WO2018212560A1

WO2018212560A1 - Apparatus and system for measuring location of vehicle

Info

Publication number: WO2018212560A1
Application number: PCT/KR2018/005557
Authority: WO
Inventors: 기석철; 이신재
Original assignee: 충북대학교 산학협력단
Priority date: 2017-05-18
Filing date: 2018-05-15
Publication date: 2018-11-22
Also published as: KR101947685B1; KR20180126879A

Abstract

Disclosed are an apparatus and a system for measuring a location of a vehicle, and the present invention provides a lane recognition apparatus and method for determining whether a vehicle has crossed a lane, through the measuring apparatus and system and using an around view monitor (AVM) camera, and delivering a determination result to the outside. Specifically, a system for measuring a location of a vehicle in a space partitioned into multiple courses may comprise: a recognition tag which is disposed outside a road, on which a vehicle travels, and has a predetermined pattern so as to allow identification of each course; and a vehicle location measurement apparatus for acquiring respective images through an AVM camera and synthesizing the images to obtain an AVM image, and recognizing a recognition tag in the AVM image and analyzing the predetermined pattern of the recognition tag, thereby determining a course along which the vehicle travels among the multiple courses.

Description

Devices and systems for measuring the position of the vehicle

The present embodiment relates to a system and an apparatus for measuring the position of a vehicle, in particular, a position of the vehicle in a place divided into a plurality of courses or sections, such as a driving license test site.

The contents described in this section merely provide background information on the present embodiment and do not constitute a prior art.

The driver may drive a vehicle and may travel in a space divided into a plurality of sections or courses. For example, if the driver is a test driver who wants to get a driver's license, the space can be a driver's license test site that is divided into courses. Alternatively, the space could be a large parking lot divided into many compartments.

As such, when a driver drives or parks a vehicle in a space divided into various courses or many sections, there is a need for the driver or the manager of the vehicle to know exactly which course or section is located in the space. .

In order to grasp the position of the vehicle, GPS has conventionally been used. However, when the space is divided into several courses or sections as described above, and the distance between each course or sections is not very far, the conventional method of determining the location of the vehicle using GPS is that the vehicle is exactly in any course or section. It was not easy to figure out if it was located.

Therefore, there is a need for a system for identifying exactly which course or section the vehicle passes or is located in a space divided into a plurality of sections or courses.

An embodiment of the present invention is to provide a lane recognition apparatus and method for determining whether a vehicle crosses a lane using an AVM camera and transmitting the determination result to the outside.

The present embodiment aims to provide a vehicle position measuring system and apparatus for identifying a course or section in which a vehicle is located by identifying a tag indicating each course or section using an AVM camera.

In addition, the present embodiment is to provide an obstacle recognition apparatus and method capable of predicting and preventing the risk of collision between the obstacle and the vehicle by detecting an obstacle approaching the vehicle in all directions of the vehicle using an AVM camera. have.

According to an aspect of the present embodiment, in the system for measuring the position of the vehicle in a space divided into a plurality of courses, located on the outside of the road where the vehicle is moving, have a predetermined pattern to identify each course Acquiring each image from the identification table and the AVM (Around View Monitor) camera, and synthesizing the AVM image, and recognizing the identification table within the AVM image to analyze the predetermined pattern, thereby allowing the vehicle to perform It provides a vehicle position measuring system comprising a vehicle position measuring device for determining which course is in progress.

Further, according to another aspect of the present embodiment, in the apparatus for measuring the position of the vehicle in the space divided into a plurality of courses, the image acquisition unit and the image acquisition unit for acquiring each image from each camera of the AVM camera An AVM image synthesizer for synthesizing each acquired image into an AVM image, and recognizes a recognition tag in the AVM image to analyze a predetermined pattern included in the identification table, and to determine which course of the plurality of courses the vehicle is in progress. Vehicle position measurement, comprising: a course detection unit for detecting course information about; a memory unit for storing information about a course corresponding to each pattern; and a result transmitter for transmitting the course information to the outside of the vehicle position measuring device; Provide the device.

As described above, according to an aspect of the present embodiment, since it is determined whether the vehicle crosses the lane using the AVM camera, it is not possible to accurately determine the lane invasion of the vehicle even if no auxiliary equipment such as a sensor is installed in the lane. There are advantages to it.

According to another aspect of the present embodiment, since the location of the vehicle is measured by identifying the identification tag indicating each course or section using an AVM camera, it is possible to determine which course or section the vehicle is passing through or within. It has the advantage of being able to measure accurately.

According to another aspect of the present embodiment, in the measurement of the position of the vehicle, since the identification table is installed on the outside of the road on which the vehicle in the course or section moves, there is no fear of being damaged by the vehicle even if used for a long time, a long-term error There is an advantage that can measure the position of the vehicle without.

In addition, according to another aspect of the present embodiment, by using the AVM camera to detect the obstacle approaching the vehicle in all directions of the vehicle, there is an advantage that can predict and prevent the risk of collision between the obstacle and the vehicle.

1 is a diagram illustrating a scene in which a vehicle recognizes a lane using a lane recognition apparatus according to an exemplary embodiment of the present invention.

2 is a diagram illustrating a configuration of a lane recognition apparatus according to an embodiment of the present invention.

3 is a diagram illustrating an image output by each camera and an AVM camera constituting a camera according to an embodiment of the present invention.

4 is a flowchart illustrating a method of recognizing a lane by a lane recognition apparatus according to an exemplary embodiment of the present invention.

5 is a view showing a vehicle position measuring system according to an embodiment of the present invention.

6 is a diagram illustrating a tag according to an embodiment of the present invention.

7 is a diagram illustrating a configuration of a vehicle position measuring apparatus according to an embodiment of the present invention.

8 is a flowchart illustrating a method of measuring a vehicle position by a vehicle position measuring apparatus according to an exemplary embodiment of the present invention.

9 is a diagram illustrating a scene in which a vehicle recognizes an obstacle using an obstacle recognition device according to an embodiment of the present invention.

10 is a view showing the configuration of the obstacle recognition apparatus according to an embodiment of the present invention.

11 is a diagram illustrating an image in which a camera recognizes an obstacle according to an embodiment of the present invention.

12 is a flowchart illustrating a method for recognizing an obstacle by an obstacle recognition device according to an embodiment of the present invention.

Hereinafter, some embodiments of the present invention will be described in detail through exemplary drawings. In adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are assigned to the same components as much as possible even though they are shown in different drawings. In addition, in describing the present invention, when it is determined that the detailed description of the related well-known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted.

In addition, throughout the specification, when a part is said to include, 'include' a certain component, which may further include other components rather than excluding other components unless otherwise stated. it means. In addition, as described in the specification. The terms 'unit' and 'module' refer to a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software.

The vehicle 110 includes a plurality of

cameras

120 a, 120 b, 120 c, a lane recognition device 130, and a scoring device 140.

The plurality of

cameras

120 a, 120 b, and 120 c are components constituting an AVM (Around View Monitor) camera and acquire images by photographing from front, rear, and both sides of the vehicle 110. The plurality of

cameras

120 a, 120 b, and 120 c transmit the image acquired by each camera to the lane recognizing device 130, so that the lane recognizing device 130 synthesizes the corresponding image or the AVM image synthesized with the corresponding image. Use it to recognize lanes. Here, each of the plurality of

cameras

120 a, 120 b, and 120 c may be implemented as a wide angle camera having an angle of view of a predetermined angle or more so that the environment around the vehicle can be photographed with a small quantity. FIG. 1 shows that the camera is provided only on the right side 120 a, the front side 120 b, and the left side 120 c in the vehicle 110, but the camera is also provided in the rear of the vehicle 110. It will be obvious by the foregoing.

The lane recognition apparatus 130 obtains images captured by the cameras from a plurality of cameras and synthesizes them into AVM images, recognizes the lanes based on the acquired images and the synthesized AVM images, and determines whether the vehicle 110 has invaded the lanes. To judge. The lane recognition device 130 transmits a determination result of whether the vehicle 110 has invaded the lane to the scoring device 140. A detailed description of the lane recognition device 130 will be described with reference to FIG. 2.

The scoring device 140 is a device for quantifying the driver's driving ability according to whether the driver of the vehicle 110 invades the lane. When the driver of the vehicle 110 drives the vehicle 110 while invading the lane, the scoring apparatus 140 quantifies the driving ability of the driver of the vehicle 110 by deducting a predetermined score from a predetermined score. The scoring device 140 determines whether a score is deducted based on a determination result of whether the vehicle 110 received from the lane recognizing device 130 has invaded the lane. The evaluator evaluates the driver's driving skills using the scoring device 140. For example, when the vehicle 110 is used at the driver's license test site, the driver's license supervisor may determine the sum or failure of the driver's license test of the driver of the vehicle 110 by using the score scored by the scoring device 140. have.

2 is a diagram showing the configuration of a lane recognition device 130 according to an embodiment of the present invention.

Referring to FIG. 2, the lane recognition apparatus 130 according to an embodiment of the present invention may include an image acquisition unit 210, an AVM image synthesis unit 220, a lane recognition unit 230, a result transmitter 240, It includes a notification unit 250 and a memory unit 260.

The image acquisition unit 210 acquires an image from each of the plurality of cameras 120. The image acquisition unit 210 acquires an image captured in each direction of the vehicle 110 from each of the plurality of cameras 120, and transfers the acquired image to the AVM image synthesis unit 220 and the lane recognition unit 230. .

The AVM image synthesizer 220 generates an AVM image by synthesizing the image received from the image acquirer 210. The AVM image synthesizer 220 receives images captured in each direction of the vehicle acquired from the image acquirer 210, and performs image processing such as image enhancement, distortion correction, image registration, and synthesis on these images. The surrounding environment of the vehicle 110 is synthesized into an AVM image, which is a top view image as if viewed from the top of the vehicle 110. Since the synthesis of images from each direction into AVM images is obvious to those skilled in the art, a detailed description thereof will be omitted.

The lane recognizing unit 230 recognizes the lane using the images received from the image obtaining unit 210 or the AVM image synthesizing unit 220, and determines whether the vehicle 110 has invaded the lane. The lane recognizing unit 230 recognizes a straight line or a curve having a predetermined curvature having a predetermined width in the image and having a predetermined length or more as a lane. Since the lane recognizing unit 230 has a predetermined width and recognizes only a straight line or a curve having a predetermined length as a lane, the lane recognizing unit 230 reduces the probability of recognizing a foreign object existing on the road as a lane.

The lane recognizing unit 230 may use not only the AVM image received from the AVM image synthesizing unit 220 but also the respective direction images of the vehicle 110 received from the image obtaining unit 210. When the AVM image synthesizer 220 synthesizes the AVM image, the AVM image synthesizer 220 may be discontinuously synthesized at the boundary of the images in each direction, and there is a concern that the lanes may be discontinuous. Accordingly, there is a fear that the lane recognizing unit 230 may not recognize a straight line or a curve having a predetermined length or more, but may not recognize the lane as a lane despite being a lane. In order to prevent such a problem, the lane recognizing unit 230 considers each direction image of the vehicle 110 received from the image obtaining unit 210 as well as the AVM image received from the AVM image synthesizing unit 220. I can recognize the next day. Accordingly, the lane recognizing unit 230 may improve the lane recognizing performance in the image.

When the lane is recognized in the image, the lane recognizer 230 determines whether the vehicle 110 has invaded the lane. When the lane is recognized in the AVM image, the lane recognizing unit 230 determines whether the vehicle has invaded the lane by determining how far the lane is from the center of the vehicle. Alternatively, when the lane is recognized in the image in the lateral direction, the lane recognizing unit 230 recognizes the lane only in the image in both the left direction and the right direction, or recognizes the lane only in one of the image in the left and right directions. It is determined whether the vehicle has invaded the lane by determining whether the vehicle is in a lane. When the vehicle 110 invades the lane, the lane is recognized only in one of the left and right images, so the lane recognizing unit 230 recognizes the lane in both the left and right images. Determine if you can.

The result transmitter 240 receives a result regarding whether the vehicle 110 has invaded the lane from the lane recognizer 230 and transmits the result to the scoring apparatus 140. The result transmitter 240 transmits the result to the scoring apparatus 140, thereby allowing the evaluator to evaluate the driver's driving ability of the vehicle 110.

If the vehicle 110 invades the lane, the notification unit 250 notifies the outside of the lane violation. The notification unit 250 may be configured as an optical device or an acoustic device. When the vehicle 110 has violated the lane, the notification unit 250 notifies the outside visually or audibly to the driver or the evaluator that the vehicle 110 has violated the lane.

The memory unit 260 stores an image and an AVM image in each direction of the vehicle 110 acquired from the image acquisition unit 210 and the AVM image synthesis unit 220. The memory unit 260 stores the images acquired from the image acquisition unit 210 and the AVM image synthesis unit 220, and provides the image so that the lane recognizing unit 230 may recognize the lane in the image.

3A to 3D show images taken by the plurality of cameras 120 in each direction of the vehicle. 3 (a) shows an image taken from the left side of the vehicle, FIG. 3 (b) shows an image taken from the right side of the vehicle, and FIG. 3 (c) shows an image taken from the front of the vehicle. d) shows an image taken from the rear of the vehicle.

The lane recognizing apparatus 130 receives an image of each direction shown in FIGS. 3A to 3D, and synthesizes the AVM image illustrated in FIG. 3E. As shown in FIG. 3E, the AVM image corresponds to a top view image as if the surrounding environment of the vehicle 110 is viewed from the top of the vehicle 110. The lane recognizing apparatus 130 recognizes a lane by using the image of each direction shown in FIGS. 3A to 3D and the AVM image shown in FIG. 3E, and determines whether the vehicle has invaded the lane. .

The lane recognizing apparatus receives an image photographed from each component of the AVM camera (S410). The lane recognizing apparatus 130 receives an image photographed in each direction of the vehicle 110 from each of the plurality of cameras 120 constituting the AVM camera.

The lane recognizing apparatus generates an AVM image by synthesizing the received image (S420). The lane recognizing device 130 synthesizes images taken from each direction of the vehicle 110 received from each of the plurality of cameras 120, and looks like the environment of the vehicle 110 is viewed from the upper side of the vehicle 110. Creates an AVM image that is a top view image.

The lane recognition apparatus recognizes a lane within each received image and the generated AVM image (S430). The lane recognition apparatus 130 recognizes the lane in the image using not only the generated AVM image but also the image photographed in each direction of the vehicle 110 received from the plurality of cameras 120. As such, by using images captured in each direction of the vehicle 110 together, the lane recognition apparatus 130 increases the accuracy of lane recognition.

The lane recognition apparatus determines whether the vehicle has invaded the recognized lane, and transmits the result to the scoring apparatus (S440). The lane recognition apparatus 130 determines whether the vehicle has invaded the recognized lane by determining whether the lane is recognized in all the side images or how far the recognized lane is left and right from the center of the vehicle. The lane recognizing apparatus 130 transmits the determined result to the scoring apparatus 140, so that an evaluator who wants to evaluate the driver's driving ability can confirm the result.

Referring to FIG. 5A, a vehicle position measuring system according to an exemplary embodiment of the present invention includes a plurality of cameras 514, a vehicle position measuring device 518, and identification tags 520a to e.

The plurality of cameras 514 constitutes an around view monitor (AVM) camera, and acquires images by photographing from front, rear, and both sides of the vehicle 510. The plurality of cameras 514 transmits the images acquired by each camera to the vehicle position measuring device 518, so that the vehicle position measuring device 518 uses the corresponding image or the AVM image synthesized with the corresponding image. To be recognized. Here, each of the plurality of cameras 514 may be implemented as a wide-angle camera having an angle of view of a predetermined angle or more, so that the environment around the vehicle can be photographed with a small quantity. Although FIG. 5 illustrates that the camera is provided only on the right side 514 of the vehicle 110, it will be apparent that the camera is also provided on the front, rear, and left sides of the vehicle 110.

The vehicle position measuring apparatus 518 is installed in the vehicle 510 and acquires images captured by the cameras from the plurality of cameras 514 and synthesizes the images into AVM images. The vehicle position measuring apparatus 518 is based on the acquired images and the synthesized AVM images. 510 detects where it is located. When the vehicle 510 is driving on a road in a space divided into a plurality of courses or sections, the vehicle position measuring device 518 may include a vehicle 510 in which a course 510 is located or a course or section. Detect if it is passing. Here, the space may be a space divided into various courses, such as a driver's license test site, or may be a space divided into a plurality of sections, such as a parking lot. The vehicle position measuring apparatus 518 detects which course or section the vehicle is located in the space and transmits the detected position to the driver or the manager of the vehicle. Detailed description thereof will be described with reference to FIG. 7.

The identification tags 520a to e correspond to marks installed in the corresponding sections to recognize each course or section. As shown in FIG. 5, the identification tags 520a to e are installed outside the road rather than the road in each course or section. When the identification tag is installed on the road in each course or section, there is a concern that the vehicle 510 repeatedly runs on the identification mark and is damaged so that recognition is difficult or impossible. Therefore, the identification tags 520a to e are installed outside the road rather than the road in each course or section. FIG. 5 shows that one course is provided for each course, but is not necessarily limited thereto, and may be installed at each of a starting point, a middle point, and a final point of each course to recognize the course.

The identification tags 520a to e have a predetermined size and include a two-dimensional bit pattern assigned to each course. The identification tags 520a through e are designed to have a pattern or color that can be easily identified by the camera. The identification tables 520a to e have different patterns for each course or section, and include two-dimensional bit patterns having an n * n size. For example, the identification tags 520a to e may be implemented with ArUco markers. The identification tags 520a to e have two-dimensional bit patterns different from each other so that the starting point, the middle point, and the last point of each course or section or each course or section are distinguished. As described above, the identification tables 520a to e have different patterns for each course or section, so that the vehicle position measuring device 518 analyzes the patterns in the identification tables 520a to e to measure where the vehicle 510 is located. can do.

6 illustrates an example of the identification tag 520. The identification table 520 shown in FIG. 6 has a size of 6 * 6 and has a constant two-dimensional bit pattern. The bright part in the dog tag 520 may represent 1, and the dark part may represent 0. Accordingly, the identification table 520 may be recognized as the two-dimensional bit matrix 610 in the vehicle position measuring apparatus 520. The vehicle position measuring apparatus 520 may determine a course or section or a recognition table installed at any point of the course or section by analyzing the 2D bit matrix 610. As such, the identification table 520 may identify each of the various courses or sections or by matching various points of the courses or sections with the two-dimensional bit pattern.

Referring to FIG. 7, the vehicle position measuring apparatus 518 according to an embodiment of the present invention may include an image acquisition unit 710, an AVM image synthesis unit 720, a course detection unit 730, a result transmitter 740, and the like. The memory unit 750 is included. Furthermore, the vehicle position measuring device 518 may further include a positioning unit 760.

The image acquisition unit 710 acquires an image from each of the plurality of cameras 514. The image acquisition unit 710 acquires an image captured in each direction of the vehicle 510 from each of the plurality of cameras 514, and transfers the acquired image to the AVM image synthesis unit 720 and the course detection unit 730.

The AVM image synthesizer 720 generates an AVM image by synthesizing the image received from the image acquirer 710. The AVM image synthesizing unit 720 receives images photographed in each direction of the vehicle acquired from the image capturing unit 710, and performs image processing such as image enhancement, distortion correction, image matching, and compositing on these images, The surrounding environment of the vehicle 510 is synthesized into an AVM image, which is a top view image as if viewed from the top of the vehicle 510. Since the synthesis of images from each direction into AVM images is obvious to those skilled in the art, a detailed description thereof will be omitted.

The course detector 730 recognizes the identification table using the images received from the image acquisition unit 710 or the AVM image synthesis unit 720, and analyzes the two-dimensional bit pattern in the identification table to detect a course or section. The course detector 730 recognizes the identification table having a predetermined size in the image photographed in each direction of the vehicle 510 or the AVM image, and having a constant two-dimensional bit pattern. The course detector 730 analyzes the recognized recognition table, converts it into a two-dimensional bit matrix, and analyzes the two-dimensional bit matrix to determine which course or section the corresponding recognition table represents. For example, when the identification table 520 is implemented with an ArUco marker, the course detector 730 converts the ArUco marker into a two-dimensional bit matrix by analyzing the ArUco marker by using an ArUco marker detection algorithm. The course detector 730 uses the information about the course or section corresponding to the two-dimensional bit matrix stored in the memory unit 750 to grasp the course or section indicated by the recognition table.

The course detector 730 determines which image the recognition table is recognized in the image photographed in each direction or the AVM image, and determines in which direction the vehicle 510 is traveling for a particular course. When the identification table is recognized in the image on the right side of the vehicle 510 or when the identification table is recognized on the right side based on the center of the vehicle in the AVM image, the course detection unit 730 uses the course or section indicated by the identification table in the vehicle 510. It can be seen that it is proceeding in the direction of entry. On the contrary, when the identification tag is recognized in the image on the left side of the vehicle 510 or when the identification tag is recognized on the left side based on the center of the vehicle in the AVM image, the course detector 730 may detect the vehicle in the course or section indicated by the identification table. It can be seen that 510 is progressing in a direction away. However, unlike the vehicle driving characteristics in Korea, in the countries having the characteristics that the vehicle passes to the left, such as the United Kingdom, Australia, Japan, it can be seen as opposed to the above. As such, the course detection unit 730 recognizes and analyzes the identification table to determine which course or section the vehicle 510 is located in, as well as which direction of the vehicle the identification table is located to determine which vehicle or the vehicle 510 is located in. Identify whether you are entering or leaving a course or section.

The course detector 730 extracts the coordinate values of the identification table detected in the image to determine how far from the vehicle 510, so as to determine how far in which direction the vehicle 510 is traveling for a particular course. do. The image received from the image acquisition unit 710 or the AVM image synthesis unit 720 corresponds to an image that can be acquired by a camera using a wide-angle lens, and the object may be distorted. The distance between objects) and the coordinates in the image have different problems. To solve this problem, the course detector 730 converts the corresponding image into the camera coordinate system using the distance transformation matrix. The course detector 730 extracts the coordinate values of the identification table in the image converted into the camera coordinate system and determines how far from the vehicle 510. Through the above-described process, the course detector 730 determines which course or section the vehicle 510 is located in, in which direction it is proceeding for a particular course or section, and in what direction and how much it is progressing in a particular course. Can be identified.

The result transmitter 740 receives the result detected by the course detector 730 from the course detector 730 and transmits the result to the terminal (not shown) of the driver or the manager of the vehicle. The result transmitter 740 transmits the result detected by the course detector 730 to the terminal of the driver, so as to determine where the vehicle is parked. Alternatively, the result transmitter 740 may determine which course or section the driver enters or exits by transmitting the result detected by the course detector 730 to the terminal of the manager.

The memory unit 750 stores an image and an AVM image in each direction of the vehicle 510 obtained from the image acquisition unit 710 and the AVM image synthesis unit 720. The memory unit 750 stores the images acquired from the image acquisition unit 710 and the AVM image synthesis unit 720, and provides an image so that the course detection unit 730 may recognize a lane in the image.

In addition, the memory unit 750 stores information of each course or section corresponding to the 2D bit pattern. When the course detector 730 analyzes the 2D bit pattern of the recognition table, the memory unit 750 transmits the course or section information corresponding to the pattern to the course detector 730.

The positioning unit 760 measures the position of the vehicle 510. The positioning unit 760 may be implemented as a GPS module or a communication module. When the positioning unit 760 is implemented as a GPS module, position information of the vehicle 510 may be directly obtained. Alternatively, when the positioning unit 760 is implemented as a communication module, the vehicle may be configured in consideration of the time when the communication module transmits radio waves to the repeater or the receiving end, the time when the response is received from the repeater or the receiving end, and the direction from the repeater or the receiving end. The location information of 510 may be obtained. The positioning unit 760 transmits the acquired location information to the course detector 730, so that the course detector 730 may measure the position of the vehicle more accurately.

The vehicle position measuring apparatus receives an image photographed from each component of the AVM camera (S810). The vehicle position measuring apparatus 518 receives images captured in each direction of the vehicle 510 from each of the plurality of cameras 514 constituting the AVM camera.

The vehicle position measuring apparatus synthesizes the received image to generate an AVM image (S820). The vehicle position measuring device 518 synthesizes images taken in each direction of the vehicle 510 received from each of the plurality of cameras 514, and views the surrounding environment of the vehicle 510 from the upper side of the vehicle 510. Create an AVM image that looks like a top view image.

The vehicle position measuring apparatus recognizes the identification table within each received image and the generated AVM image (S830). In recognizing the identification table 520 in the image, the vehicle position measuring apparatus 518 may determine whether the identification table is recognized from the image of the direction of the vehicle or in which direction the recognition table is recognized based on the center of the vehicle. In addition, the coordinates of the recognition table may be determined by converting the image into a camera coordinate system using a distance transformation matrix.

The vehicle location measuring apparatus analyzes the identification table to determine which course the vehicle is currently entering (S840). The vehicle position measuring apparatus 518 analyzes the recognized identification table 520 to acquire the course or section information corresponding to the two-dimensional bit pattern in the identification table 520 to determine the position of the vehicle 510. The vehicle position measuring apparatus 518 may determine a course or section of the vehicle based on a result of whether the identification table is recognized from the image of the direction of the vehicle 510 or in which direction the identification table is recognized based on the center of the vehicle 510. Know which way you're driving. Furthermore, the vehicle position measuring device 518 may determine how much the vehicle 510 has traveled in a certain course or section using the coordinate values of the identification table.

The vehicle 910 includes a plurality of

cameras

920 a, 920 b, 920 c, an obstacle recognition device 930, and an electronic braking device 940.

The plurality of

cameras

920 a, 920 b, and 920 c are components constituting an around view monitor (AVM) camera, and acquire images by photographing from front, rear, and both sides of the vehicle 910. The plurality of

cameras

920 a, 920 b, and 920 c transmits the image acquired by each camera to the obstacle recognition device 930, so that the obstacle recognition device 930 may combine the image or the AVM image that synthesized the image. To recognize the obstacle 950. Here, each of the plurality of

cameras

920 a, 920 b, and 920 c may be implemented as a wide-angle camera having an angle of view of a predetermined angle or more so that the environment around the vehicle can be photographed with a small quantity. 9 illustrates that the camera is provided only on the right side 920 a, the front side 920 b, and the left side 920 c of the vehicle 910, but the camera is also provided behind the vehicle 910. It will be obvious by the foregoing.

The obstacle recognition apparatus 930 obtains images captured by the cameras from the plurality of cameras, synthesizes the images into AVM images, and recognizes the obstacle 950 based on the acquired images and the synthesized AVM images. In addition, when the obstacle 950 approaches the vehicle 910 within a predetermined distance, the obstacle recognition device 930 transmits a brake signal to the electronic braking device 940 to stop the vehicle 910. A detailed description thereof will be described with reference to FIG. 10.

Electronic Stability Control (ESC) 940 is a device for controlling the braking of the vehicle, and receives the braking signal of the obstacle recognition device 930 to stop the vehicle.

In FIG. 9, a person is shown as an obstacle 950, but is not necessarily limited thereto, and all objects that hinder the driving of the vehicle 910 may be included as an obstacle 950 except for intentionally installed in a road such as a bump. have.

Referring to FIG. 10, the obstacle recognition apparatus 930 according to an embodiment of the present invention may include an image acquisition unit 1010, an AVM image synthesis unit 1020, an obstacle recognition unit 1030, a coordinate calculation unit 1040, and a distance. The determination unit 1050, a notification unit 1060, and a control signal transmission unit 1070 are included.

The image acquisition unit 1010 acquires an image from each of the plurality of cameras 920. The image acquisition unit 1010 acquires images captured in each direction of the vehicle 910 from each of the plurality of cameras 920, and transfers the acquired images to the AVM image synthesis unit 1020 and the obstacle recognition unit 1030. .

The AVM image synthesizer 1020 generates an AVM image by synthesizing the image received from the image acquirer 1010. The AVM image synthesizing unit 1020 receives images photographed in each direction of the vehicle acquired from the image capturing unit 1010, and performs image processing such as image enhancement, distortion correction, image matching and compositing on these images, The surrounding environment of the vehicle 910 is synthesized into an AVM image, which is a top view image as if viewed from the top of the vehicle 910. Since the synthesis of images from each direction into AVM images is obvious to those skilled in the art, a detailed description thereof will be omitted.

The obstacle recognition unit 1030 recognizes an obstacle in an image or an AVM image photographed from each direction of the vehicle 910 received from the image acquisition unit 1010 or the AVM image synthesis unit 1020. The obstacle recognition unit 1030 may recognize an object having a predetermined length or more or a predetermined width or more as an obstacle in the image. If even one small object is recognized as an obstacle, since it may interfere with normal driving of the vehicle 910, only an object having a predetermined length or width or more may be recognized as an obstacle. The obstacle recognition unit 1030 may recognize all as obstacles when there are a plurality of objects having a predetermined length or width in the image. An example of such an image is shown in FIG. 11.

FIG. 11 is a diagram illustrating an image in which a camera recognizes an obstacle 950 according to an embodiment of the present invention.

The obstacle 950 in the image 1110 taken from one side of the vehicle 910 is recognized. Since the recognized obstacle 950 has a predetermined length or more, the obstacle recognition unit 1030 recognizes the obstacle 950 in the image.

Referring back to FIG. 9, the coordinate calculator 1040 calculates coordinate values of an obstacle recognized by the obstacle recognizer 1030 in the image. The image received from the image acquisition unit 1010 or the AVM image synthesis unit 1020 corresponds to an image that can be acquired by a camera using a wide-angle lens, and thus an object may be distorted. The distance between objects) and the coordinates in the image have different problems. To solve this problem, the coordinate calculation unit 1040 converts the image into a camera coordinate system using a distance transformation matrix. The coordinate calculator 1040 extracts the coordinate value of the obstacle in the image converted into the camera coordinate system. The coordinate calculator 1040 may calculate all coordinate values of each obstacle when the obstacle recognizer 1030 recognizes a plurality of obstacles in the image.

The distance determiner 1050 determines the distance between the vehicle 910 and the obstacle using the coordinate values of the obstacle calculated by the coordinate calculator 1040, and determines whether the distance between the vehicle 910 and the obstacle is less than a preset reference value. do. The distance determiner 1050 determines a distance to the obstacle 950 around the vehicle 910 by using the coordinate value of the obstacle. In this case, the distance determiner 1050 determines whether the determined distance between the vehicle 910 and the obstacle is less than a preset reference value. In this case, the preset reference value may vary depending on the speed of the vehicle. The distance determiner 1050 receives the speed of the vehicle from a speed sensor (not shown) included in the vehicle 910, and may set a reference value differently according to the received speed. Since the braking distance increases as the speed of the vehicle increases, the preset reference value increases as the speed of the vehicle increases. That is, when the speed of the vehicle 910 is fast, even if the obstacle 950 is relatively far from the vehicle, the distance between the vehicle 910 and the obstacle may be determined to be less than the preset reference value. On the contrary, when the speed of the vehicle 910 is slow, it may be determined that the distance between the vehicle 910 and the obstacle exceeds a preset reference value even when the obstacle 950 is relatively close to the vehicle. When there are a plurality of obstacles, the distance determination unit 1050 determines the distance between each obstacle and the vehicle by using each coordinate value, and determines whether the distance between the obstacle and the vehicle located closest to each other is less than the preset reference value. Judge.

The notification unit 1060 notifies whether the obstacle in the image is recognized or whether the distance between the vehicle 910 and the obstacle is less than a preset reference value. The notification unit 1060 may be configured of an optical device or an acoustic device. When an obstacle in the image is recognized, the notification unit 1060 may inform the driver visually or audibly to the outside, and when the obstacle is a person, a person near the vehicle 910 may recognize the vehicle. Make sure On the other hand, if there is an obstacle whose distance to the vehicle 910 is less than the predetermined reference value, the notification unit 1060 may notify the outside with a greater light and sound than when the obstacle in the image is recognized.

When there is an obstacle whose distance from the vehicle 910 is less than a preset reference value according to the determination of the distance determiner 1050, the control signal transmitter 1070 controls the vehicle to brake the vehicle with the electronic brake 940. Send it. If there is an obstacle whose distance from the vehicle 910 is less than a predetermined reference value, there is a possibility of collision between the vehicle and the obstacle, and human and physical damage may occur. Accordingly, the control signal transmitter 1070 controls the electronic brake 940 to stop the vehicle without the driver's operation by transmitting a control signal for braking the vehicle to the electronic brake 940.

12 is a flowchart illustrating a method of recognizing an obstacle 950 by an obstacle recognition device according to an exemplary embodiment of the present invention.

The obstacle recognition apparatus receives the captured image from each component of the AVM camera (S1210). The obstacle recognition apparatus 930 receives an image photographed in each direction of the vehicle 910 from each of the plurality of cameras 920 constituting the AVM camera.

The obstacle recognition apparatus generates an AVM image by synthesizing the received image (S1220). The obstacle recognition device 930 synthesizes the images photographed in each direction of the vehicle 910 received from each of the plurality of cameras 920, and looks like the environment of the vehicle 910 is viewed from the upper side of the vehicle 910. Creates an AVM image that is a top view image.

The obstacle recognition apparatus recognizes an obstacle in each of the received images and the generated AVM image (S1230).

The obstacle recognition apparatus calculates coordinate values of the recognized obstacle in the image (S1240). The obstacle recognition apparatus 930 calculates coordinate values for all obstacles when there are a plurality of recognized obstacles in the image.

The obstacle recognition apparatus detects an obstacle located closest to the vehicle among the recognized obstacles (S1250). The obstacle recognition device 930 calculates a distance between the vehicle 910 and the obstacle 950 using the calculated coordinate values. When there are a plurality of obstacles recognized in the image, the obstacle recognition apparatus calculates the distance between all the obstacles 950 and the vehicle 910. In this case, when there are a plurality of obstacles recognized in the image, the obstacle 950 that detects the closest distance between the obstacle 950 and the vehicle 910 is detected. The obstacle recognition apparatus determines whether the calculated distance between the vehicle 910 and the obstacle 950 or the detected distance is the distance between the nearest obstacle 950 and the vehicle 910 is less than a preset reference value.

The obstacle recognition apparatus notifies that there is an obstacle approaching within a preset reference value, and transmits a control signal to the electronic brake device to control the braking (S1260). The obstacle recognition device 930 notifies that when an obstacle 950 having a distance between the vehicle 910 and the obstacle 950 is less than a preset reference value exists, an obstacle approaching to the outside within the preset reference value exists. In addition, the obstacle recognition device 930 transmits a control signal for braking the vehicle to the electronic braking device 940, thereby controlling the electronic braking device 940 to stop the vehicle without the driver's operation.

4, 8 and 12 are described as sequentially executing each process, which is merely illustrative of the technical idea of an embodiment of the present invention. In other words, a person of ordinary skill in the art to which an embodiment of the present invention belongs may change the order of the processes described in each drawing or execute one or more of the processes without departing from the essential characteristics of the embodiment of the present invention. 4 and 8 and 12 are not limited to the time series since the processes may be applied in various ways.

4, 8, and 12 may be implemented as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. That is, the computer-readable recording medium may be a magnetic storage medium (for example, ROM, floppy disk, hard disk, etc.), an optical reading medium (for example, CD-ROM, DVD, etc.) and a carrier wave (for example, the Internet Storage medium). The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

The above description is merely illustrative of the technical idea of the present embodiment, and those skilled in the art to which the present embodiment belongs may make various modifications and changes without departing from the essential characteristics of the present embodiment. Therefore, the present embodiments are not intended to limit the technical idea of the present embodiment but to describe the present invention, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present embodiment.

CROSS-REFERENCE TO RELATED APPLICATION

This patent application claims priority to Patent Application No. 10-2017-0061892 filed in Korea on May 18, 2017, which is hereby incorporated by reference in its entirety.

Claims

In the system for measuring the position of the vehicle in the space divided into a plurality of courses,

A identification table positioned outside of a road on which the vehicle moves and having a predetermined pattern to identify each course; And

By acquiring each image from an AVM (Around View Monitor) camera and synthesizing the AVM image, and recognizing the identification tag in the AVM image and analyzing the preset pattern, the vehicle proceeds to any one of the plurality of courses. Vehicle positioning system

Vehicle position measuring system comprising a.
The method of claim 1,

The tag is,

And a two-dimensional bit pattern provided for each of the plurality of courses.
The method of claim 1,

The vehicle position measuring device,

An image acquisition unit for acquiring each image from each camera of the AVM camera;

An AVM image synthesis unit for synthesizing each image acquired by the image acquisition unit into an AVM image;

A memory unit which stores information about a course corresponding to each pattern;

Recognizing the identification table in the AVM image, analyzing a predetermined pattern included in the identification table, and using the information stored in the memory unit, the course information on which course of the plurality of courses the vehicle is running A course detector for detecting; And

And a result transmitter for transmitting the course information to the outside of the vehicle position measuring apparatus.
The method of claim 3,

The course detection unit,

By determining whether the identification tag is recognized in an image obtained from which camera of each of the AVM cameras, or in which direction the identification tag is recognized based on the center of the vehicle in the AVM image, the vehicle is selected from among the plurality of courses. A vehicle position measuring system, characterized in that it is determined in which direction the course is going.
The method of claim 4, wherein

The course detection unit,

Vehicle location measurement, characterized in that by determining the distance of the course of the plurality of courses in which direction by the vehicle by extracting the coordinates of the identification table from the image of the recognition table using a distance transformation matrix system.
In the apparatus for measuring the position of the vehicle in the space divided into a plurality of courses,

An image acquisition unit which acquires each image from each camera of the AVM camera;

An AVM image synthesis unit for synthesizing each image acquired by the image acquisition unit into an AVM image;

A course detector configured to recognize a tag in the AVM image, analyze a predetermined pattern included in the tag, and detect course information regarding which course of the plurality of courses the vehicle is running;

A memory unit which stores information about a course corresponding to each pattern; And

Result transmission unit for transmitting the course information to the outside of the vehicle position measuring device

Vehicle position measuring apparatus comprising a.
The method of claim 6,

The tag is,

And a two-dimensional bit pattern provided to each of the plurality of courses.
The method of claim 6,

The course detection unit,

By determining whether the identification tag is recognized in an image obtained from which camera of each of the AVM cameras, or in which direction the identification tag is recognized based on the center of the vehicle in the AVM image, the vehicle is selected from among the plurality of courses. A vehicle position measuring device, characterized in that it is determined in which direction the course is going.
The method of claim 8,

The course detection unit,

Vehicle location measurement, characterized in that by determining the distance of the course of the plurality of courses in which direction by the vehicle by extracting the coordinates of the identification table from the image of the recognition table using a distance transformation matrix Device.