion B. Here, the plurality of LED units 10 may not necessarily have different colors, but may be formed of only one color. In addition, the colors of the plurality of LED portions 11 are not limited to red (R), green (G), and blue (B), and may be set to various colors.

The light emitting unit 10 may be controlled by the control unit 30, and the plurality of LED units 11 may sequentially emit light during the light emitting operation. For example, during the light emitting operation, the red LED portion R, the green LED portion G and the blue LED portion B are sequentially and repeatedly turned on and off sequentially at intervals of a predetermined time (for example, one second).

For example, a plurality of light emitting units 10 may be provided at predetermined intervals. For example, as shown in FIG. 3, the image recognition sensing apparatus according to the present invention may be used to capture an image at a second predetermined distance from the first camera 20 as a position different from that of the light emitting unit 10, And a second light emitting unit 40 that emits light while sequentially flashing. In this case, the first camera 20 alternately captures the light emitting unit 10 and the second light emitting unit 40 at predetermined time intervals (for example, at intervals of 30 seconds). At this time, the camera 20 is adjusted in focus in a manner set to acquire image data of a similar frame from the light emitting unit 10 and the second light emitting unit 40 at the time of photographing alternately, or may be rotated Can be driven.

The light emitting portion 10 may be installed around a driveway in an area where mist or the like frequently appears or disappears. However, it is preferable that the light emitting portion 10 is installed so as not to face the vehicle passing the roadway so that the vehicle passing the roadway does not mistakenly recognize the light emitting portion 10 as a traffic light.

Next, the first camera 20 is arranged at a position a certain distance from the light emitting portion 10, and photographs the light emitting portion 10.

The distance between the first camera 20 and the light emitting unit 10 may be set in consideration of the performance of the camera, the light emitting size of the light emitting unit 10, and the light emission luminance. For example, And may be installed at intervals of 5 to 10 meters from the main body 20. 3, the second light emitting unit 40 may be installed at an interval of 15 to 20 meters from the first camera 20. In this case, The first light emitting unit 10 and the second light emitting unit 40 are not necessarily arranged in a line from the first camera 20 as shown in FIG. 3, but may be installed at various positions at predetermined intervals.

The first camera 20 is driven by the control unit 30 and photographs the light emitting unit 10 for a predetermined time and transmits the photographed image data to the control unit 30. [ The first camera 20 may photograph the light emitting unit 10 to capture the image data or may photograph the image at the consecutive blinking time intervals of the light emitting unit 10. [

Then, the control unit 30 receives the image data photographed by the first camera 20, analyzes the received image data, and determines the in-the-view state.

Specifically, the control unit 30 analyzes at least two of the luminous intensity of each of the plurality of positions, the movement of the luminous position by the progressive blinking, and the size of the luminous object by the plurality of positions from the received image data.

For example, FIGS. 4 (a) to 4 (b) are illustrations for explaining an image analysis process of photographed image data. 4A is an image data image taken when all the plurality of LED portions 11 of the light emitting portion 10 are turned off (that is, when the LED portions 11 are turned off). 4B is an image data image photographed when only the red LED portion R among the plurality of LED portions 11 of the light emitting portion 10 is turned on (i.e., in the ON state). 4C is an image data image taken when only the green LED portion G among the plurality of LED portions 11 of the light emitting portion 10 is turned on (i.e., when the LED portion G is ON). 4C is an image data image taken when only the blue LED portion B among the plurality of LED portions 11 of the light emitting portion 10 is turned on (that is, when the blue LED portion B is ON). 4 (b) to 4 (d) is repeated for the following image data image.

At this time, the controller 30 recognizes the LED portion of the ON state in the image data image. In FIG. 4, a rectangle image object recognition box is displayed for the sake of understanding, and the recognized portion is displayed.

Accordingly, the control unit 30 performs image analysis using each LED portion of the recognized ON state. For example, the control unit 30 compares the light emission luminances of a plurality of positions, the light emission movement by the sequential blinking, and the light emission image size of each of the plurality of positions with a reference value according to a predetermined in-the-view visual field state, It is possible to calculate the visual field state in the step by step by summing the satisfactory scores.

For example, the light emission luminance (i.e., the average luminance of the object) from the image data is determined as the average brightness of the recognized LED portion, and the movement of the light emission position from the image data (i.e., movement detection) The size of the light emitting entity (i.e., the size of the sensing object) can be determined from the image data as the pixel size of the recognition area. To this end, the control unit 30 stores in advance a program for performing each judgment function. Alternatively, the control unit 30 may further include a module device in which a program for performing the respective determination functions is stored, and may be controlled according to the set method. Alternatively, as another example, the first camera 20 may include the control unit 30 so that the image capturing and the image analysis may be collectively processed from the first camera 20. [

Here, each of the reference values according to the preset in-the-view visual field state is determined by the distance between the first camera 20 and the light emitting portion 10, the luminance of the light emitting portion 10, and the light emission size in accordance with the occurrence of rain, And may have a score for each level corresponding to the reference value for each level according to the visual field state.

For example, Table 1 is a table showing reference values according to the visual field conditions.

Object average luminance Move Detection Sense object size LED part burglar score LED part Number of times score LED part Recognition size score Red
LED part Level 1 One Red
LED part Level 1 One Red
LED part Level 1 One Level 2 2 Level 2 2 Level 2 2 Level 3 3 Level 3 3 Level 3 3 Green
LED part Level 1 One Green
LED part Level 1 One Green
LED part Level 1 One Level 2 2 Level 2 2 Level 2 2 Level 3 3 Level 3 3 Level 3 3 blue
LED part Level 1 One blue
LED part Level 1 One blue
LED part Level 1 One Level 2 2 Level 2 2 Level 2 2 Level 3 3 Level 3 3 Level 3 3

However, in Table 1, the reference value for each level is classified into three levels, but it is not limited to this, and various levels can be set.

Here, the stepwise determination of the in-the-view field state can be determined as the attention step, the alarm step, and the emergency step according to the summed score. For example, assuming that image data is obtained by sequentially flashing the light emitting unit 10 at intervals of 1 second for 30 seconds based on Table 1, since three reference values are classified into three levels, 27 Acquisition information can be obtained, and the total score for each level can be obtained from 0 to 54 points.

Thus, the visibility state at the time of occurrence of fog can be determined as shown in Table 2.

Visibility level Total points by level Judge status One 20 or less  Fog 2 21 ~ 40  Fog alarm 3 40 or more  Fog emergency

In Table 2, the reference value for each level is determined to be a high level (i.e., level 3) as the visual state is worse, and the lower the reference level for each level is judged as the lower level Level 1 to level 3), the score for each level is determined to be 0. However, the reference value for each level can be set in various ways. Also, although the state judgment is determined to be the caution phase, the alarm phase, and the emergency phase, it can be set in various stages.

Accordingly, the image recognition sensing apparatus according to the present invention performs image analysis of a plurality of elements using the image data of the light emitting unit 10 photographed through the first camera 20, (For example, the degree of fog generation or the correcting distance) can be accurately determined. In addition, the present invention can determine the visual field state more effectively because the visual field state is determined in a similar manner to the human visual sense by using the image data image photographed by the camera.

1, the image recognition sensing apparatus according to the present invention may further include at least one of the guide light 50, the electric sign board 60, and the second camera 70. [

First, the guide light 50 is a lighting device arranged at regular intervals along the roadway. For example, as shown in FIG. 5 (a), the guide lights 50A may be provided at regular intervals on the ground surface of the vehicle, and may be ON / OFF controlled by the control unit 30. FIG. Alternatively, as shown in FIG. 5 (b), the guide lights 50B may be provided at predetermined intervals on the guide rail 90 of the driveway, and may be ON / OFF controlled by the control unit 30.

The guide lights 50 may be installed at 8M intervals in the case of the curve section and at 16M intervals in the case of the straight section, for example.

Further, the guide light 50 may be an LED or the like by way of example, and the guide light 50 may be made of a three-color guide light including, for example, a red guide light, an orange guide light, and a yellow guide light.

In this manner, when the guide light 50 is provided, the control unit 30 can control the guide light 50 in a stepwise manner according to the view state determined step by step. For example, the control unit 30 controls the color conversion, the blinking speed conversion, and the like of the guide light 50 according to the determined visual field state to induce the driver of the vehicle passing through the road to recognize and recognize the current visual field state At the same time, to recognize the traveling direction of the lane. For example, if the visual field is foggy, the yellow indicator lights will flash at set intervals. If the visual field is foggy, the orange indicator lights will flash at set intervals. If the visual field is in a foggy emergency, can do.

Then, the electric sign board 60 is disposed at a predetermined position around the roadway, and can display at least one of the visual state information and the vehicle safety speed information in a stepwise manner according to the visual field state. For example, the electric sign board 60 can arrange a plurality of LEDs on a plane and display a figure (or a symbol) or a letter or the like under the control of the controller 30. [

Accordingly, when the electric sign board 60 is provided, the control unit 30 controls to display at least one of the visual field state information and the vehicle safety speed information according to the visual field state determined step by step, So that the driver of the vehicle can recognize the current visual state more accurately.

Accordingly, the image recognition sensing apparatus according to the present invention controls the guide light 50 and / or the electric signboard 60 according to the determined visual field state, thereby allowing the driver of the vehicle passing through the area to accurately determine the current standby state .

Then, the second camera 70 is a photographing device that photographs a certain section of the road map. For example, the second camera 70 photographs a predetermined section of the road map on the road map or photographs a predetermined section of the road map on the road map, and transmits the photographed image data to the control section 30.

In this case, the control unit 30 receives the image data photographed by the second camera 70, analyzes the image of the vehicle speed within a certain interval (that is, the flow state of the vehicle) from the received image data, The traffic condition can be judged step by step.

In addition, the control unit 30 can control to display the determined step-by-step traffic state information on the electric sign board 70. [ Alternatively, the control unit 30 may control the guide light 50 to display the traffic state. For example, in the case of a three-color induction lamp, in order to guide a vehicle approaching from the rear to a traffic state in front of the road, if the flow of the vehicle is slower than the set speed (that is, For example, the red light and the orange light can be alternately flickered.

Also, when it is determined that a specific event (for example, a traffic collision accident) has occurred according to the determined traffic condition, the control unit 30 may transmit the corresponding event information to the terminal of the designated contact person or to the established control server.

Accordingly, the image recognition sensing apparatus according to the present invention can recognize a traffic situation in addition to a vision state, thereby recognizing a situation in front of a road where a mist occurs, thereby preventing an accident in advance.

Referring again to FIG. 1, the image recognition sensing apparatus according to the present invention may further include a control server 80. The control server 80 is a server for managing road traffic conditions, for example, a system server of Korea Highway Corporation.

In this case, the control unit 30 may transmit the determined waiting state information to the control server 80 in real time, and use the information in the control server 80 in various ways using the information. For example, the control server 80 may broadcast the received visibility state information to a radio broadcast that guides traffic conditions in the area.

The image recognition sensing apparatus according to the present invention can be understood as an image recognition sensing system.

6 and 7, a method of detecting image recognition according to the present invention will be described. FIG. 6 is a flowchart illustrating an image recognition detection method according to an embodiment of the present invention, and FIG. 7 is a flowchart illustrating an image recognition detection method according to another embodiment of the present invention.

6, a plurality of LED units 11 of the light emitting unit 10 are sequentially turned on and off at a plurality of positions under the control of the control unit 30, (S10). Next, the first camera 20 captures the light emitting unit 10 that flashes at a predetermined distance from the light emitting unit 10 (S20). At this time, the first camera 20 transmits the photographed image data to the controller 30. Then, the control unit 30 receives the photographed image data and determines the in-the-view state by analyzing the image data (S30). At this time, the controller 30 performs image analysis of at least two elements among the light emission luminances per a plurality of positions, the light emission movement by the blinking and the light emission image sizes by the plurality of positions from the received image data, . Then, the control unit 30 can control at least one of the guide light 50 and the electric sign board 60 (S40) so as to be displayed in the determined visual field state.

Alternatively, as shown in FIG. 7, in the image recognition sensing method according to the present invention, the light emitting unit 10 performs the RGB light control operation by the control unit 30 (S110). That is, the plurality of LED units 11 are sequentially controlled to be flickered. Then, the first camera 20 and the control unit 30 are used to perform image observation and image recognition (S120). That is, the first camera 20 performs image observation by photographing the light emitting unit 10, and the control unit 30 recognizes the light emitting object within the acquired image data. Then, the control unit 30 analyzes the image data, stores the image data, and transmits the image data to the general situation room as the control server 80 (S130). Then, the traffic condition for each road is remotely controlled and operated using data analyzed in the general situation room (S140). For example, the visibility state information determined according to the image analyzed data may be displayed in a predetermined format on the electric signboard 60 for each road, or the three-color guide light may be remotely controlled (S150). If the visual field is in the normal state according to the image analyzed in the general situation room, control of the guide light 50 and the electric sign board 60 may be terminated in a normal standby state. That is, in FIG. 7, the guide light 50 and the electric sign board 60 can be controlled by the control server 80.

According to the present invention, by analyzing a plurality of elements using image data of a light emitting unit photographed through a camera, it is possible to accurately determine a visual state of an atmospheric phenomenon such as a mist, It is possible to effectively control the electric sign board.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

10: light emitting unit 20: first camera
30: control unit 40: second light emitting unit
50: Guide light 60: Display board
70: second camera 80: control server

Claims (12)

A light emitting portion which emits light while repeatedly and repeatedly flashing at a plurality of positions;
A first camera for photographing the light emitting unit at a distance from the light emitting unit;
And a controller for receiving image data photographed by the first camera, performing image analysis of the movement of the light emission position by sequentially flashing from the received image data, and the size of the light emitting object for each of a plurality of positions according to the movement of the light emission position, And a controller for determining a state of the image. The method according to claim 1,
Wherein the control unit compares the size of the light emitting entity for each of a plurality of positions according to the movement of the light emitting position by the sequential blinking and the size of the light emitting unit for each position according to the movement of the light emitting position, And a step of judging the state of the atmospheric state step by step. The method of claim 2,
Each of the reference values according to the predetermined in-waiting-view state is set to have a level-specific reference value based on the distance between the first camera and the light-emitting portion, the luminance of the light-emitting portion, and the light emission size according to the occurrence of rain, the occurrence of fog, or the occurrence of smog, And a score for each level corresponding to the level-specific reference value according to the level. The method of claim 3,
Wherein the step of judging the atmospheric visibility state is judged to be a caution step, an alarm step and an emergency step according to the summed score. The method according to claim 1,
Wherein the light emitting unit includes a plurality of LEDs of different colors arranged at regular intervals. The method of claim 5,
Wherein the plurality of LED units include a red LED unit, a green LED unit, and a blue LED unit. The method according to claim 1,
Further comprising a second light emitting unit that is photographed at a second predetermined distance from the first camera as a position different from the light emitting unit and emits light while sequentially flashing at a plurality of positions,
Wherein the first camera alternately captures the light emitting unit and the second light emitting unit at predetermined time intervals. The method of claim 4,
And a guide light disposed at regular intervals along the roadway,
Wherein the control unit controls the operation of the guide light step by step according to a state of view determined stepwise. The method according to claim 1,
Further comprising a display board disposed at a predetermined position of the driveway,
Wherein the control unit controls the display unit to display at least one of visual field status information and vehicle safety speed information in accordance with the visual field condition determined stepwise. The method of claim 9,
And a second camera for photographing a predetermined section of the road map,
The control unit receives the image data photographed by the second camera, performs image analysis of the vehicle speed within the predetermined interval from the received image data, determines the traffic state according to the vehicle speed step by step, And displays it on the electric sign board. The method according to claim 1,
And the control unit transmits the determined waiting state information to the control server managing the road traffic situation in real time. Emitting the light emitting portion repeatedly and sequentially in a plurality of positions;
Capturing a light emitting portion at a distance from the light emitting portion;
The image pickup device receives the photographed image data, performs image analysis of the size of the light emitting object according to a plurality of positions according to movement of the light emitting position by sequentially flashing from the received image data, and movement of the light emitting position, The method comprising the steps of:

Images