KR101995223B1 - System, module and method for detecting pedestrian, computer program - Google Patents

Abstract

A pedestrian detection system and modules, methods, and computer programs are disclosed. A pedestrian detection system according to an embodiment of the present invention includes a pedestrian detection unit for detecting a pedestrian by removing noise from the lattice data obtained by sensing a target area by a laser sensor; Dimensional coordinate system, transforming the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, mapping the Lidar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image generation unit for generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional rectangular coordinate system format as distance information; A boundary box setting unit for setting the size of the boundary box corresponding to the pedestrian using the distance information; And a bounding box display unit for displaying the bounding box of a size set to a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area by the camera sensor.

Description

TECHNICAL FIELD [0001] The present invention relates to a pedestrian detection system and a module, a method, and a computer program.

Embodiments of the present invention relate to a pedestrian detection technique in an image.

Recently, the development and research of intelligent automobiles capable of autonomous driving have been increasing. Especially, automatic emergency braking (AEBS), which is one of Advanced Driving Assistance System (ADAS) System) has been growing.

AEBS is a technology to prevent collision in the front of the car. It aims at detecting pedestrians in front of the vehicle accurately and quickly in various environments. Recently, the pedestrian detection is performed through the laser sensor. Since the Lada data acquired by the laser sensor is mapped to the grid map, the distance information is lost. Therefore, it is difficult to set the size of the bounding box for detecting the pedestrian through the laser sensor . Further, there is a problem that the noise is erroneously determined as a pedestrian in the process of detecting the pedestrian through the laser sensor.

Recently, research has been conducted to prevent the erroneous determination of a pedestrian as an object other than a pedestrian by appropriately setting the size of the bounding box.

The embodiments of the present invention set the size of the bounding box appropriately and prevent an object other than a pedestrian from being erroneously determined as a pedestrian.

According to an exemplary embodiment of the present invention, there is provided an image processing apparatus including: a sensor registration unit for registering LADID data acquired by sensing a target region by a laser sensor and an image captured by photographing the target region; A pedestrian detector for detecting a pedestrian by removing noise from the Lattice data; Dimensional coordinate system, transforming the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, mapping the Lidar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image generation unit for generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional rectangular coordinate system format as distance information; A boundary box setting unit for setting the size of the boundary box corresponding to the pedestrian using the distance information; And a boundary box display unit for displaying the boundary box of a size set to a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area by the camera sensor.

The pedestrian detecting unit may judge the data including the signal intensity information that is less than or equal to the set value in the rider data as noise and remove the noise.

The pedestrian detecting unit may judge the data including the distance information out of the set range in the rider data as noise and remove the noise.

The pedestrian detecting unit may determine data including azimuth angle information that deviates from the set range in the rider data as noise and remove the noise.

The boundary box setting unit may set the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapped image data.

According to another exemplary embodiment of the present invention, a pedestrian detection unit detects a pedestrian by removing noise from Lattice data obtained by sensing a target area by a laser sensor; Dimensional coordinate system, transforming the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, mapping the Lidar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image generation unit for generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional rectangular coordinate system format as distance information; A boundary box setting unit for setting the size of the boundary box corresponding to the pedestrian using the distance information; And a boundary box display unit for displaying the boundary box of a size set in a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area by the camera sensor.

The pedestrian detecting unit may judge the data including the signal intensity information that is less than or equal to the set value in the rider data as noise and remove the noise.

The pedestrian detecting unit may judge the data including the distance information out of the set range in the rider data as noise and remove the noise.

The pedestrian detecting unit may determine data including azimuth angle information that deviates from the set range in the rider data as noise and remove the noise.

The boundary box setting unit may set the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapped image data.

According to another exemplary embodiment of the present invention, there is provided a method of detecting a pedestrian by detecting noise in a ladder data obtained by sensing a target area by a laser sensor in a pedestrian detecting unit, thereby detecting a pedestrian; In the mapping image generation unit, the noise-removed Lidar data is converted into a two-dimensional Cartesian coordinate system, the Lidar data converted into the two-dimensional Cartesian coordinate system is mapped to a grid map to obtain mapping image data Generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional Cartesian coordinate system to the mapping image data as distance information; Setting a size of a bounding box corresponding to the pedestrian using the distance information in a bounding box setting unit; And displaying the boundary box of a size set in a portion corresponding to the pedestrian in a shot image obtained by photographing the object area by the camera sensor in a boundary box display section.

The step of detecting the pedestrian may include eliminating data including signal intensity information that is less than or equal to a predetermined value from the rider data by judging noise.

The step of detecting the pedestrian may remove data including distance information that deviates from the set range in the rider data by judging it as noise.

The step of detecting the pedestrian may remove data including azimuth angle information that deviates from the set range in the rider data by judging it as noise.

The step of setting the size of the bounding box may set the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapping image data.

According to another exemplary embodiment of the present invention there is provided a computer program stored in a non-transitory computer readable storage medium, the computer program comprising one or more instructions, The pedestrian detection system detects a pedestrian by removing noise from the ladder data acquired by sensing the object area by the laser sensor, and outputs the ladder data from which the noise has been removed to the grid Maps the image data to obtain mapping image data, adds the distance information of the noise-removed Lidar data to the mapping image data to generate a mapping image, and uses the distance information included in the mapping image data The width of the bounding box for the pedestrian And set a, and that the camera sensor is to indicate the size of the bounding box is set to the portion corresponding to the pedestrian in the captured image acquired by photographing the target area, a computer program.

By setting the size of the bounding box using the distance information of the Raider data, the size of the bounding box can be appropriately set.

By detecting the pedestrian by removing noise from the ladder data, it is possible to prevent the object other than the pedestrian from being erroneously determined as a pedestrian.

1 is a block diagram showing a detailed configuration of a pedestrian detection system according to an embodiment of the present invention;
Figure 2 shows a target for calibration in one embodiment of the present invention.
FIG. 3 is a diagram illustrating Lada data according to an embodiment of the present invention; FIG.
4 is a view showing a matching point according to an embodiment of the present invention;
5 is a view for explaining a process of extracting a matching point according to an embodiment of the present invention;
6 is a view for explaining a noise removing process according to an embodiment of the present invention;
7 is a diagram showing visualized Latticed Data and a mapping image according to an embodiment of the present invention;
8 is a view showing a photographed image in which a bounding box is displayed according to an embodiment of the present invention;
9 is a diagram for explaining a process of adding distance information to mapping image data according to an embodiment of the present invention
10 is a view for explaining a process of setting a size of a bounding box according to an embodiment of the present invention;
11 is a flowchart for explaining a pedestrian detection method according to an embodiment of the present invention.
12 is a block diagram illustrating and illustrating a computing environment including a computing device suitable for use in the exemplary embodiments.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals are used to designate identical or similar elements, and redundant description thereof will be omitted. The suffix "module" and " part "for the components used in the following description are given or mixed in consideration of ease of specification, and do not have their own meaning or role. In the following description of the embodiments of the present invention, a detailed description of related arts will be omitted when it is determined that the gist of the embodiments disclosed herein may be blurred. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are intended to provide further explanation of the invention as claimed. , ≪ / RTI > equivalents, and alternatives.

Terms including ordinals, such as first, second, etc., may be used to describe various elements, but the elements are not limited to these terms. The terms are used only for the purpose of distinguishing one component from another.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The singular expressions include plural expressions unless the context clearly dictates otherwise.

In the present application, the terms "comprises", "having", and the like are used to specify that a feature, a number, a step, an operation, an element, a component, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. It will be apparent to those skilled in the art that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.

1 is a block diagram showing a detailed configuration of a pedestrian detection system 100 according to an embodiment of the present invention. Referring to FIG. 1, a pedestrian detection system 100 according to an embodiment of the present invention includes a sensor matching unit 110 and a pedestrian detection module 200. The pedestrian detection module 200 according to an embodiment of the present invention includes a pedestrian detection unit 120, a mapping image generation unit 130, a bounding box setting unit 140, and a bounding box display unit 150.

Although the above configurations are described or shown as being separate configurations, they are either described or illustrated as separate configurations to describe each functional section separately. Therefore, it is apparent to those skilled in the art that the functions of the respective configurations can be performed in different parts, and that the functions of the respective configurations can be performed in the same part. In addition, each of the components may be connected in a wired or wireless form including an electrical connection to each other so as to transmit or receive data.

The laser sensor 10 emits a laser beam to a target area and generates three-dimensional spherical coordinate type LIDAR (Light Detection And Ranging) information including distance information from the laser sensor 10 to the target object, Can be obtained. The camera sensor 20 can acquire a photographed image in the form of a two-dimensional Cartesian coordinate system by photographing an object area.

The sensor matching unit 110 aligns the LR data obtained by sensing the target area by the laser sensor 10 and the photographed image obtained by photographing the target area by the camera sensor 20. [ Specifically, the sensor matching unit 110 extracts the feature points of the target region, converts the Lattice data acquired by the laser sensor 10 into a two-dimensional Cartesian coordinate system, And matched with the photographed image acquired by the camera sensor 20. For the specific image registration process, refer to the calibration operation below.

The calibration work will be discussed in detail. In order to set the size of the bounding box corresponding to the pedestrian using the distance information included in the Lada data acquired from the laser sensor 10 by the pedestrian detection system 100, A calibration operation is required to fuse the data acquired from the camera sensor 20. This is because the distance information included in the Ladder data acquired by the laser sensor 10 is composed of data in the form of a three-dimensional spherical coordinate system, while the captured image acquired by the camera sensor 20 is a two-dimensional rectangular coordinate system image .

In order to perform the calibration, it is necessary to convert the radar data of the three-dimensional spherical coordinate system into the two-dimensional rectangular coordinate system. The following equation 1 can be used to convert the radar data of the three-dimensional spherical coordinate system format into the two-dimensional orthogonal coordinate system. r is the distance covered by the data and θ is the azimuth angle. x _i represents the horizontal coordinate on the two-dimensional space, and y _i represents the vertical coordinate on the two-dimensional space.

[Equation 1]

x _i = r _i * cos θ _i , y _i = r _i * cos θ _i

In order to obtain the matching points in the two plane spaces, the corresponding feature points of the two sensors must be extracted. A detailed description thereof will be given later with reference to FIG.

The transformation matrix for the calibration operation should be obtained. Singular value decomposition (SVD) is used to find a transformation matrix according to an embodiment of the present invention. The last column of the U matrix (M * N) after the singular value decomposition of the A matrix (M * N) for the image photographed by the camera sensor 20 is used as a homography (transformation matrix). A person skilled in the art can easily carry out the calibration work by using the minutiae when the homography is obtained by using Equation (2). U and V are orthogonal matrixes, S is a diagonal matrix, H is homography, X _C is mapping image data, and X _L is Lattice data in which coordinate system transformation is performed.

&Quot; (2) "

SVD (A) = U * S * V

H = U (:, end)

X _C = H * X _L

The pedestrian detecting unit 120 detects the pedestrian by removing noise from the Raid data obtained by sensing the object area by the laser sensor 10. [ This is because the Lada data obtained by sensing the object area by the laser sensor 10 contains data on the noise outside the pedestrian. Specifically, the pedestrian detecting unit 120 may judge the data including the signal intensity information that is equal to or less than a set value (for example, noise) in the Lada data as noise and remove the noise. In addition, the pedestrian detecting unit 120 may judge and remove data including distance information that deviates from the set range (for example, 1 mm to 40 m) from the Lada data as noise. In addition, the pedestrian detecting unit 120 may judge the data including the azimuth angle information that deviates from a predetermined range (for example, not a road) from the Raid data as noise and remove the noise.

The mapping image generation unit 130 converts the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, maps the Ladar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image is generated by adding the horizontal coordinates of the Lidar data converted into the two-dimensional Cartesian coordinate system to the mapping image data as distance information. The mapping image data refers to the Lada data mapped to the grid map, and the mapping image refers to the image generated by adding the distance information to the mapping image data. Hereinafter, a process of adding distance information to mapping image data will be described in detail.

The lidar data is lost in the process of mapping to the grid map. Therefore, the mapping image data does not include distance information. Thus, it should have a vertical coordinate of the distance information by including the x _i on the two-dimensional space as shown in Equation 3 below on the mapping image data. When the distance information is added to the mapping image data, the distance information can be used in the mapping image. LIDAR _cartesian is _Lada data, H is homography, X _C is mapping image data, and X _D is mapping image data including distance information.

&Quot; (3) "

LIDAR _cartesian = (x _i , y _i )

X _C = H * LIDAR _cartesian = (x _j , y _j )

X _D = (x _j , y _j , x _i )

(i, j = 1, 2, 3, ..., n)

The bounding box setting unit 140 sets the size of the bounding box for the pedestrian using the distance information included in the mapped image data. Specifically, the boundary box setting unit 140 can set the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapping image data.

The bounding box display unit 150 displays a bounding box of a size set in a portion corresponding to the pedestrian in the photographed image obtained by the camera sensor 20 photographing the target area.

2 is a diagram illustrating a target for calibration in one embodiment of the present invention.

In order to fuse two sensors with different characteristics, such as a laser sensor and a camera sensor, a calibration operation is required. In performing calibration according to an embodiment of the present invention, a 'c' shape and a 'v' shape target may be used as shown in FIG. Target 1 is a 'c' target, and Target 2 is a 'v' target. Feature points can be easily extracted by using a 'c' target or a 'v' target. However, it is apparent to those skilled in the art that the shape of the target is not limited to this, and other shapes of targets may be used in performing the calibration operation.

3 is a diagram illustrating Lattice Data obtained from a laser sensor according to an embodiment of the present invention.

As shown in the left drawing of FIG. 3, the distance information included in the Lattice Data acquired by the laser sensor is composed of three-dimensional spherical coordinate system data (azimuth angle and distance DISTANCE). On the other hand, the Raidata data includes signal strength (SIGNAL_STRANGTH) information. In order to convert the radar data of the three-dimensional spherical coordinate system into the two-dimensional orthogonal coordinate system, Equation 1 may be used as described above. As shown in the right drawing of FIG. 3, when the coordinate system conversion is performed, the LAD data can be represented visually. The x-axis represents x _i and the y-axis represents y _i .

4 is a view illustrating matching points according to an embodiment of the present invention.

Feature points can be both end points and corner points. The feature points (P1 - P4, P5 - P7) can be used as matching points as corresponding feature points of two sensors for extracting a matching point in two plane spaces.

5 is a view for explaining a process of extracting a matching point according to an embodiment of the present invention.

5, the coordinate values (P1, P2, P3, and P4) of the corresponding point are feature points in the image captured by the camera sensor. As shown in the right drawing of FIG. 5, the characteristic points of the laser sensor include the end points P1 and P4 after the LIDAR data has been segmented and the intersection points P2 of the straight lines after the line- , P3).

6 is a diagram for explaining a noise removal process according to an embodiment of the present invention.

The pedestrian detection unit detects the pedestrian by removing noise from the Lada data acquired by sensing the object area by the laser sensor. Specifically, the pedestrian detecting unit may judge the data including the signal intensity information that is equal to or less than a set value (for example, noise) in the Lada data as noise and remove the noise. In addition, the pedestrian detecting unit 120 may judge and remove data including distance information that deviates from the set range (for example, 1 mm to 40 m) from the Lada data as noise. In addition, the pedestrian detecting unit can determine and remove data including azimuth information that deviates from a predetermined range (for example, a road other than a road) in the Raidata data as noise. The upper diagram in Fig. 6 is a diagram before noise is removed, and the lower diagram in Fig. 6 is a diagram after noise is removed. The portion indicated by '+' in FIG. 6 indicates a portion detected as a pedestrian, and a comparison between the box regions shown in FIG. 6 shows that the portion detected as a pedestrian in addition to the actual pedestrian is removed.

7 is a diagram illustrating visualized Lidar data and a mapping image according to an embodiment of the present invention.

If the Lidar data converted into the two-dimensional Cartesian coordinate system is mapped to the grid map, a mapping image can be generated. The mapping image data can be treated as an image having pixels. 7, the distance information is lost in the process of mapping the grid data to the grid map. As shown in the lower diagram of FIG. 7, the mapping image data includes distance information I never do that.

FIG. 8 is a view showing a photographed image showing a bounding box according to an embodiment of the present invention.

FIG. 8 shows a problem in setting the size of the bounding box using the mapping image data to which the distance information is not added. The size of the boundary box in the case where the pedestrian exists at a distance of 5 m from the laser sensor (upper drawing) and the case where the pedestrian exists at a distance of 10 m (lower drawing) is the same.

9 is a diagram for explaining a process of adding distance information to mapping image data according to an embodiment of the present invention.

9, distance information (Dist.) Is added to the mapping image data (x, y, Width, Height) having no distance information, as shown in the lower drawing of Fig. 9 . You may have a vertical coordinate information, by including the distance x _i on the two-dimensional space as shown in the equation (3) above in the mapping image data. If the distance information is added to the mapping image data, the distance information can be used in the mapping image.

FIG. 10 is a view for explaining a process of setting a size of a bounding box according to an embodiment of the present invention.

The bounding box setting unit sets the size of the bounding box for the pedestrian using the distance information included in the mapping image data. Specifically, the boundary box setting unit may set the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapping image data. As shown in FIG. 10, the size of the bounding box can be set to be small in the order of 5m targets, 10m targets, and 15m targets.

11 is a flowchart illustrating a pedestrian detection method according to an embodiment of the present invention. In the illustrated flow chart, the method is described as being divided into a plurality of steps, but at least some of the steps may be performed in reverse order, combined with other steps, performed together, omitted, divided into detailed steps, One or more steps may be added and performed.

First, a pedestrian is detected (S202). Specifically, the pedestrian detecting section detects no pedestrians by removing noise from the Lattice data.

Next, a mapping image is generated (S204). Specifically, the mapping image generation unit converts the noise-removed Lattice data into a two-dimensional Cartesian coordinate system, maps the Lattice data transformed into the two-dimensional Cartesian coordinate system to the grid map to obtain mapping image data, A mapping image is generated by adding the horizontal coordinates of the Lidar data converted into the two-dimensional Cartesian coordinate system to the mapping image data as distance information.

Next, the size of the bounding box is set (S206). Specifically, the boundary box setting unit sets the size of the bounding box corresponding to the pedestrian using the distance information.

Finally, a bounding box is displayed (S208). Specifically, in the boundary box display section, a boundary box of a size set in a portion corresponding to a pedestrian is displayed on the photographed image acquired by the camera sensor by photographing the target area.

12 is a block diagram illustrating and illustrating a computing environment 10 that includes a computing device suitable for use in the exemplary embodiments.

In the illustrated embodiment, each of the components may have different functions and capabilities than those described below, and may include additional components in addition to those described below.

The illustrated computing environment 10 includes a computing device 12. In one embodiment, the computing device 12 may be a pedestrian detection system 100. The computing device 12 includes at least one processor 14, a computer readable storage medium 16, The processor 14 may cause the computing device 12 to operate in accordance with the exemplary embodiment discussed above. For example, processor 14 may execute one or more programs stored on computer readable storage medium 16. The one or more programs may include one or more computer-executable instructions, which when executed by the processor 14 cause the computing device 12 to perform operations in accordance with the illustrative embodiment .

The computer-readable storage medium 16 is configured to store computer-executable instructions or program code, program data, and / or other suitable forms of information. The program 20 stored in the computer-readable storage medium 16 includes a set of instructions executable by the processor 14. In one embodiment, the computer-readable storage medium 16 may be any type of storage medium such as a memory (volatile memory such as random access memory, non-volatile memory, or any suitable combination thereof), one or more magnetic disk storage devices, Memory devices, or any other form of storage medium that can be accessed by the computing device 12 and store the desired information, or any suitable combination thereof.

Communication bus 18 interconnects various other components of computing device 12, including processor 14, computer readable storage medium 16.

The computing device 12 may also include one or more input / output interfaces 22 and one or more network communication interfaces 26 that provide an interface for one or more input / output devices 24. The input / output interface 22 and the network communication interface 26 are connected to the communication bus 18. The input / output device 24 may be connected to other components of the computing device 12 via the input / output interface 22. The exemplary input and output device 24 may be any type of device, such as a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touch pad or touch screen), a voice or sound input device, An input device, and / or an output device such as a display device, a printer, a speaker, and / or a network card. The exemplary input and output device 24 may be included within the computing device 12 as a component of the computing device 12 and may be coupled to the computing device 12 as a separate device distinct from the computing device 12 It is possible.

The foregoing detailed description should not be construed in any way as being restrictive and should be considered illustrative. The scope of the present invention should be determined by rational interpretation of the appended claims, and all changes within the scope of equivalents of the present invention are included in the scope of the present invention.

10: Laser sensor
20: Camera sensor
100: Pedestrian detection system
110: sensor matching portion
120: Pedestrian detection unit
130:
140: boundary box setting unit
150:

Claims (16)

A sensor matching unit for matching the Lada data obtained by sensing the target region with the laser sensor and the photographed image obtained by photographing the target region by the camera sensor;
A pedestrian detector for detecting a pedestrian by removing noise from the Lattice data;
Dimensional coordinate system, transforming the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, mapping the Lidar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image generation unit for generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional rectangular coordinate system format as distance information;
A boundary box setting unit for setting the size of the boundary box corresponding to the pedestrian using the distance information; And
And a boundary box display unit for displaying the boundary box of a size set in a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area by the camera sensor.
The method according to claim 1,
Wherein the pedestrian detection unit judges, as noise, data including signal intensity information that is less than or equal to a set value in the Läda data, and removes the noise.
The method according to claim 1,
Wherein the pedestrian detecting unit judges, as noise, data including distance information that deviates from a predetermined range in the Lidar data and removes it.
The method according to claim 1,
Wherein the pedestrian detection unit judges, as noise, data including azimuth angle information that deviates from a predetermined range in the Lidar data and removes it.
The method according to claim 1,
Wherein the boundary box setting unit sets the size of the bounding box to be smaller as the distance to the pedestrian increases as a result of the distance information included in the mapping image data.
A pedestrian detector for detecting a pedestrian by removing noise from the Lattice data;
Dimensional coordinate system, transforming the noise-removed Lidar data into a two-dimensional Cartesian coordinate system, mapping the Lidar data converted into the two-dimensional Cartesian coordinate system to a grid map to obtain mapping image data, A mapping image generation unit for generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional rectangular coordinate system format as distance information;
A boundary box setting unit for setting the size of the boundary box corresponding to the pedestrian using the distance information; And
And a bounding box display unit for displaying the bounding box of a size set to a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area by the camera sensor.
The method according to claim 6,
Wherein the pedestrian detecting unit judges, as noise, data including signal intensity information that is less than or equal to a set value in the LR data to remove the pedestrian.
The method according to claim 6,
Wherein the pedestrian detecting unit judges, as noise, data including distance information that deviates from a predetermined range in the Läda data, and removes it.
The method according to claim 6,
Wherein the pedestrian detecting unit judges, as noise, data including azimuth angle information that deviates from the set range in the Lidar data and removes it.
The method according to claim 6,
Wherein the bounding box setting unit sets the size of the bounding box to be smaller as the distance to the pedestrian increases, based on the distance information included in the mapping image data.
Detecting a pedestrian by removing noise from the ladder data at a pedestrian detecting unit;
In the mapping image generation unit, the noise-removed Lidar data is converted into a two-dimensional Cartesian coordinate system, the Lidar data converted into the two-dimensional Cartesian coordinate system is mapped to a grid map to obtain mapping image data Generating a mapping image by adding horizontal coordinates of the Lidar data converted into the two-dimensional Cartesian coordinate system to the mapping image data as distance information;
Setting a size of a bounding box corresponding to the pedestrian using the distance information; And
And displaying the boundary box of the size set in a portion corresponding to the pedestrian in the shot image obtained by photographing the target area by the camera sensor in the boundary box display section.
12. The method of claim 11,
Wherein the step of detecting the pedestrian comprises the step of judging, as noise, data including signal intensity information that is equal to or lower than a value set in the Lattice data.
12. The method of claim 11,
Wherein the step of detecting the pedestrian judges that data including distance information that deviates from a predetermined range in the Lattice data is noise and removes the pedestrian.
12. The method of claim 11,
Wherein the step of detecting the pedestrian judges that data including azimuth information that deviates from a predetermined range in the Lidar data is noise and removes the pedestrian.
12. The method of claim 11,
Wherein the step of setting the size of the bounding box sets the size of the bounding box to be smaller as the distance to the pedestrian increases as a result of the distance information included in the mapped image data.
A computer program stored on a non-transitory computer readable storage medium,
The computer program comprising one or more instructions, wherein the instructions, when executed by a pedestrian detection system having one or more processors, cause the pedestrian detection system to:
The laser sensor detects the pedestrian by removing noise from the Lada data obtained by sensing the target area,
Generating a mapping image by adding distance information of the noise-removed Lidar data to the mapping image data by mapping the noise-removed Lidar data to a grid map to obtain mapping image data,
Setting a size of a bounding box for the pedestrian using the distance information included in the mapped image data, and
And the camera sensor displays a boundary box of a size set in a portion corresponding to the pedestrian in the photographed image obtained by photographing the target area.

Images