KR101956891B1 - apparatus for expelling animals based on image recognition - Google Patents

Abstract

[0001] The present invention relates to an animal repelling apparatus based on image recognition, which comprises a tower main body supported on the ground and extending upward, a light irradiation unit mounted on the tower main body, A sound output unit mounted on the tower main body for outputting sound for exterminating an animal; a motion object detecting unit for detecting a motion object from an image picked up by the camera unit; And a control unit for controlling the light irradiating unit to irradiate only the light for exterminating the animal if the movement pattern of the motion object corresponds to the warning extermination condition. According to the image recognition-based animal suppression apparatus, an animal movement pattern can be analyzed from an image captured by a camera, thereby providing an advantage of adjusting an elimination condition.

Description

[0001] The present invention relates to an animal-

The present invention relates to an animal suppression apparatus based on image recognition, and more particularly, to an animal suppression apparatus based on image recognition capable of adjusting an elimination condition according to an animal movement pattern.

Generally, in agricultural areas and orchards where crops and fruit trees grow, small animals such as boars, pheasants and sparrows, as well as large animals such as wild boar, elk, etc., are increasingly infested with crops and fruit trees.

In the case of wild boar, there are no natural enemies at present, and the population is rapidly increasing. Therefore, the population of the wild boar is rising to the urban area beyond the mountain area, and the crops and fruits are damaged by their larger size than the damage caused by eating food, The damaged crops and fruit trees are worthless as a product because of the secondary damage such as broke the branches of the water. In some cases, these wild animals seriously threatens farmers' means of livelihood.

In order to curb wildlife encroachment, agricultural workers have used scarecrows, sparkles, and explosives to fight wildlife, but these are also only temporary effects, and aging rural communities It is difficult to implement the activities continuously.

Although a method of applying a fence to block intrusion of wild animals has been variously published, such as Korean Patent Laid-Open No. 10-2014-0106093, it is difficult to apply the fence in reality due to the economic reason that the construction cost is high.

Disclosure of Invention Technical Problem [8] Accordingly, the present invention has been made to solve the above problems, and it is an object of the present invention to provide an animal repelling device based on an image recognition which can detect movement of an animal by image recognition, There is a purpose.

According to an aspect of the present invention, there is provided an animal repelling apparatus based on an image recognition, comprising: a tower main body supported on a ground and extending upward; A light irradiating unit mounted on the tower main body and irradiating light for exterminating an animal to a monitored area; A camera unit mounted on the tower main body for capturing an image of a surveillance target area; An acoustic output unit mounted on the tower main body and outputting sounds for exterminating animals; The moving object detecting unit detects the moving object from the image captured by the camera unit and processes both the sound output unit and the light irradiating unit when the moving pattern of the moving object corresponds to the strong erasure condition, And a controller for controlling the light irradiating unit to irradiate only the light for exterminating the animal if the condition is met.

According to an aspect of the present invention, the control unit determines that the moving object is moving away from the set proximity position and corresponds to the warning firing condition, and operates the light irradiation unit to be activated.

The control unit determines that the motion object corresponds to the proximity horizontal movement in which the motion object is moved in the horizontal direction at the set proximity position, and processes the sound output unit and the light irradiation unit to be both activated.

In addition, the controller determines that the motion object corresponds to the alarm firing condition and corresponds to the remote horizontal movement in which the moving object is moved in the remote range out of the set proximity position.

In addition, if the movement object corresponds to the approach movement moving in the direction approaching the set proximity position, the controller determines that the motion object corresponds to the strong erasure criterion and processes both the sound output unit and the light irradiation unit.

The animal repelling device based on the image recognition according to the present invention provides an advantage of being able to analyze an animal movement pattern from an image picked up by a camera and adjust an elimination condition.

FIG. 1 is a side view schematically showing an animal repelling device based on an image recognition according to the present invention,
FIG. 2 is a block diagram showing a control system of an animal suppression apparatus based on the image recognition of FIG. 1,
FIG. 3 is a flow chart showing an animal repelling process of the control unit of FIG. 2,
Figs. 4 to 8 are views for explaining an example of an animal movement pattern in a screen captured through a camera. Fig.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image capturing device based on an image recognition according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a side view schematically showing an animal repelling apparatus based on an image recognition according to the present invention, and FIG. 2 is a block diagram showing a control system of an animal repelling apparatus based on the image recognition shown in FIG.

1 and 2, an animal suppression apparatus 100 based on image recognition according to the present invention includes a tower main body 110, a light irradiation unit 120, an acoustic output unit 130, a camera unit 140, (Not shown).

The tower main body 110 is supported on a ground surface for monitoring the monitored area and is installed to extend upward.

The tower main body 110 may be constructed so that the camera 142 and the light irradiation unit 120 described later can mount the light source 125 at an appropriate height.

The light irradiating unit 120 is mounted on the tower main body 110 and irradiates light for exterminating the animal to the surveillance target area.

The light irradiating unit 120 includes a first relay 121 for turning on and off the power supply from the power supply unit 160 to the light source 125 and a second relay 121 for supplying power to the light source 125 provided between the first relay 121 and the light source And a dimming adjusting unit 123 for adjusting brightness through power adjustment.

Here, the first relay 121 and the dimming adjusting unit 123 are controlled by the controller 152.

The light source 125 is constructed so as to emit light with an output suitable for animal extermination.

The sound output unit 130 is mounted on the tower main body 110 and outputs sound for exterminating the animal.

The sound output unit 130 includes a second relay 132 for turning on and off the power supply from the power supply unit 160 to the speaker 135 and a second relay 132 which is provided between the second relay 132 and the speaker 135, And a volume adjusting unit 133 that adjusts the volume through power adjustment supplied to the speaker 135.

Here, the second relay 132 and the volume adjusting unit 133 are controlled by the controller 152.

The camera unit 140 is mounted on the tower main body 110 to capture a surveillance target area.

The camera unit 140 includes a camera 142 for capturing an image of a surveillance target area and a camera driver 144 for adjusting the image capturing direction of the camera vertically and horizontally under the control of the control unit 152. [

It is needless to say that, unlike the example shown in the drawing, the camera unit 140 is not controlled in the rotation of the camera 142, but only in the fixed direction.

The storage unit 154 is controlled by the control unit 152 to record the recording object information.

The control unit 152 controls the imaging direction of the camera 142 through the camera driving unit 144 according to the set monitoring pattern, detects motion objects from the images captured by the camera 142, Stores it in the storage unit 154, analyzes the movement pattern of the motion object, and adjusts the elimination condition.

The control unit 152 processes both the sound output unit 130 and the light irradiation unit 120 so that the motion pattern of the motion object corresponds to the strong erasure condition. If the movement pattern of the motion object corresponds to the warning erasure condition, Only the light irradiating unit 120 is operated so that only the irradiation light is irradiated.

The control unit 152 judges that the object corresponds to the strong erasure condition if it corresponds to the proximity horizontal movement in which the motion object moves in the horizontal direction at the set proximity position and the access movement in the direction in which the motion object approaches the set proximity position .

In addition, when the moving object moves in a direction moving away from the set proximity position and corresponds to a remote horizontal movement in which the moving object moves in the remote area away from the set proximity position, As shown in FIG.

An example of the control process of the controller 152 will be described with reference to FIG.

Hereinafter, the control unit 152 determines that the original distance in the image screen captured by the camera 142 is determined based on the Y coordinate value of the pixel and the horizontal movement is determined based on the X coordinate value.

First, an image of the camera 142 is acquired (step 210).

Next, it is determined whether there is a motion object from the image acquired by the camera 142 (step 220).

Here, the determination of whether there is a motion object applies a normal image recognition processing method of extracting a motion object through comparison between image frames.

If it is determined in step 220 that there is a motion object, the object frame is generated by performing outline processing such as a rectangle on the area of the motion object (step 230).

In other words, reference numeral 149 in FIG. 4, in which an image screen captured by the camera 142 is illustrated, becomes an object frame.

Next, it is determined whether the Y value of the object frame is a neighboring area (step 240).

Here, the neighboring area may be appropriately set. For example, when the screen is divided into three in the vertical direction, i.e., the Y direction, the area from the bottom of the screen to the 1/3 position may be set as the adjacent area.

If it is determined in step 240 that the Y value of the object frame is not a proximity area, it is determined whether the object frame Y corresponds to an approaching movement in a direction approaching the tower body 110 (step 250).

That is, as shown in FIG. 4, when a moving object moves in alphabetical letters a, b, and c with respect to the same imaging direction over time, and the Y value of the object frame 149 is moved to the bottom of the screen, .

If it is determined in step 250 that the approach movement as shown in FIG. 4 is determined, it is determined that the power is the strong power failure condition, and the sound output unit 130 is activated to perform sound erasure (step 260) (Step 270).

Otherwise, if it is determined in step 250 that the access movement is not the movement, the movement of the object frame 149 determines the horizontal movement (step 280).

Here, as illustrated in FIGS. 5 and 6, in the horizontal movement, the motion object is moved to a, b, and c with a lapse of time at a position deviated from the proximity area, and the X value of the object frame 149 corresponds to the left The horizontal movement is determined.

Similarly, if the X value of the object frame 149 moves from the right side of the screen to the left side, it is judged as the horizontal movement.

If it is determined in step 280 that there is a horizontal movement, it is determined as a warning prevention condition and step 270 is performed.

Here, the control unit 152 may apply the output of light differently according to the access distance of the motion object through the adjustment of the dimming adjustment unit 123.

That is, the controller 152 may be constructed to adjust the output of the light applied in FIG. 5 higher than in FIG.

On the other hand, if it is determined in step 240 that the Y value of the object frame is a neighboring area, it is determined whether the object frame X corresponds to a horizontal movement (step 290)

That is, as shown in FIG. 7, when the moving object moves to a, b, c according to a lapse of time at an X value corresponding to the object frame 149 corresponding to the motion object, .

Also, even if the X value moves from right to left as time elapses at a position where the object frame 149 corresponding to the motion object corresponds to the adjacent area, it is determined that the object is a horizontal movement.

If it is determined in step 290 that the horizontal movement is determined, the process proceeds to step 260. [

Otherwise, it is determined in step 290 whether the Y value of the object frame 149 corresponds to a retreat movement to move away from the tower main body 110 (step 300).

That is, as shown in FIG. 8, when a moving object moves in alphabetical letters a, b, and c with respect to the same imaging direction over time, and when the Y value of the object frame 149 corresponding to the moving direction moves to the top of the screen, .

If it is determined in step 300 that the movement is to be retired, it is determined that the condition corresponds to the warning removal condition, and the process proceeds to step 270.

On the other hand, the control unit 152 also applies the volume of the sound output unit 130 through the volume adjustment unit 133 according to the approach distance of the motion object.

The animal recognition system based on the above-described image recognition operates in a severe eradication condition for the acoustic output and does not apply the acoustic output in the warning erasure condition, thereby minimizing the noise damage to the residential area.

In addition, an animal moving pattern can be analyzed from an image picked up by a camera to adjust an elimination condition, thereby providing an advantage of simplifying installation scale.

110: Tower main body 120:
130: Audio output unit 140: Camera unit
152:

Claims (5)

A tower main body supported by the ground and extending upward;
A light irradiating unit mounted on the tower main body and irradiating light for exterminating an animal to a monitored area;
A camera unit mounted on the tower main body for capturing an image of a surveillance target area;
An acoustic output unit mounted on the tower main body and outputting sounds for exterminating animals;
The moving object detecting unit detects the moving object from the image captured by the camera unit and processes both the sound output unit and the light irradiating unit when the moving pattern of the moving object corresponds to the strong erasure condition, And a controller for controlling the light irradiating unit to irradiate only the light for exterminating the animal if the condition is met,
The control unit
When the moving object moves in a direction away from the set proximity position, it is determined that the motion object corresponds to the warning firing condition and the light irradiation unit is activated,
The control unit
When the moving object corresponds to a proximate horizontal movement that moves in a horizontal direction at a predetermined proximity position, it is determined that the motion object corresponds to the strong erasure condition, and both the sound output unit and the light irradiation unit are activated,
The control unit
When the moving object corresponds to a remote horizontal movement in which the moving object moves in the horizontal direction in a remote area deviated from the set proximity position, the light irradiation unit is determined to correspond to the warning firing condition,
The control unit
When the movement object moves in a direction approaching the set proximity position, it is determined that the motion object corresponds to the robust fighting condition and the both of the sound output unit and the light irradiation unit are activated,
Wherein the control unit judges the near distance of the motion object in the image screen captured by the camera through the Y coordinate value of the pixel and determines the horizontal movement through the X coordinate value, Animal control system.
delete delete delete delete

Images