KR20180130377A - Smart helmet for fire-fighting - Google Patents

Abstract

The present invention relates to a smart firefighting helmet comprising: an image photographing unit attached to an outer surface of a helmet to photograph the front of the helmet to obtain an image; an image processing unit formed in the helmet to receive the image obtained by the image photographing unit in real time, to perform image processing to remove smoke included in the obtained image so as to generate a two-dimensional image, and so as to integrate the generated two-dimensional image to generate a three-dimensional image; and a display unit attached to an inner surface of the helmet to receive and output the three-dimensional image generated by the image processing unit. Accordingly, the three-dimensional image in which a scattering medium is removed from the photographed image is outputted to the display unit formed inside the helmet. Thus, the present invention can be efficiently used to perform a rescue work in a space where smoke or flame is full by a firefighter wearing the smart firefighting helmet.

Description

Smart fire helmet {SMART HELMET FOR FIRE-FIGHTING}

FIELD OF THE INVENTION The present invention relates to a smart fire helmet, and more particularly, to a smart fire helmet which removes smoke from a video image taken in a fire place and makes it available for fire fighting rescue operations.

In the fire fighting activity to rescue the rescue person in the field of fire, it is difficult to secure visibility of the firefighter due to flame or smoke, which limits the rescue operation.

Korean Patent No. 10-1656873 discloses a technology for acquiring an image through thermal imaging in a helmet worn by a firefighter in a three-dimensional recognition firefighting helmet and displaying the acquired image so that a firefighter wearing the helmet can confirm it have.

However, in the conventional helmet, there is a limit in that it is not possible to properly detect a structure object or to detect a structural object due to smoke by detecting a flame of a fire scene as heat.

Accordingly, it is necessary to develop a fire helmet capable of obtaining a clear image in order to quickly and accurately structure the object in a fire scene where it is difficult to obtain a visibility due to flame or smoke.

SUMMARY OF THE INVENTION The present invention provides a fire fighting helmet for firefighters, which performs rescue operations in a fire scene, to output a fire or smoke-free image to improve fire fighting activity efficiency.

The smart fire fighting helmet according to an embodiment of the present invention includes an image capturing unit attached to an outer surface of a helmet to capture an image of the front of the helmet to acquire an image; The helmet may include an image capturing unit that receives the image captured by the image capturing unit in real time, performs image processing for removing smoke contained in the captured image to generate a two-dimensional image, An image processor for generating a three-dimensional image; The display unit is attached to an inner surface of the helmet, and receives and outputs the 3D image generated by the image processing unit.

The display unit includes a first output unit and a second output unit for receiving the left eye image and the right eye image for outputting the augmented reality image and outputting the left eye image and the right eye image, respectively.

And a communication unit attached to the helmet and receiving the 3D image generated by the image processing unit and delivering the 3D image to an external server through wireless communication.

The external server generates a control signal for controlling the image processing unit and transfers the generated control signal to the communication unit. The external server receives and stores the 3D image generated by the image processing unit through the communication unit.

And an LED formed outside the front surface of the helmet to output light.

The LED detects the illuminance around the helmet and is automatically turned on or off.

The image processing unit obtains a perfluorogram from the image acquired by the image capturing unit, estimates a scattering medium, removes an estimated scattering medium from the acquired perpetogram, and detects a photon to generate a two-dimensional original image And reconstructing the three-dimensional image using the generated two-dimensional original image to generate the three-dimensional image.

According to another aspect of the present invention, there is provided a smart fire fighting helmet according to the present invention, wherein the smart fire fighting helmet is attached to an upper part of a front surface of a helmet to process an image acquired by the image capturing unit, So that the working efficiency of the firefighter performing the structural work can be improved.

1 is a schematic view of a smart fire helmet according to an embodiment of the present invention.
2 is a block diagram showing a schematic structure of a smart fire helmet according to an embodiment of the present invention.
3 is a flowchart illustrating an image processing flow of a smart fire helmet according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Hereinafter, the structure of a smart fire helmet according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a view showing a schematic configuration of a smart fire helmet according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a schematic structure of a smart fire helmet according to an embodiment of the present invention.

The smart fire helmet 1 according to the present embodiment includes a first protection part 11, a second protection part 12 and a third protection part 13 and is a helmet shape for protecting a face and a head of a person, The first protection portion 11 protects both the upper face of the face and the back of the head, the second protection portion 12 protects the face portion of the person, and the third protection portion 13 protects the lower face of the person do.

The second protection unit 12 is coupled to the first protection unit 11 and is mounted on the first protection unit 11 or on the first protection unit 11 based on a portion coupled to the first protection unit 11, And the shape can be changed by an externally applied force.

At this time, the second protection part 12 is formed to extend from the first protection part 11, and the end of the second protection part 12 may be protected by the third protection part 11 unless the first protection part 11 is connected thereto. The smart fire fighting helmet 1 is formed so as to contact with the smart fire fighting helmet 1 such that smoke or noxious gas does not enter into the smart fire fighting helmet 1 when the smart fire fighting helmet 1 is worn, good.

As described above, the smart fire helmet 1 including the first protection part 11, the second protection part 12 and the third protection part 13 is in the form of a general helmet shape, and in one example, The protective portion 12 may be formed of a translucent polarizing plate.

In one example, the smart fire helmet 1 includes a display unit 120, an image capturing unit 20, an LED 30, an image processing unit 40, a control unit 50, and a communication unit 60, 120 are coupled to the second protection unit 12 to receive an image from the control unit 50 and output the image.

The display unit 120 includes a first display unit 121 and a second display unit 122. The display unit 120 is formed at a position corresponding to the left and right eyes of a user wearing the smart fire helmet 1, (120) may be attached to the inner surface of the second protective portion (12).

The first display unit 121 and the second display unit 122 constituting the display unit 120 are spaced apart from each other and receive a left-eye output image and a right-eye output image from the control unit 50, respectively.

As the first display unit 121 and the second display unit 122 output the images received from the controller 50, the user wearing the smart fire helmet 1 can display the images output to the display unit 120 in three dimensions Or an augmented reality image.

The image capturing unit 20 is a camera attached to the outer surface of the smart firefighter helmet 1 and capturing an image. The camera captures the front of a user wearing the smart firefighter helmet 1 as a moving picture, .

The image capturing unit 20 can capture a moving image according to a control signal received from the outside.

The LED (light emitting diode) 30 is attached to the outer surface of the smart fire helmet 1 and outputs light. The light emitting diode 30 outputs light toward the front of the user wearing the smart fire helmet 1.

The LED 30 may be controlled to be turned on or off according to a control signal, or may be turned on or off automatically by sensing ambient illuminance, but is not limited thereto.

The image processing unit 40 receives the moving image photographed by the image photographing unit 20 in real time from the image photographing unit 20 and processes the processed moving image to the control unit 50, And control the moving picture based on the control signal.

Referring to FIG. 3, a method of processing the image received from the image capturing unit 20 by the image processing unit 40 will be described in detail. 3 is a flowchart illustrating an image processing method of a smart fire helmet according to an embodiment of the present invention. The image processing method of the smart fire helmet according to the present embodiment includes the steps of acquiring a single perpetogram (S1), estimating a scattering medium (S2), removing a scattering medium in a single perfluorogram (S3) An original image reconstruction step S4, and a 3D image reconstruction step S5.

An image received by the image processing unit 40 from the image capturing unit 20 is acquired as a single peak image (S1) as an image recorded by a natural light source and an image sensor (S1) The image is a superposition of scattered rays generated by light rays from an object passing through the scattering medium, and the scattered rays for each scattering medium having a unit area shape are statistically independent of each other.

Next, the image processing unit 40 estimates the scattering medium included in the single perfluorogram formed by superposition of the scattering medium (S2) using the central limit theorem.

이고, 가우시안(Gaussian) 분포를 따른다.In the scattering medium estimation step S2, the scattering medium included in the single perfluorogram is a form in which a plurality of scattering media in the unit area are overlapped, and the scattering medium in the unit area is

And follows a Gaussian distribution.

인 경우, 단일 페플로그램(

)은

로 표현되고, 이때,

,

,

, and

이다. Then, the pixel resolution of the image capturing unit 20

, A single perphogram (

)silver

, ≪ / RTI >

,

,

, and

to be.

와 분산

을 갖는 가우시안 분포를 따르는 산란 매질을 모델링한다.More specifically, in the scattering medium estimation step S2, the image processing unit 40 is a central limit theorem for estimating a scattering medium from a single per-flowgram. The maximum likelihood estimation (MLE) The scattering medium is estimated by applying the statistical inference method,

And dispersion

Lt; RTI ID = 0.0 > Gaussian < / RTI >

And, i and j are indexes of the photon parts scattered in the x and y directions, respectively, and the Gaussian variable is expressed by the following equation (1).

[Equation 1]

In Equation 1 above, I is the pixel intensity of the perpetogram.

)를 이용하여 산란 매질의 부분 영역으로서 가우시안 분포에 확률적으로 실제와 최대한 가까운 값(

)을 추정하고, 다음의 수식 2와 같이 산란 매질의 부분 영역으로 추정되는 변수

는 산란 빛의 각 부분의 샘플 평균값으로 산출된다.The image processing unit 40 calculates the average value of the Gaussian distribution

) Is used as a partial region of the scattering medium to give a probability distribution of the Gaussian distribution as close to the actual value as possible

), And as shown in the following equation (2), a variable estimated as a partial region of the scattering medium

Is calculated as a sample average value of each part of the scattered light.

[Equation 2]

The maximum likelihood estimation (MLE), which is a statistical estimation method used in estimating the scattering medium using the central limit theorem in Equation (2), is expressed by the following Equations (3) to (5).

[Equation 3]

[Equation 4]

[Equation 5]

)상의 표본 평균을 산란 매질 평균값으로 추정함을 알 수 있다.In Equations (4) and (5), the scattering medium plane

) Is estimated as the scattering medium mean value.

Next, the image processing unit 40 performs the step (S3) of removing the estimated scattering medium in the obtained single perfogram according to the following Equation (6).

[Equation 6]

Then, the image processing unit 40 performs a step S4 of recovering the two-dimensional original image by detecting the photons in the perturbation-processed cepstrum. At this time, the normalized perpetogram is expressed as Equation 7 To detect the living photons from the processed Peplogram.

[Equation 7]

는 검출하고자 하는 광자의 수, c는 색상 채널(R채널, G채널 및 B채널), 그리고

는 개개의 색상 채널에 대한 탄도 광자의 계수(coefficient)이고, 정규화된 페플로그램은

이다.In Equation (7) above,

Is the number of photons to detect, c is the color channel (R channel, G channel and B channel), and

Is the coefficient of the trajectory photon for the individual color channel, and the normalized perphogram is

to be.

More specifically, the processed Pepflogram model photon counts using the photon count imaging method modeled by the Poisson distribution because it has a small number of surviving photons, where the image must have a single energy Therefore, the image processing unit 40 performs normalization using the sum of the total pixel intensities for the object or the scenery passing through the scattering medium by the following Equation (8) to generate a normalized image having a single energy.

[Equation 8]

In order to extract a desired number of photons from the above equation (7), the image processing unit 40 can apply the Poisson random generation method as expressed by the following equation (9).

[Equation 9]

In this manner, the image processing unit 40 generates a perfluorogram including the photons of survival from the equations (8) and (9). However, since the generated perphograms do not contain a sufficient number of surviving photons to visualize the object, it is possible to use the multiple restored perpetograms and maximum likelihood estimation (MLE) . A likelihood function can be constructed from a plurality of reconstructed perphograms, and the scenery can be estimated using Equations (10) to (14).

[Equation 10]

[Equation 11]

[Equation 12]

[Equation 13]

[Equation 14]

The image processing unit 40 estimates the two-dimensional original image scenery that has passed through the scattering medium, and finally, the image processing unit 40 performs the above-described step S4 by using the computer- Dimensional original images are restored as a three-dimensional image (S5).

In the above step S5, the image capturing unit 20 can record a plurality of two-dimensional images having a perspective by applying a lens array or a camera array to the three-dimensional object through the integrated image technique, and the image processing unit 40 Can reconstruct an image acquired by the image capturing unit 20 at each depth and reconstruct it into a three-dimensional image.

In one example, the image processing unit 40 may perform a computer integrated image restoration process from the following equations (15) and (16).

[Equation 15]

[Equation 16]

는 요소 영상에서 X축 방향으로의 픽셀의 수이고,

는 요소 영상에서 Y축 방향의 픽셀 수이다. 또한, p는 가상 핀홀 사이의 간격이고,

, 는 X축 방향으로의 이미지 센서의 크기,

는 Y축 방향으로의 이미지 센서의 크기이다. z 는 재생된 이미지의 깊이이고,

는 X축 방향으로 중첩되는 이동 픽셀의 수,

는 Y축 방향으로 중첩되는 이동 픽셀의 수이다.

는 k번째 행(row), l번째 열(columm)의 복원된 페플로그램이고, O(x, y)는 중첩되는 수를 나타내는 매트릭스이다.In the above Equation 15 and Equation 16,

Is the number of pixels in the X-axis direction in the element image,

Is the number of pixels in the Y axis direction in the element image. Also, p is the interval between virtual pinholes,

, Is the size of the image sensor in the X-axis direction,

Is the size of the image sensor in the Y-axis direction. z is the depth of the reproduced image,

The number of moving pixels superimposed in the X-axis direction,

Is the number of moving pixels superimposed in the Y-axis direction.

(X, y) is a reconstructed perphogram of the kth row and the lth column (columm), and O (x, y) is a matrix representing the number of overlaps.

As described with reference to FIG. 3, the image processing unit 40 reconstructs a three-dimensional image and transfers the reconstructed image to the control unit 50. FIG.

At this time, the image processing unit 40 can adjust and process the parameters necessary for image restoration according to the control signal received from the control unit 50, but it is not limited thereto.

The control unit 50 controls the image processing unit 40 to transmit the three-dimensional image processed by the image processing unit 40 to the external server 100 via the communication unit 60 or to transmit the image to the display unit 120, And controls the first display unit 121 and the second display unit 122, which form the display unit 120, to output appropriate images, respectively.

Alternatively, the control unit 50 can be used for generating a control signal by receiving the signal or the like received from the external server 100 from the communication unit 60, but is not limited thereto.

The communication unit 60 receives the three-dimensional image processed by the image processing unit 40 through the control unit 50 and transmits the received three-dimensional image to the external server 100 or transmits the control signal of the control unit 50 to the external server 100 And delivers signals and the like received from the external server 100 to the controller 50. [

In one example, the communication unit 60 may include a wireless communication module for performing communication with an external server 100, and may be connected to the nearby server 100 through wired communication, but is not limited thereto.

The server 100 may control the smart fire helmet 1 by storing a three-dimensional image received from the communication unit 60 or transmitting an input control signal to the communication unit 60.

As described above with reference to FIGS. 1 to 3, the smart fire helmet 1 according to the embodiment of the present invention outputs a three-dimensional image obtained by removing a scattering medium from a photographed image to a display unit 120 formed inside the helmet , A firefighter wearing the smart firefighter helmet 1 can be effectively used to perform a rescue operation in a smoke or flame-covered space.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

1: smart fire helmet 20: image shooting part
40: image processing unit 50:
60: communication unit 120:

Claims (7)

An image capturing unit attached to an outer surface of the helmet to capture an image of the front of the helmet to acquire an image;
The helmet may include an image capturing unit that receives the image captured by the image capturing unit in real time, performs image processing for removing smoke contained in the captured image to generate a two-dimensional image, An image processor for generating a three-dimensional image; And,
And a display attached to an inner surface of the helmet to receive and output the 3D image generated by the image processing unit.
The method according to claim 1,
Wherein the display unit comprises a first output unit and a second output unit for receiving the left eye image and the right eye image for outputting the augmented reality image and outputting the left eye image and the right eye image, respectively.
The method according to claim 1,
And a communication unit attached to the helmet and receiving the 3D image generated by the image processing unit and delivering the 3D image to an external server through wireless communication.
The method of claim 3,
Wherein the external server generates a control signal for controlling the image processing unit and transmits the control signal to the communication unit, or receives the three-dimensional image generated by the image processing unit through the communication unit and stores the received three-dimensional image.
The method according to claim 1,
Further comprising an LED formed outside the front surface of the helmet to output light.
6. The method of claim 5,
Wherein the LED is automatically turned on or off by detecting the illuminance around the helmet.
The method according to claim 1,
The image processing unit obtains a perfluorogram from the image acquired by the image capturing unit, estimates a scattering medium, removes an estimated scattering medium from the acquired perpetogram, and detects a photon to generate a two-dimensional original image Dimensional image, and the three-dimensional image is generated by restoring the three-dimensional image using the generated two-dimensional original image.

Images