KR101993445B1 - Visibility meter of image analysis using artificial intelligence - Google Patents

Abstract

Disclosed is an image analysis visibility meter using artificial intelligence. The image analysis visibility meter using artificial intelligence includes: an image information obtaining part obtaining image information about subjects set by distance; and an information processing part calculating the definition of image information about each of the subjects and determining image information, of which the definition is no less than a preset value, if the image information about the subjects is obtained, and calculating a visible distance in a specific area based on the determined image information. Therefore, a visible distance is quantitatively calculated by analyzing an image, obtained through equipment such as a CCTV camera, based on artificial intelligence technology, and if visibility is deteriorated due to fine or ultrafine dust, information about a visible distance in an area with the deteriorated visibility is delivered to drivers in the area, so they may use more caution when driving to reduce death and injury from accidents and subsequent economic losses due to the deterioration of visibility.

Description

[0001] The present invention relates to an image analysis system using artificial intelligence,

The present invention relates to an image analysis system, and more particularly, to an image analysis system using artificial intelligence.

Visibility is the maximum distance that can identify the terrain or object at a long distance in the horizontal direction during the day, usually the naked eye, and the maximum distance at which the target can be identified at night, assuming a bright state such as daytime .

Visual, optical, and imaging measurements are used to measure these visibilities. The correction distance can serve as an indicator for quantitatively grasping atmospheric pollution, and can prevent the economic loss due to the deterioration of the air conditioner, the road, and the ocean in advance.

In recent years, fine dust or ultrafine dust is caused by serious social problems such that the forecast of fine dust is included in the weather forecast, and when the deterioration of the security due to fine dust or ultrafine dust occurs, the occurrence of fine dust or ultrafine dust Quantitatively, and calculating the visibility distance is becoming an important issue.

Conventionally, in order to calculate such a correction distance, a method has been developed in which an optical measurement is performed directly on an object, or a correction distance is measured using geographical information, a perspective and a nonlinear function. However, due to the development of nonlinear conversion techniques such as deep running , The demand for measuring the corrective distance in the case of the deterioration of the visibility in the more precise manner than the existing ones is gradually increasing, but the development of the effective technology is insufficient.

Accordingly, it is required to find a way to quantitatively calculate a corrective distance in a specific area by using artificial intelligence technology, and to prevent an economic loss when the deterioration of the visibility occurs due to fine dust or ultrafine dust.

Korean Patent Laid-Open No. 10-2006-0101990 (Title: Mist Disaster Prevention System) Korean Patent No. 10-1503213 (entitled " Device for Measuring Distance and Distance Using Geographic Information and Pattern Recognition Technology of Video Image "

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to provide a method and apparatus for analyzing a video image obtained through a device such as a CCTV camera using artificial intelligence technology, And to provide an image analysis system using intelligence.

Another object of the present invention is to provide a directional voice outputting means for informing drivers who travel in a region where a time-worsening deterioration has occurred when a time-worsening deterioration occurs due to fine dust or ultrafine dust or the like, The present invention provides an image analysis system using an artificial intelligence that can prevent human injury and economic loss due to deterioration in munition.

According to an aspect of the present invention, there is provided an image analysis system using artificial intelligence, comprising: an image information acquisition unit for acquiring image information of objects set in a distance; The image processing method according to claim 1, further comprising the steps of: calculating image clarity of each image information when the image information on the objects is acquired; determining image information whose calculated sharpness is equal to or greater than a predetermined value; And an information processing unit for calculating an output value.

The information processing unit may store and store image information of a subject positioned at predetermined distance intervals based on the image information obtaining unit, distance information on the distance between the image information obtaining unit and the subject, The image information acquiring unit sequentially acquires the image information of the first object located at the farthest position with respect to the image information acquiring unit based on the stored image information and the distance information, It is possible to acquire the image information of the objects set by the distance.

When the sharpness of each of the pieces of image information is calculated, the information processing unit calculates the sharpness of the image information of the first subject among the pieces of image information, and determines whether the calculated sharpness is equal to or greater than the predetermined value And if the calculated sharpness is equal to or greater than the preset value, the sharpness calculation procedure of the image information for the other subjects may be interrupted, and the corrective distance may be calculated based on the distance information about the first subject.

If the calculated sharpness is less than the predetermined value, the information processing unit may calculate image information on the second subject, which is closer to the first subject than the first subject, based on the image information obtaining unit, And determining the sharpness of the second subject based on the result of the determination of the sharpness of the image information with respect to the second subject so that the distance to the second subject The correcting distance may be calculated on the basis of the information or the sharpness of the image information of the third subject nearer to the predetermined distance than the second subject may be calculated based on the image information obtaining unit.

In order to calculate the sharpness of each of the image information, the information processing unit may previously store the minutia point setting information for setting some pixels among the plurality of pixels constituting the respective pieces of image information as minutiae The sharpness of the specific image information can be calculated by calculating the number of recognizable feature points from the obtained specific image information based on the feature point setting information for the specific image information.

The image analysis timeline using the artificial intelligence according to the present embodiment is characterized in that the image analysis system using the artificial intelligence includes a directional voice output unit for outputting voice information on the corrective distance in the specific area identified by the information processing unit, And an information output unit.

The first area may be a region where a road is located or a region where a vehicle traveling along the road is located.

The directional voice information output unit includes a radar that emits a radiation wave toward a vehicle traveling along the traveling road and receives a reflected wave reflected by the vehicle so as to travel along the traveling road The vehicle can be detected.

Accordingly, it is possible to quantitatively calculate the correcting distance by analyzing the image obtained through the equipment such as a CCTV camera by using the artificial intelligence technology. When the deterioration of the corrective state occurs due to fine dust or ultrafine dust, Information on the corrective distance of the area can be transmitted to the drivers traveling in the area where the deterioration has occurred through the directional voice output means to improve the alertness and to prevent the loss of life and the economic loss due to the deterioration of the visibility.

FIG. 1 is a diagram illustrating a configuration of an image analysis timeline using artificial intelligence according to an embodiment of the present invention. Referring to FIG.
FIG. 2 is a diagram illustrating image information obtained by distance through an image information obtaining unit according to an embodiment of the present invention.
3 is a diagram illustrating image information for a specific region obtained through the image information obtaining unit according to an embodiment of the present invention.
4 is a view for explaining a process of calculating the sharpness of image information obtained by the information processing unit according to an embodiment of the present invention.
FIG. 5 is a view illustrating a state in which a corrected distance calculated through an information processing unit according to an exemplary embodiment of the present invention is output through a separate screen.
FIG. 6 is a diagram illustrating a process of calculating a correction distance using an image analysis timeline using artificial intelligence according to an embodiment of the present invention. Referring to FIG.

Hereinafter, the present invention will be described in detail with reference to the drawings. The embodiments described below are provided by way of example so that those skilled in the art will be able to fully understand the spirit of the present invention. The present invention is not limited to the embodiments described below and may be embodied in other forms.

FIG. 1 is a view for explaining a configuration of an image analysis timeline using artificial intelligence according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating an image information acquisition unit 100 according to an embodiment of the present invention. FIG. 3 is a diagram illustrating image information for a specific region obtained through the image information obtaining unit 100 according to an embodiment of the present invention. Referring to FIG.

The image analysis timeline using the artificial intelligence according to the present embodiment analyzes the image image using the artificial intelligence technology to quantitatively calculate the corrective distance, and when the deterioration of the corrective state occurs due to fine dust or ultrafine dust, And is provided to transmit information on the corrective distance of the area to the drivers who travel in the area where the deterioration occurs due to the directional voice output means.

To this end, the image analysis timeline using the artificial intelligence may include an image information obtaining unit 100, an information processing unit 200, and a directional audio information output unit 300.

The image information obtaining unit 100 is provided for obtaining image information on the subjects. Specifically, the image information obtaining unit 100 may be implemented as a photographing unit such as a CCTV camera and an IP camera, and may acquire image information by photographing a subject by a distance as shown in FIG.

Here, the object means a subject positioned at a predetermined distance interval by the information processing unit 200 among objects arranged at a predetermined position regardless of time, such as a structure, a facility, and a sign.

The information processing unit 200 calculates the sharpness of each image information when the image information about the subjects is obtained through the image information obtaining unit 100 and determines image information whose calculated sharpness is equal to or larger than a preset value , A sharpness calculation module 210, a corrective distance calculation module 220 and a storage module 230 to calculate a correction distance in a specific area based on the discriminated image information.

Specifically, the sharpness calculation module 210 is connected to one or more image information acquisition units 100, and acquires image information about a subject positioned at a predetermined distance interval based on the connected image information acquisition unit 100 and image information The distance information on the distance between the unit 100 and the subject is matched and stored in the storage module 230 so that the connected image information acquisition unit 100 can acquire image information for each distance according to a predetermined schedule.

3A to 3D, when the image information acquiring unit 100 acquires image information by distance according to a preset schedule, the image information acquiring unit 100 acquires the image information stored in the image information acquiring unit 100 and the distance information The photographing direction can be adjusted to photograph a subject related to the stored image information. 3A to 3D are diagrams showing image information obtained by photographing the same subject at different times.

On the other hand, the acquisition order of each image information located at different distances from the image information of the first object located farthest from the image information acquiring unit 100, And a third subject positioned closer by a predetermined distance than the second subject. Thus, it is possible to acquire image information on the subjects sequentially set in the distance by the second subject.

The sharpness calculation module 210 calculates the sharpness of each image information from the image information of the first object located farthest from the image information obtaining unit 100 according to the order in which each image information is obtained Can be calculated.

When the image information whose calculated sharpness is equal to or greater than a preset value is determined, the correction amount distance calculation module 220 stops the procedure of calculating the sharpness of the image information for the other subjects, The correcting distance can be calculated.

That is, the information processing unit 200 calculates the sharpness from the image information of the first object through the sharpness calculation module 210, and if the calculated sharpness is greater than a predetermined value, If the sharpness of the image information for the first subject is less than the predetermined value, the correcting distance calculating module calculates the correcting distance based on the distance information for the first subject, and stops the procedure for calculating the sharpness of the image information for the other subjects. And calculates the sharpness of the image information for the second subject nearer to the image information obtaining unit 100 than the first subject by a predetermined distance interval through the correction distance calculating module 210, The correcting distance in the specific area can be calculated by discriminating whether the sharpness of the image information for the second subject is equal to or greater than a preset value.

Here, the sharpness calculation module 210 may calculate the correction distance based on the distance information of the second subject or may calculate the correction distance based on the image information obtaining unit 100 as a reference based on the determination result of the sharpness of the image information about the second subject. The sharpness calculation procedure can be repeatedly performed until image information whose sharpness is equal to or greater than a predetermined value is determined by calculating the sharpness of the image information for the third subject nearer to the subject than the predetermined distance interval.

For example, the information processing unit 200 calculates the sharpness of the image information about the first subject 2 km (2,000 m) away from the image information obtaining unit 100, and if the value is less than the predetermined value, The sharpness calculation procedure can be repeatedly performed until the image information of the second subject positioned near 10m is calculated and compared with the preset value until the image information whose sharpness is equal to or greater than a preset value is discriminated , And if the sharpness of the image information for the second subject is equal to or greater than a predetermined value, the correction distance can be calculated to be 1,990m based on the distance information of the second subject.

The storage module 230 may store identification information for one or more image information acquisition units 100, image information for a subject positioned at predetermined distance intervals based on each image information acquisition unit 100, The distance information on the distance between the subject 100 and the subject can be matched and stored. In particular, when the image information obtaining unit 100 is implemented as an IP camera, IP address information of the image information obtaining unit 100 may be stored in the storage module 230 as identification information.

In addition, the storage module 230 stores the angle information about the photographing angle, which is used for efficiently photographing a subject positioned at a predetermined distance interval through the specific image information obtaining unit 100, And stored. Accordingly, when the specific image information obtaining unit 100 arrives at the time of acquiring the image information by distance, the information processing unit 200 can effectively photograph the objects by using the angle information about the objects set for each distance.

The directional voice information output unit 300 is provided for outputting voice information about the corrective distance to the first area set by a predetermined limited directivity angle.

Specifically, when the corrective distance to a specific area is calculated by the information processing unit 200, the directional voice information output unit 300 outputs the directional voice information corresponding to the area where the vehicle is traveling The audio information on the corrected distance can be output to the first area.

Here, the first area set by the predetermined limited directivity angle means an area belonging to the specific area in which the corrective distance is calculated, whereby the area where the driving road is located or the vehicle traveling along the driving road is located It is possible to listen to the voice information about the correcting distance only in the area.

The directional voice information output unit 300 is provided with a radar (not shown), and can sense a vehicle traveling along the traveling road using the radar.

Specifically, the directional audio information output unit 300 outputs a radiation wave toward a vehicle traveling along a traveling road using a radar, and when a reflected wave reflected by the vehicle is received, a vehicle traveling along the traveling road And only when the vehicle is detected along the traveling road, the voice information on the corrected distance can be outputted.

Through this, it is possible to inform the driver who is traveling in the area where the deterioration is caused by the information about the visibility of the area, thereby improving the alertness and preventing the loss of life and economic loss due to the deterioration of visibility.

FIG. 4 is a diagram illustrating a process of calculating the sharpness of image information obtained by the information processing unit 200 according to an exemplary embodiment of the present invention.

In order to calculate the sharpness of each piece of image information, the information processing unit 200 according to this embodiment arbitrarily designates some pixels among a plurality of pixels constituting each piece of image information before selecting sharpness, And stores the feature point setting information for some pixels set as the feature points in advance.

Here, some of the pixels set as the minutiae can be set as one minutiae by allowing a plurality of pixels to designate one group. Specifically, FIG. 4A shows a case where some pixels among a plurality of pixels are designated as three minutiae, And FIG. 4B is a diagram showing an example in which some pixels among a plurality of pixels are designated as four groups and set as minutiae.

In other words, the information processing unit 200 may designate one group of pixels, which are difficult to identify due to similarities in color and shape among a plurality of pixels constituting each image information, to be set as one feature point. By storing the feature point setting information for the specific image information in advance in the storage module 230 so that the number of recognizable feature points is calculated from the specific image information obtained based on the feature point setting information for the specific image information , The sharpness of specific image information can be calculated.

Here, the process of determining whether the information processing unit 200 can recognize a feature point or the number of feature points that can be recognized as a recognizable feature point can be implemented using a CUDA (Compute Unified Device Architecture) technology.

FIG. 5 is a view illustrating a state in which the corrected distance calculated through the information processing unit 200 according to an exemplary embodiment of the present invention is output through a separate screen.

When the correction distance in the specific area is calculated, the information processing unit 200 according to the present embodiment can output the correction distance in the calculated specific area through the screen of the user terminal provided separately. In this way, the user can easily and easily check the correcting distance in a desired area of the area where the information processing unit 200 is installed.

FIG. 6 is a diagram illustrating a process of calculating a correction distance using an image analysis timeline using artificial intelligence according to an embodiment of the present invention. Referring to FIG.

In order to quantitatively calculate the correcting distance by analyzing the image image using the artificial intelligence technology, the image processing system using the artificial intelligence according to the present embodiment may be configured such that the information processing unit 200 Some pixels among the plurality of pixels constituting the image information per distance are arbitrarily set as the minutiae points (S610), and the minutiae setting information for some pixels set as the minutiae points may be stored in advance.

When the image information acquiring unit 100 acquires the time information of distance-based image information according to a predetermined schedule, the photographing direction is adjusted based on the image information and the distance information to photograph a subject related to the stored image information, (S620). ≪ / RTI >

When each image information is obtained, the information processing unit 200 calculates the number of recognizable feature points from the specific image information obtained based on the feature point setting information for specific image information among the respective image information, (Step S630).

The information processing unit 200 determines whether the sharpness of the specific image information is equal to or greater than a preset value (S640). If the calculated sharpness is greater than or equal to a preset value (S640-Yes) If the sharpness of the image information for the first object is less than the predetermined value (S640-No), the image information obtaining unit 100 (Step S650), and determines whether the calculated sharpness of the image information of the second subject is equal to or greater than a predetermined value, The correcting distance in the area can be calculated (S670).

Here, the information processing unit 200 may calculate the correction distance based on the distance information of the second subject according to the determination result of the sharpness of the image information about the second subject, or may calculate the correction distance based on the image information obtaining unit 100 The sharpness calculation procedure can be repeatedly performed until image information whose sharpness is equal to or greater than a predetermined value is determined by calculating the sharpness of the image information for the third subject nearer to the subject than the predetermined distance interval.

When the corrective distance to the specific area is calculated by the information processing unit 200, the directional audio information output unit 300 outputs the first distance corresponding to the area in which the traveling road is located or the area where the vehicle traveling along the traveling road is located The voice information on the corrected distance can be outputted to the area (S680).

Through this, it is possible to inform the driver who is traveling in the area where the deterioration is caused by the information about the visibility of the area, thereby improving the alertness and preventing the loss of life and economic loss due to the deterioration of visibility.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be construed as limiting the scope of the invention as defined by the appended claims. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

100: Image information obtaining unit
200: Information processor
210: Sharpness calculation module
220: Time constant calculation module
230: storage module
300: directional voice information output unit

Claims (8)

delete An image information acquiring unit for acquiring image information of objects set for each distance;
The image processing method according to claim 1, further comprising the steps of: calculating image clarity of each image information when the image information on the objects is acquired; determining image information whose calculated sharpness is equal to or greater than a predetermined value; And an information processing unit for calculating an output value,
The information processing unit,
Wherein the image information acquiring unit matches the distance information with respect to the distance between the image information acquiring unit and the subject, the image information of the subject positioned at predetermined distance intervals based on the image information acquiring unit,
The image information obtaining unit obtains the image information of the first object located at the farthest position with reference to the image information obtaining unit based on the stored image information and the distance information, And acquires image information of the objects set sequentially by the distance from the start point of the image. 3. The method of claim 2,
The information processing unit,
Wherein the image processing method further comprises the steps of calculating the sharpness of the image information of the first subject among the respective pieces of image information to determine whether the calculated sharpness is equal to or greater than the predetermined value when the sharpness of each of the pieces of image information is calculated, Wherein the correcting step of the image information for the other objects is stopped and the corrective distance is calculated on the basis of the distance information about the first object if the difference is greater than or equal to the predetermined value, system. The method of claim 3,
The information processing unit,
Calculating a sharpness of image information for a second subject nearer to the first subject than the first subject based on the image information acquiring unit among the respective pieces of image information when the calculated sharpness is less than the predetermined value, Determining whether the calculated sharpness of the image information for the second subject is equal to or greater than a predetermined value,
The correcting distance may be calculated on the basis of the distance information of the second subject or the predetermined distance may be calculated based on the determination result of the sharpness of the image information with respect to the second subject, And calculating the sharpness of the image information for the third subject close to the third distance. The method of claim 3,
The information processing unit,
In order to calculate the sharpness of each of the image information, the minutiae setting information for a certain pixel among a plurality of pixels constituting the image information is arbitrarily set as the minutiae point, is stored in advance, and when specific image information is obtained And calculating the number of recognizable feature points from the obtained specific image information based on the feature point setting information for the specific image information to calculate the sharpness of the specific image information. . An image information acquiring unit for acquiring image information of objects set for each distance;
The image processing method according to claim 1, further comprising the steps of: calculating image clarity of each image information when the image information on the objects is acquired; determining image information whose calculated sharpness is equal to or greater than a predetermined value; And an information processing unit for calculating an output value,
And a directional audio information output unit for outputting audio information on the corrective distance in the specific area identified by the information processing unit in a first area set by a predetermined limited directional angle. Using Image Analysis System. The method according to claim 6,
Wherein the first region comprises:
An area where a driving road is located or an area where a vehicle running along the driving road is located. 8. The method of claim 7,
Wherein the directional voice information output unit comprises:
A radar for emitting a radiation wave toward a vehicle traveling along the traveling road and receiving reflected waves reflected by the vehicle is provided to detect a vehicle traveling along the traveling road, Image analysis system using artificial intelligence.

Images