KR101949968B1 - Apparatus and method for determining the visibility distance of an image containing a fog component - Google Patents

Abstract

The present invention provides a method and apparatus for determining the corrective distance of an image containing a fog component. According to an embodiment, the correction distance determining apparatus may calculate a transmission amount of an input image by patch region by applying a DCP (Dark Channel Prior) algorithm to an input image including a fog component. Further, the correction distance determining apparatus may correct the calculated amount of transmission using a guided filter, and determine a fog boundary line in the input image based on the corrected transmission amount. The corrective distance determining apparatus can determine the corrective distance of the input image based on the fog boundary line.

Description

[0001] APPARATUS AND METHOD FOR DETERMINING THE VISIBILITY DISTANCE OF AN IMAGE CONTAINING A FOG COMPONENT [0002]

The present disclosure provides an apparatus and method for determining the corrective distance of an image containing a fog component.

To prevent this, the marine accidents are increased when the marine accident occurs. However, as the vessel operation is controlled without quantitative and scientific grounds, there is a need to continuously improve the regulations of the ship 's people.

In order to solve this problem, it is necessary to calculate a quantitative correction distance. However, in the case of the currently used optical visual system, only the correction distance for the observation point is measured.

Accordingly, there is a need for a technique for calculating a correction distance based on an image taken by a camera for a two-dimensional visibility observation as a human eye observes.

And an apparatus and method for determining a corrective distance of an image including a fog component. The technical problem to be solved by this embodiment is not limited to the above-mentioned technical problems, and other technical problems can be deduced from the following embodiments.

According to a first aspect of the present disclosure, there is provided a method of determining a corrective distance of an image including a fog component, the method comprising: receiving an input image including a fog component; Calculating a value of a dark channel for each patch region of the input image by applying a DCP (Dark Channel Prior) algorithm to the input image; Calculating an amount of transmission per patch area based on the dark channel value; Correcting the calculated transmission amount using a guided filter; Binarizing the input image based on the corrected amount of transmission and determining a fog boundary line in the binarized input image; And determining a corrective distance of the input image based on the fog boundary line.

According to a second aspect of the present disclosure, there is provided an apparatus for determining a corrective distance of an image including a fog component, the apparatus comprising: a memory in which at least one program is stored; A communication unit for receiving an input image including a fog component; And applying a DCP (Dark Channel Prior) algorithm to the input image to calculate a value of a dark channel for each patch region of the input image by executing the at least one program stored in the memory, Calculating a transmission amount per patch area, correcting the calculated amount of transmission using a guided filter, binarizing the input image based on the corrected transmission amount, and calculating a fog boundary line from the binarized input image And determining a corrective distance of the input image based on the fog boundary line.

The third aspect of the present disclosure can provide a computer-readable recording medium on which a program for causing a computer to execute the method of the first aspect is recorded.

According to the present invention, a fog component of an image can be determined using a DCP (Dark Channel Prior) algorithm. In order to improve the calculation speed, a method and an apparatus using a guided filter in applying the DCP (Dark Channel Prior) algorithm are proposed. In addition, a more sophisticated distance map is generated to calculate the correction distance of the image, thereby proposing a scientific and quantitative correction distance calculation method and apparatus.

1 is a flow chart of a method for determining the corrective distance of an image containing a fog component according to an embodiment.
FIG. 2 is a diagram for explaining a DCP (Dark Channel Prior) algorithm according to an embodiment.
3 is a diagram for explaining an example of correcting the amount of transmission using a guided filter according to an embodiment.
4 is a diagram for explaining an example of determining a fog boundary line in an input image according to an embodiment.
5A to 5D are diagrams for explaining an example of generating a distance map according to an embodiment.
6 is a view for explaining an example of correcting the distance value in consideration of the tide difference according to the embodiment.
FIG. 7 is a diagram for explaining an example of performing an element-wise product operation between a fog boundary line and a distance map according to an embodiment.
8 is a diagram for explaining an example of determining a correction distance of an input image according to an embodiment.
FIG. 9 is a block diagram illustrating a hardware configuration of a fixation distance determining apparatus according to an embodiment.

The phrases " in some embodiments " or " in one embodiment " appearing in various places in this specification are not necessarily all referring to the same embodiment.

Some embodiments of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented with various numbers of hardware and / or software configurations that perform particular functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors, or by circuit configurations for a given function. Also, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented with algorithms running on one or more processors. In addition, the present disclosure may employ conventional techniques for electronic configuration, signal processing, and / or data processing, and the like. Terms such as " mechanism, " " element, " " means, " and " configuration " and the like are widely used and are not limited to mechanical and physical configurations. Also, the terms " part ", "? Module ", etc. in the specification mean a unit for processing at least one function or operation, which may be implemented by hardware or software or by a combination of hardware and software.

Also, the connection lines or connection members between the components shown in the figures are merely illustrative of functional connections and / or physical or circuit connections. In practical devices, connections between components can be represented by various functional connections, physical connections, or circuit connections that can be replaced or added.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1 is a flow chart of a method for determining the corrective distance of an image containing a fog component according to an embodiment.

Referring to FIG. 1, in step 110, the corrective distance determining apparatus may receive an input image including a fog component.

The corrective distance determining apparatus can receive an input image from an external camera for a photographing region photographed by an external camera. In one embodiment, the external camera generates an input image composed of RGB for the shooting area for monitoring, security, or information providing purposes, and the correction distance determining device can receive the input image from the external camera. The external camera may be, but is not limited to, Closed-Circuit Television (CCTV).

The input image received by the correction distance determining apparatus may include a fog component in accordance with the weather condition of the photographing region photographed by the external camera. In one embodiment, when the corrective distance determining apparatus receives an input image from an external camera that photographs a specific area of the sea, the input image may include a fog component, i.e., a sea component.

In the meantime, the camera may be mounted on the correction distance determining device. In this case, the correction distance determining device can photograph a specific area using the mounted camera and directly generate an input image for the photographing area.

In step 120, the corrective distance determining apparatus may determine a dark channel value for each patch region of the input image by applying a DCP (Dark Channel Prior) algorithm to the input image.

The correction distance determining apparatus can separate an input image into RGB channels. The dark channel means a channel having the minimum brightness value among the respective color channels, that is, the R, G, and B channels. Further, the correction distance determining apparatus may divide the input image into patch regions and determine a channel having a minimum brightness value among the R, G, and B channels for each patch region as a dark channel for the patch region. In one embodiment, the dark channel value may refer to a pixel value of a dark channel for a particular patch region, but is not limited thereto.

In step 130, the corrective distance determining device may calculate the amount of transmission per patch area based on the dark channel value.

The amount of transmission can be calculated using the property that the light is not scattered and reaches the camera and the dark channel value of the patch region in which the fog component exists has a large value.

In step 140, the corrective distance determining apparatus can correct the calculated amount of transmission using a guided filter.

Since the process of determining the dark channel value in step 120 and calculating the amount of transmission in step 130 is performed for each patch area, stepping may occur in a boundary part of the patch area in the process of restoring the fog removed image. That is, when the fog removed image is restored, the accuracy may be lowered.

Conventionally, a soft matting algorithm is applied to prevent a reduction in precision, but a soft matting algorithm has a slow computation speed because it performs Laplacian transformation requiring a high computational complexity.

Instead of the soft matting algorithm, the correction distance determining apparatus can correct the amount of transmission calculated using the guarded filter in real time, thereby preventing a reduction in accuracy.

In one embodiment, the input image may include a light source (e.g., solar, sunlight, headlight, etc.). When the size of the light source included in the input image is larger than the size of the patch area, the corrective distance determining apparatus can classify the light source as a mist component. In order to prevent the light source from being misclassified as mist components, the correction distance determining apparatus can adjust the amount of transmission of a specific patch region when the dark channel value for a specific patch region in the input image is equal to or greater than a predetermined threshold value. That is, in this case, the corrective distance determining apparatus can prevent the light source from being misclassified as a mist component by adjusting the amount of transmission of the specific patch region.

In step 150, the corrective distance determining device can determine the fog boundary line in the input image based on the corrected amount of transmission.

In one embodiment, the corrective distance determining device may binarize the input image based on the corrected transmission amount and determine a fog boundary line in the binarized input image.

For example, the corrective distance determining apparatus applies a first color to a patch region whose corrected transmission amount is equal to or greater than a predetermined threshold value in a patch region of an input image, and applies a second color to a patch region whose corrected transmission amount is less than a predetermined threshold value Can be applied. For example, the first color may be white and the second color may be black.

The corrective distance determining apparatus can determine the boundary line between the patch region of the first color and the patch region of the second color as the fog boundary line.

In step 160, the corrective distance determining device can determine the corrective distance of the input image based on the fog boundary.

In one embodiment, the corrective distance determining device can determine the corrective distance by calculating an average of the corrective distance on the fog boundary line or by selecting the minimum corrective distance on the fog boundary line.

Further, the correction distance determining apparatus may divide the input image into a plurality of regions, and then provide an average correction distance and / or a minimum correction distance for a region of interest selected by the user among the plurality of regions.

On the other hand, the correction distance determining apparatus can use the distance map to determine a more accurate correction distance for the input image. The corrective distance determining apparatus performs an element-wise product operation between a matrix value representing a fog boundary line of an input image and a matrix value representing a distance value of the distance map, thereby calculating a distance of a pixel corresponding to a fog boundary of the input image Value can be determined. The corrective distance determining device can determine the corrective distance for the input image by using the distance value of the fog boundary line. The description of the distance map will be described later in Figs. 5A to 5D.

FIG. 2 is a diagram for explaining a DCP (Dark Channel Prior) algorithm according to an embodiment.

Referring to Fig. 2, the correction distance determining apparatus is capable of receiving an input image from a camera 210 (a camera mounted on an external camera or a corrective distance determining apparatus) to a photographing area 220 photographed by the camera 210 have. The input image may contain fog components.

First, an input image including a fog component can be expressed by the following equation (1). (X) is the original image from which the fog component is removed, or the scene radiance of the subject, A is the global atmospheric light, t (x) is the input image containing the fog component, ) Means transmission.

The transmittance t (x) depends on the depth of the image, and can be expressed as shown in Equation 2 below. Where β is the scattering coefficient of the atmosphere and d (x) is the depth at the x position of the image.

On the other hand, β is a value that varies with the weather, and if there is no fog component in the input image, β = 0 (ie t (x) = 1, I (x) = J The larger the recording β.

The Dark Channel Prior (DCP) algorithm (hereinafter referred to as the DCP algorithm) is an algorithm for obtaining a relative image depth instead of actually obtaining the absolute image depth in the input image. The input image can be separated into RGB channels, and a channel having the minimum brightness value among the respective color channels, that is, the R, G, and B channels, is called a dark channel.

)은 아래 수학식 3을 통해 결정될 수 있다. 아래 수학식 3에서, x 및 y는 2차원 벡터로서 이미지에서의 위치,

는 입력 이미지의 R, G 및 B 채널 별 이미지,

는 x를 중심으로 하는 입력 이미지의 패치 영역을 의미한다. 예를 들어, 는

패치 영역을 의미할 수 있다.Dark channel value (

Can be determined through Equation (3) below. In the following Equation (3), x and y are positions in the image as two-dimensional vectors,

Is an image for each R, G and B channel of the input image,

Is the patch area of the input image centered at x. For example,

May refer to a patch area.

)은 0이 된다. 반면, 안개 성분이 존재하는 입력 이미지에서는 A(1-t(x)) 성분이 더해짐에 따라(수학식 1 참조) 다크 채널 값(

)은 큰 값을 갖는다.Through empirical observations, the dark channel value of the input image without the fog component (

) Becomes zero. On the other hand, as the A (1-t (x)) component is added in the input image in which the fog component exists (see Equation 1), the dark channel value

) Has a large value.

The amount of transmission represents the extent to which the light is not scattered and reaches the camera, and the amount of transmission can be calculated using the fact that the dark channel value of the patch region in which the fog component is large is large.

는 구하고자 하는 전달량을 의미하며, 패치 영역(

)내부에서는 동일한 전달량을 갖는 것으로 가정한다.In order to estimate the amount of transmission, a min operation is performed on each of the R, G, and B channels and patch regions in both sides of Equation (3). In Equation (4) below,

Means the amount of transmission to be obtained, and the patch area (

) Are assumed to have the same amount of transmission.

)이 0이라는 것에 의해, 수학식 4의 우변에서 J가 포함된 항은 0으로 근사할 수 있다. 결과적으로, 전달량(

)은 아래 수학식 5와 같이 표현할 수 있다. 아래 수학식 5에서, I는 입력 이미지, A는 대기 산란광(global atmospheric light)을 의미한다.(2) and the dark channel value of the input image in which the fog component does not exist

) Is 0, the term including J in the right side of Equation (4) can be approximated to zero. As a result,

) Can be expressed by the following Equation (5). In Equation (5), I denotes an input image and A denotes a global atmospheric light.

)을 산출할 수 있다. 결과적으로 전달량(

)은 다크 채널이 가장 벗어난 영역, 즉 다크 채널의 화소값이 큰 영역으로부터 구할 수 있다.That is, referring to Equation (5), the input image I is divided by the atmospheric scattering light A and the dark channel value as shown in Equation (3)

) Can be calculated. As a result,

) Can be obtained from the region where the dark channel is the outermost, that is, from the region where the pixel value of the dark channel is large.

Based on the above description, the input image J from which the fog component is removed can be summarized by the following equation (6).

3 is a diagram for explaining an example of correcting the amount of transmission using a guided filter according to an embodiment.

Since the DCP algorithm described above in FIG. 2 is performed for each patch region of the input image, a staircase phenomenon may occur at the boundary portion of the patch region when restoring the fog removed image. That is, in the process of restoring the fog removed image, the accuracy may be lowered.

Conventionally, the soft matting algorithm is applied to prevent the degradation of the precision, but the soft matting algorithm performs the Laplacian transformation which requires a high computational complexity, so that the computation speed is slow.

Instead of the soft matting algorithm, the correction distance determining apparatus can correct the amount of transmission calculated using the guarded filter in real time, thereby preventing a reduction in accuracy.

A filtering method using a guided filter is a method of obtaining an output corresponding to a guidance by filtering an input image against a predetermined guidance.

The main premise of the guided filter is the local linearity between the guidance (I) and the output image (q) passing through the filter. q is, the center pixel is considered in the window w k _k linear transformation of I, it can be expressed as shown in Equation (7). Here, the window w _k denotes a patch in this bonded filter.

In Equation (7), a _k , b _k is a linear coefficient considered constant at w _k . To obtain the linear coefficient, the cost function as shown in Equation (8) below must be minimized.

In Equation (8),? Is a parameter for preventing a _{k from} becoming infinitely large, and p represents an input image. By solving Equation (8) by linear regression, the following Equation (9) can be obtained.

는 w_k에서의 픽셀의 개수,

는 w_k에서의 p의 평균값이다. 수학식 9를 수학식 7에 적용하여 가이디드 필터의 결과를 수학식 10과 같이 도출할 수 있다.In Equation (9), u _k and σ _k are the mean and variance of w _k extracted from I,

Is the number of pixels at w _k ,

Is the average value of p in the _k w. Equation (9) can be applied to Equation (7) to derive the result of the guided filter as shown in Equation (10).

Referring to FIG. 3, an intermediate image 320 may be generated by applying a DCP algorithm to the input image 310. Since the DCP algorithm is performed for each patch region of the input image 310, a stepping phenomenon may occur in the intermediate image 320 generated by applying the DCP algorithm. That is, a partial area of the intermediate image 320 may be crushed.

The final image 330 can be generated by applying the guiding filter described above to the intermediate image 320. [ A guiding filter may be applied to the intermediate image 320 to produce a sharper final image 330 that is stepped away. That is, the blurred area of the intermediate image 320 is restored by the guided filter, so that the final image 330 in which the detail has survived can be generated.

4 is a diagram for explaining an example of determining a fog boundary line in an input image according to an embodiment.

Referring to FIG. 4, in step 410, the corrective distance determining device may receive an input image from a camera for a photographing area photographed by the camera. The input image may contain fog components.

In one embodiment, the camera may be a CCTV, the correction distance determining apparatus may receive the streaming image from the CCTV, and extract the still cut from the streaming image to generate an input image.

In step 420, the corrective distance determining apparatus determines a dark channel value for each patch region of the input image by applying a DCP (Dark Channel Prior) algorithm to the input image, and calculates a transmission amount per patch region based on the dark channel value can do.

In step 430, the corrective distance determining apparatus determines whether the light source (e.g., sunlight, sunlight, headlight, etc.) included in the input image is misclassified as a fog component, Can be applied.

In one embodiment, when the size of the light source included in the input image is larger than the size of the patch area, the light source may be misclassified as a mist component, so that the correction distance determining device determines that the dark channel value If it is equal to or greater than the threshold value, the amount of transmission of the specific patch area can be adjusted.

For example, the correction distance determining apparatus can normalize the dark channel value of the input image to be between 0 and 1. When the dark channel value of the specific patch region in the input image is 0.7 or more, the correction distance determining apparatus can be set so that the specific patch region is determined as the fog component. At this time, the fixation distance determining apparatus sets a predetermined threshold value to 0.92, and when the dark channel value of the specific patch region is 0.92 or more, the amount of the specific patch region can be adjusted by classifying the specific patch region as a light source. That is, when the dark channel value of a specific patch region is 0.92 or more, the correction distance determining apparatus can prevent the light source from being misclassified as a mist component by adjusting the amount of transmission of a specific patch region. On the other hand, the dark channel value in which the specific patch region is determined as the fog component and the dark channel value classified in the light source are not limited thereto.

In step 440, the corrective distance determining apparatus can correct the calculated amount using the guided filter. Since the DCP algorithm used in the second step 420 is applied to each patch region of the input image, a staircase phenomenon may occur at the boundary portion of the patch region when the fog removed image is restored. That is, in the process of restoring the fog removed image, the accuracy may be lowered.

Instead of a soft matting algorithm that uses Laplacian transformations that require a high computational complexity, the Corrected Distance Correction Device can prevent the degradation of precision by correcting the calculated amount of transmission using a guided filter in real time.

In step 5, the corrective distance determining device may binarize the input image based on the corrected transmission amount.

The corrective distance determining apparatus applies a first color to a patch region whose corrected transmission amount is equal to or greater than a preset threshold value in a patch region of an input image and applies a second color to a patch region whose corrected transmission amount is less than a predetermined threshold value have. For example, the first color may be white and the second color may be black.

In one embodiment, when a user confirms an input image to which a guided filter is applied, the value of an area that can be determined as fog is set as a threshold value, and an input image to which a guided filter is applied is binarized can do.

In addition, the predetermined threshold value, which is a reference for binarizing the input image, can be determined based on the correlation with the correcting distance of the optical visual system.

In step 660, the corrective distance determining unit may determine a fog boundary line 461 in the binarized input image.

The corrective distance determining apparatus can determine the fog boundary line 461 based on the patch region-specific hue in the binarized image. In one embodiment, the corrective distance determining apparatus may determine a boundary line between the patch region of the first color and the patch region of the second color by the fog boundary line 461.

5A to 5D are diagrams for explaining an example of generating a distance map according to an embodiment.

Referring to Figs. 5A to 5C, the correction distance determining apparatus is capable of receiving a captured image 510a to 510c obtained by photographing moving means 520a to 520c for moving the photographing region of the camera from a camera that photographs an input image have. The shot images 510a to 510c may be still cuts of the streaming image received from the camera.

Referring to FIG. 5A, the correction distance determining apparatus can calculate the distance value between the moving means 520a and the camera based on the position of the moving means 520a and the position of the camera according to the GPS (Global Positioning System). The fixation distance determining apparatus can set the area 530a in which the moving means 520a is located in the shot image 510a as the calculated distance value.

More specifically, the corrective distance determining apparatus can compare the time at which the position of the moving means 520a used for calculating the distance value is measured and the time at which the photographing image 510a is photographed. If the measured time and the photographed time are the same, the area 530a in which the moving means 520a is located in the shot image 510a can be set to the calculated distance value.

Similarly, in Figs. 5B to 5C, the corrective distance determining apparatus can set the areas 530b to 530c in which the moving means 520b to 520c are located in the shot images 510b to 510c as the calculated distance values.

Referring to FIG. 5D, the fixation distance determination apparatus may generate a distance map 540 composed of distance values set in the regions 530a to 530c in which the moving means 520a to 520c are located in FIGS. 5A to 5C. The corrective distance determining apparatus can generate the distance map 540 composed of the distance value between the moving means and the camera, which is calculated according to the degree of trajectory of the moving means.

The correction distance determining apparatus may calculate a remaining distance value constituting the distance map 540 in addition to the distance value set in the distance map 540 by applying a homography technique to the distance map 540. [ The homography technique is applicable to a planar object as a theory that a certain transformation relation is established between projected corresponding points when one plane is projected on another plane. On the other hand, when the sky is included in the shot image, the correction distance determining apparatus can complete the distance map 540 by setting the sky portion of the shot image to the observation maximum value.

According to the present embodiment, a more refined distance map can be generated. In order to calculate the correcting distance of the photographed image without considering the camera distortion correction or the characteristics of the camera by using the cameras (for example, CCTVs) of various standards installed in the past, And a distance map can be completed by applying a homography technique.

6 is a view for explaining an example of correcting the distance value in consideration of the tide difference according to the embodiment.

The corrective distance determining apparatus can receive, from the CCTV 610, an input image including the moving means taken by the CCTV 610. [

Referring to FIG. 6, when the moving means is a ship, the GPS observation value for the moving means can be changed with time by the steep changing tide 630. For example, if the moving means is at the first height 620a according to the tilt angle 630, the distance between the CCTV 610 and the moving means becomes the first distance value 640a, The distance between the CCTV 620 and the mobile means may be the second distance value 640b.

When the moving means is at the second height 620b, the corrective distance determining apparatus can obtain the time at which the position of the moving means used to calculate the distance value is measured. The corrective distance determining apparatus can acquire the tide difference 630 at the obtained time based on the position of the moving means according to the GPS.

The corrective distance determining apparatus can correct the second distance value 640b to the first distance value 640a by using the tilt difference 630. [ First, the fixation distance determining apparatus can convert the installation height of the CCTV 610 and the height of the moving means into a datum level using the tilt gauge 630. That is, the fixation distance determining apparatus can convert the height of the moving means from the second height 620b to the first height 620a using the tilt angle difference 630. [ The corrective distance determining apparatus can calculate the distance value 640a between the moving means and the CCTV 610 at the converted first height 620a.

The corrective distance determining apparatus can generate a distance map composed of corrected distance values.

FIG. 7 is a diagram for explaining an example of performing a matrix multiplication operation between a fog boundary line and a distance map according to an embodiment.

Referring to FIG. 7, the distance map 710 is a matrix having a distance value as a component, and the fog boundary line map 720 is a matrix having components corresponding to fog components and non-fog components.

Specifically, the components of the distance map 710 are distance values from the camera to the actual area where the moving means is located in the shot image, the fog component is' 1 ', the fog component is' Quot; 0 ".

The fixation distance determining apparatus can perform an element-wise product operation between the distance map 710 and the fog boundary line map 720 to determine the distance value of the pixel corresponding to the fog boundary line.

8 is a diagram for explaining an example of determining a correction distance of an input image according to an embodiment.

The corrective distance determining apparatus can determine the corrective distance of the input image 800 based on the fog boundary line 840. In one embodiment, the correction distance determining apparatus may divide an input image into a plurality of ROIs, and then provide an average ROI and / or a minimum ROI for the ROI selected by the user.

Referring to FIG. 8, the correction distance determining apparatus may divide an input image 800 into three regions of interest 810 to 830. FIG. In one embodiment, if the user selects the first region of interest 810, the corrective distance of the input image 800 may be determined based on the fog boundary 840 in the first region of interest 810.

In one embodiment, the corrective distance determining device can determine the corrective distance by calculating an average of the corrective distance on the fog boundary line or by selecting the minimum corrective distance on the fog boundary line.

FIG. 9 is a block diagram illustrating a hardware configuration of a fixation distance determining apparatus according to an embodiment.

9, the correction distance determining apparatus 900 may include a controller 910, a communication unit 920, and a memory 930. Only the components related to the embodiment are shown in the corrective distance determination device 900 of Fig. Therefore, it will be understood by those skilled in the art that other general-purpose components other than the components shown in FIG. 9 may be further included.

The corrective distance determining apparatus 900 may be implemented by various types of devices such as a personal computer (PC), a server device, a mobile device, and an embedded device.

The control unit 910 may control a series of processes for determining the corrective distance of the image including the fog component described above with reference to Figs. The control unit 910 controls the overall functions for controlling the corrective distance determining apparatus 900. For example, the control unit 910 generally controls the corrective distance determining apparatus 900 by executing programs stored in the memory 930 in the corrective distance determining apparatus 900. [ The controller 910 may be implemented by a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), or the like, provided in the fixation distance determination apparatus 900, but is not limited thereto.

The communication unit 920 may include a local communication unit, a mobile communication unit, and a broadcast receiving unit. The communication unit 920 can receive an input image for a shooting region captured by an external camera from an external camera. The input image received by the communication unit 920 may include a moving means for moving the photographing area of the camera. In one embodiment, the communication unit 920 may receive the streaming video from the CCTV.

The memory 930 is a hardware for storing various data processed in the corrective distance determining apparatus 900. For example, the memory 930 stores an input image including a mist component received by the communication unit 920, And can store the processed data and the data to be processed in the determination apparatus 9000. [ In addition, the memory 930 may store applications, drivers, and the like to be driven by the corrective distance determining apparatus 900. [ The memory 930 may be a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a dynamic random access memory (DRAM) ROM, Blu-ray or other optical disk storage, a hard disk drive (HDD), a solid state drive (SSD), or a flash memory.

The embodiments may be implemented in the form of an application stored on a recording medium readable by an electronic device that stores instructions and data executable by the electronic device. The command may be stored in the form of program code, and when executed by the processor, may generate a predetermined program module to perform a predetermined operation. In addition, the instructions, when executed by a processor, may perform certain operations of the disclosed embodiments.

The embodiments may also be embodied in the form of a recording medium including instructions executable by a computer, such as program modules, being executed by a computer. Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically comprise any information delivery media, including computer readable instructions, data structures, other data of a modulated data signal such as program modules, or other transport mechanisms.

Also, in this specification, the term " part " may be a hardware component such as a processor or a circuit, and / or a software component executed by a hardware component such as a processor.

It will be understood by those skilled in the art that the foregoing description of the specification is for illustrative purposes only and that those skilled in the art will readily understand that various changes in form and details may be made therein without departing from the spirit and scope of the invention. It will be possible. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the foregoing detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be interpreted as including the claims.

Claims (10)

A method for determining a corrective distance of an image including a fog component,
Receiving an input image including a fog component;
Calculating a value of a dark channel for each patch region of the input image by applying a DCP (Dark Channel Prior) algorithm to the input image;
Calculating an amount of transmission per patch area based on the value of the dark channel;
Wherein when the value of the dark channel for a specific patch region in the input image is equal to or greater than a first threshold value to prevent the light source included in the input image from being misclassified as the mist component, step;
Correcting the calculated transmission amount using a guided filter;
Determining a fog boundary line in the input image based on the corrected transmission amount;
Receiving a shot image of a moving means for moving a shooting region of the camera from a camera that shoots the input image;
Calculating a distance value between the moving means and the camera based on the position of the moving means and the position of the camera according to a GPS (Global Positioning System);
Setting an area where the moving means is located in the shot image as the calculated distance value;
Generating a distance map composed of the set distance values; And
Determining a corrective distance of the input image based on the fog boundary line and the distance map;
/ RTI > delete The method according to claim 1,
Wherein the step of determining a fog boundary in the input image comprises:
Binarizing the input image based on the corrected transmission amount;
Lt; / RTI >
Wherein the step of binarizing the input image comprises:
Applying a first color to a patch region of the input image in which the corrected transmission amount is equal to or greater than a second threshold value and applying a second color to a patch region in which the corrected transmission amount is less than a second threshold value; And
Determining a boundary line between the patch region of the first color and the patch region of the second color as the fog boundary line;
/ RTI > delete The method according to claim 1,
Wherein the setting step comprises:
Comparing the time at which the position of the moving means used to calculate the distance value is measured and the time at which the taken image was taken; And
Setting an area where the moving means is located in the shot image as the calculated distance value when the measured time and the taken time are the same;
/ RTI > The method according to claim 1,
Wherein the generating the distance map comprises:
Obtaining a time at which the position of the moving means used to calculate the distance value is measured;
Obtaining a range of tide at the acquired time based on the position of the moving means according to the GPS;
Correcting the set distance value using the tidal elevation; And
Generating a distance map composed of the corrected distance values;
/ RTI > The method according to claim 1,
Wherein the generating the distance map comprises:
Calculating a remaining distance value constituting the distance map in addition to the distance value set in the distance map by applying a homography technique to the distance map; And
Generating the distance map using the set distance value and the remaining distance value;
/ RTI > The method according to claim 1,
Wherein the step of determining the corrective distance comprises:
Calculating an average of the correcting distance on the fog boundary line or determining the correcting distance by selecting a minimum correcting distance on the fog boundary line;
/ RTI > An apparatus for determining a corrective distance of an image including a fog component,
A memory in which at least one program is stored;
A communication unit for receiving an input image including a fog component; And
Calculating a value of a dark channel for each patch region of the input image by applying a DCP (Dark Channel Prior) algorithm to the input image by executing the at least one program stored in the memory, Wherein when a value of a dark channel for a specific patch region in the input image is equal to or greater than a first threshold value in order to prevent the light source included in the input image from being misclassified as the mist component, Adjusting a transmission amount of the specific patch region, correcting the calculated amount of transmission using a guided filter, determining a fog boundary line in the input image based on the corrected transmission amount,
A camera for photographing an input image, a camera for capturing an image of a moving means for moving a photographing area of the camera, and for receiving, based on the position of the moving means and the position of the camera according to a GPS (Global Positioning System) A distance value between the moving means and the camera is calculated, an area in which the moving means is located in the shot image is set as the calculated distance value, a distance map is formed with the set distance value,
A processor for determining a corrective distance of the input image based on the fog boundary line and the distance map;
. A computer-readable recording medium storing a program for executing the method of claim 1 in a computer.

Images