KR102423334B1 - Taillight detection method and apparatus - Google Patents

Abstract

A tail light detection method and apparatus are disclosed. The tail light detection method includes the steps of: (a) determining a tail light ROI by using a tail light color threshold range in an input image, wherein the tail light color threshold range is a threshold value range based on an HSV color space; (b) allocating a label to each point included in the region of interest for the tail light, when there is a preceding neighboring point, allocating a label using the label of the neighboring neighboring point; and (c) determining points having the same label among the assigned labels as the same tail light candidate area.

Description

Taillight detection method and apparatus

The present invention relates to a method and apparatus for detecting a tail light area of a vehicle.

In the VOCC (Vehicular Optical Camera Communication) system, data is transmitted using the vehicle's tail light and data is received through the high-speed camera inside the vehicle. At this time, in order to receive the transmitted data, it is necessary to detect the tail light area existing in the image.

In the conventional case, a certain threshold value is applied based on color and brightness control to detect the tail light area.

Republic of Korea Patent Publication No. 10-2013-0045664 (2013.05.06.)

The present invention is to provide a method and apparatus for detecting a tail light.

An object of the present invention is to provide a method and apparatus for detecting a tail light capable of determining a tail light area using a clustering technique.

According to one aspect of the present invention, a method for detecting a tail light is provided.

According to an embodiment of the present invention, (a) determining a region of interest for a tail light by using a range of a color threshold for a tail light in an input image; the color threshold range for a tail light is a threshold range based on an HSV color space; (b) allocating a label to each point included in the region of interest for the tail light, when there is a preceding neighboring point, allocating a label using the label of the neighboring neighboring point; and (c) determining points having the same label among the assigned labels as the same tail light candidate area.

The step (b) includes: (b1) allocating a label to a first point of the region of interest in the tail lamp; (b2) allocating a label for the second point as the smallest label among the labels of the preceding neighboring point when there is the preceding adjacent neighbor point for the second point, wherein (b2) includes: It may be sequentially and repeatedly performed until label assignment to all points in the region of interest of the tail light is completed.

In the step (b2), when all of the neighboring points preceding the second point are included in the background area, the method may include increasing and allocating a label.

The preceding adjacent neighboring points are an upper left point, a left point, an upper point, and a left and right point adjacent to each point.

Prior to step (b), the method may further include removing discontinuities and small holes in the region of interest of the tail lamp using morphological proximity transformation.

After step (c), determining the maximum and minimum values of each of the horizontal and vertical points based on the determined candidate tail light area, and determining the tail light area using the maximum and minimum values of each of the horizontal and vertical points may further include.

Through the verification process for the determined tail light area, if the area is greater than the standard value, the ratio of the width and height of the tail light is less than the standard, or the ratio of the height of the input image to the height of the tail light area is less than the standard value, it is determined that the tail light is not and removed It may include further steps.

According to another aspect of the present invention, a tail light detection device is provided.

According to an embodiment of the present invention, there is provided an apparatus for detecting a tail light, comprising: a memory for storing at least one instruction; a processor for executing an instruction stored in the memory, wherein the instruction comprises: (a) determining a region of interest for a tail light by using a range of a color threshold for a tail light in an input image; based threshold range; (b) allocating a label to each point included in the region of interest for the tail light, when there is a preceding neighboring point, allocating a label using the label of the neighboring neighboring point; and (c) determining points having the same label among the assigned labels as the same tail light candidate area may be provided.

By providing a method and apparatus for detecting a tail light according to an embodiment of the present invention, there is an advantage in that a tail light area can be accurately determined by applying a clustering technique.

1 is a flowchart illustrating a method for detecting a tail light area according to an embodiment of the present invention.
2 is a diagram illustrating a tail light color threshold range according to an embodiment of the present invention;
Figure 3 is a view illustrating a tail light region of interest according to an embodiment of the present invention.
Figure 4 is a view for explaining the enhancement of the region of interest in the tail light according to an embodiment of the present invention.
5 is a view for explaining a labeling method according to an embodiment of the present invention.
6 is a diagram illustrating a pseudo code for a labeling method according to an embodiment of the present invention.
7 is a diagram illustrating a minimum bounding box for a tail light area according to an embodiment of the present invention;
8 is a view showing a tail light detection result according to an embodiment of the present invention.
9 is a view showing a result of removing noise through tail light verification according to an embodiment of the present invention;
10 is a block diagram schematically illustrating an internal configuration of a tail light detection device according to an embodiment of the present invention.

As used herein, the singular expression includes the plural expression unless the context clearly dictates otherwise. In this specification, terms such as “consisting of” or “comprising” should not be construed as necessarily including all of the various components or various steps described in the specification, some of which components or some steps are It should be construed that it may not include, or may further include additional components or steps. In addition, terms such as "...unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software, or a combination of hardware and software. .

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

1 is a flowchart illustrating a tail light area detection method according to an embodiment of the present invention, FIG. 2 is a view illustrating a tail light color threshold range according to an embodiment of the present invention, and FIG. 3 is an embodiment of the present invention It is a view illustrating a region of interest in a tail light according to an example, and FIG. 4 is a diagram illustrating enhancement of a region of interest in a tail light according to an embodiment of the present invention, and FIG. 5 is a labeling method according to an embodiment of the present invention. It is a diagram for explanation, and FIG. 6 is a diagram illustrating a pseudo code for a labeling method according to an embodiment of the present invention, and FIG. 7 is a minimum bounding box for the tail light area according to an embodiment of the present invention. 8 is a view showing a tail light detection result according to an embodiment of the present invention, and FIG. 9 is a view showing a result of removing noise through tail light verification according to an embodiment of the present invention.

In operation 110 , the tail light detection apparatus 100 detects a tail light ROI by using the tail light color threshold value range from the input image. The tail light detection apparatus 100 converts the input image from the RGB color space to the HSV color space model. Since this is obvious to those skilled in the art, a detailed description thereof will be omitted.

With respect to the input image converted into the HSV color space model, the region of interest of the tail lamp may be determined using the color threshold range of the tail lamp. Here, the tail light color threshold range is as shown in FIG. 2 . 2 illustrates an example, but is not necessarily limited thereto.

As shown in FIG. 2 , the HSV color space may include hue, saturation, and value. Since this is obvious to those skilled in the art, a detailed description thereof will be omitted.

3 illustrates an example of determining a region of interest for a tail lamp by applying a color threshold range for a tail lamp to an input image.

In step 115, the tail light detection apparatus 100 removes discontinuities and small holes in the tail light ROI through morphological proximity transformation.

When the ROI of the tail lamp is derived after converting the input image into the HSV color space model, it can be seen that a missing region exists in the ROI of the tail lamp as shown in FIG. 4A . Accordingly, the apparatus 100 for detecting a tail light according to an embodiment of the present invention merges a discontinuous or hole region existing in the region of interest in the tail lamp with an adjacent region through morphological proximity transformation, and fills in points in the surrounding low-color region to the region of interest in the tail lamp. can improve

Figure 4 (b) illustrates the result of removing the discontinuity and small hole in the region of interest of the tail light.

In operation 120 , the tail light detection apparatus 100 clusters the tail light candidate area by allocating a label to each point (pixel) included in the tail light ROI.

This will be described in more detail.

Since the first point included in the tail light ROI does not have any preceding adjacent neighboring points, the first label is assigned.

From the second point on, the label may be assigned using the label of the preceding neighboring point.

For example, when a neighboring point preceding the second point exists, a label for the corresponding second point may be assigned as the smallest label among labels assigned to the preceding neighboring point.

However, if all neighboring points preceding the second point are backgrounds, in this case, since there is no preceding assigned label, the second point may be assigned a label with a new label index.

It will be described in more detail with reference to FIG. 5 .

It is assumed that 510 of FIG. 5 is a current point. In an embodiment of the present invention, the preceding neighboring point may be an upper left point, a left point, an upper point, and an upper right point based on the current point.

Accordingly, neighboring points preceding the current point may be 520a, 520b, 520c, and 520d.

As shown in FIG. 5 , for the preceding neighboring points 520a , 520b , 520c , 520d of the current point 510 , the labels are “1”, “3”, “1”, “2”, respectively. Assume that it is assigned.

In an embodiment of the present invention, since the minimum label among the labels assigned to the neighboring points 520a, 520b, 520c, and 520d preceding the current point is “1”, the label of the current point 210 is assigned as “1”. can do.

However, if all neighboring points preceding the current point are background regions, a label may be assigned to the current point by increasing the label index.

If it is assumed that the maximum value among the label indices allocated so far is "4", and it is assumed that all neighboring points preceding the current point are backgrounds, the current point may be assigned a label of "5".

6 illustrates a pseudocode for this.

In this way, each point included in the region of interest for the tail lamp may be labeled and clustered using label information of a preceding neighboring point, and the region may be classified according to the label.

For example, the above-described current point p is allocated to the q region, and the remaining regions of each preceding neighbor of p may also be merged into the q region. If not, the current point p may be assigned as a new labeled area.

In operation 125 , when label assignment to each point of the tail light ROI is completed, the tail light detection apparatus 100 determines points having the same label among the assigned labels as the same tail light candidate area.

In operation 130 , the tail light detection apparatus 100 determines the tail light area based on the coordinates of each point of the tail light candidate area.

For example, a maximum value and a minimum value of the horizontal and vertical points are respectively determined based on the determined tail light candidate area.

For example, it is assumed that the list of tail light candidate areas is T. In this case, the maximum and minimum values of the horizontal and vertical points of the tail light candidate area may be derived using Equation 1.

는 수평 포인트의 좌표값들 중 최소값(즉, X축 좌표값들 중 최소값)을 나타내고,

는 수평 포인트의 좌표값들 중 최대값(예를 들어, X축 좌표값들 중 최대값)을 나타내며,

는 수직 포인트의 좌표값들 중 최소값(즉, Y축 좌표값들 중 최소값)을 나타내며,

는 수직 포인트의 좌표값들 중 최대값(즉, Y축 좌표값들 중 최대값)을 나타낸다. Here, T represents a list of points corresponding to the tail light candidate area,

represents the minimum value among the coordinate values of the horizontal point (that is, the minimum value among the X-axis coordinate values),

represents the maximum value among the coordinate values of the horizontal point (for example, the maximum value among the X-axis coordinate values),

represents the minimum value among the coordinate values of the vertical point (that is, the minimum value among the Y-axis coordinate values),

denotes a maximum value among coordinate values of a vertical point (ie, a maximum value among Y-axis coordinate values).

In addition, i represents the index of each point, and min and max represent functions for deriving the minimum and maximum values, respectively.

The tail light detection apparatus 100 may derive the rear light pixel coordinates by using the minimum and maximum values of the derived horizontal and vertical points.

For example, the tail light pixel coordinates may be derived using Equation (2).

The tail light detection apparatus 100 determines the tail light area including the tail light pixel coordinates.

For example, the tail light detection apparatus 100 may determine the tail light area (minimum boundary area) by using the minimum and maximum values of the horizontal and vertical points. That is, the tail light detection apparatus 100 may determine the tail light area by using the minimum coordinates of horizontal and vertical points including the rear light pixel coordinates and the maximum coordinates of the horizontal and vertical points.

,

)를 좌상단 코너로 하며, (

,

)를 우하단 코너로 하는 후미등 영역을 결정할 수 있다. For example, (

,

) as the upper left corner, (

,

) as the lower right corner, it is possible to determine the tail light area.

Also, the tail light detection apparatus 100 may derive the width and height of the tail light area using Equation 3 below.

는 후미등 영역의 폭을 나타내고,

는 후미등 영역의 높이를 나타낸다. here,

represents the width of the tail light area,

indicates the height of the taillight area.

7 the derived tail light area (ie the minimum bounding box) is illustrated.

8 shows the tail light area detection result. Fig. 8 (a) shows the input image, and Fig. 8 (b) is the tail light region detection result.

In step 135, the tail light detection apparatus 100 verifies the tail light area and removes noise.

After determining the tail light area, the tail light detection apparatus 100 removes the tail light verification when the tail light is not the tail light.

For example, the tail light detection apparatus 100 verifies the tail light area, and when the area is less than or equal to the first reference value or greater than or equal to the second reference value, the tail light detection apparatus 100 may determine that the area is not a tail light such as a brake lamp or a signal, and remove it.

This will be described in more detail.

)을 기초로 검출된 후미등 영역의 폭과 높이 사이의 비율이 제1 기준값 이하인 경우 브레이크 램프로 판단하여 제거할 수 있다. The tail light detection device 100 sets a first reference value (

), when the ratio between the width and the height of the tail light area detected based on the first reference value is less than or equal to the first reference value, it may be determined as a brake lamp and removed.

That is, when the tail light detection apparatus 100 satisfies Equation 4, it may be determined as a brake lamp and removed.

)을 이용하여 검출된 후미등 영역 중 면적이 너무 작은 영역은 노이즈나 반사된 영역으로 판단하여 제거할 수 있다. In addition, the tail light detection apparatus 100 sets the second reference value (

), an area with too small an area among the tail light areas detected using ) can be determined as noise or a reflected area and removed.

That is, the tail light detecting apparatus 100 may remove the tail light area satisfying Equation 5.

) 이상인 경우 신호등, 건물 불빛으로 판단하여 제거할 수 있다. 즉, 후미등 검출 장치(100)는 수학식 6을 만족하는 후미등 영역을 제거할 수 있다. In addition, the tail light detection apparatus 100 has a Y-axis coordinate value based on the height of the input image as a third reference value (

), it can be removed by judging it as a traffic light or building light. That is, the tail light detection apparatus 100 may remove the tail light area satisfying Equation (6).

Here, H represents the height of the input image.

9 is a view showing the result after the tail light verification. 9 (b) is a result of removing the area other than the tail light by using the first reference value to the third reference value through the tail light verification process.

10 is a block diagram schematically illustrating an internal configuration of a tail light detection apparatus according to an embodiment of the present invention.

Referring to FIG. 10 , the tail light detection apparatus 100 according to an embodiment of the present invention includes a memory 1010 and a processor 1015 .

The memory 1010 stores at least one instruction.

The processor 1015 is a means for controlling the memory 1010 . In addition, the instructions executed by the processor 1015 may perform each step for tail light detection as described with reference to FIGS. 1 to 9 . Since this is the same as that described above, the overlapping description will be omitted. In addition, the tail light detection apparatus 100 may further include a configuration such as a communication unit, but since it is irrelevant to the main point of the present invention, a description of these configurations will be omitted. It should be understood that the tail light detection apparatus 100 according to an embodiment of the present invention includes general components of electronic devices included in a computer, a notebook computer, an image processing device, a camera, and a vehicle.

The apparatus and method according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the computer readable medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floppy disks. - Includes magneto-optical media and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

Up to now, the present invention has been looked at focusing on the embodiments thereof. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention may be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

100: tail light detection device
1010: memory
1015: Processor

Claims (10)

In the tail light detection method of the tail light detection device,
(a) determining a region of interest for a tail lamp by using a range of a threshold value of a color of a tail light in an input image;
(b) allocating a label to each point included in the region of interest for the tail light, allocating a label using a label of the neighboring point when there is a preceding neighboring point;
(c) determining points having the same label among the assigned labels as the same tail light candidate area;
(d) determining a maximum value and a minimum value of each of horizontal and vertical points based on the determined candidate tail light area, and determining a tail light area using the maximum and minimum values of each of the horizontal and vertical points; and
(e) if the ratio of the height of the input image to the height of the tail light area is less than or equal to the reference value through the verification process for the tail light area, it is determined that it is not a tail light and removing it;
Step (b) is,
(b1) allocating a label to a first point of the region of interest in the tail light;
(b2) If there is the preceding neighboring point for the second point, the label for the second point is assigned as the smallest label among the labels of the preceding neighboring point, and the neighboring neighboring point that precedes the second point When all the points are included in the background area, including the step of allocating by increasing the label,
The method (b2) is a tail light detection method, characterized in that it is sequentially and repeatedly performed until the label assignment to all points in the region of interest of the tail light is completed.
delete delete The method of claim 1,
The preceding adjacent neighboring points are an upper left point, a left point, an upper point, and a left and right point adjacent to each point.
The method of claim 1,
Before step (b),
and removing discontinuities and small holes in the tail light region of interest using morphological proximity transformation.
delete The method of claim 1,
Through the verification process for the determined tail light area, if the area is greater than the standard value, the ratio of the width and height of the tail light is less than the standard, or the ratio of the height of the input image to the height of the tail light area is less than the standard value, it is determined that the tail light is not and removed A tail light detection method further comprising the step.
A computer-readable recording medium recording program code for performing the method according to claim 1 .
In the tail light detection device,
a memory storing at least one instruction;
A processor for executing instructions stored in the memory,
The command is
(a) determining a region of interest for a tail lamp by using a range of a threshold value of a color of a tail light in an input image;
(b) allocating a label to each point included in the region of interest for the tail light, allocating a label using a label of the neighboring point when there is a preceding neighboring point;
(c) determining points having the same label among the assigned labels as the same tail light candidate area;
(d) determining a maximum value and a minimum value of each of horizontal and vertical points based on the determined candidate tail light area, and determining a tail light area using the maximum and minimum values of each of the horizontal and vertical points; and
(e) if the ratio of the height of the input image to the height of the tail light area is less than the reference value through the verification process for the tail light area, it is determined that it is not a tail light and removed,
Step (b) is,
(b1) allocating a label to a first point of the region of interest in the tail light; and
(b2) performing the step of allocating a label for the second point as the smallest label among the labels of the preceding neighboring point when there is the preceding neighboring neighboring point with respect to the second point,
The tail light detection apparatus, characterized in that (b2) is sequentially and repeatedly performed until label assignment to all points in the region of interest of the tail light is completed.
delete

Images