KR20160055413A - Automatic scoring system for driving license test using around-view monitering video, and method for the same - Google Patents

Abstract

The present invention generally relates to a technology of effectively and automatically marking a driver′s license test and, more specifically, to a technology of effectively and automatically marking a driver′s license test: obtaining an image of the periphery of a test vehicle through a camera; recognizing a section where the vehicle is proceeding through an image process; determining whether or not the vehicle goes into a test course; detecting a division line inside the test course; and determining whether or not the test vehicle accurately performs a task of the driver′s license test without deviating from the division line. According to the present invention, in a driver′s license test, an automatic marking system for a driver′s license test can be established at a relatively inexpensive price.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic scoring system,

The present invention generally relates to a technique for effectively scoring a driver's license test. More particularly, the present invention relates to a method and apparatus for acquiring an image around a test vehicle through a camera, recognizing an interval that the vehicle is proceeding through image processing, determining whether the test course has entered the inside of the test course, The present invention relates to a technique for automatically determining whether a vehicle carries out a test of driving license test without departing from a zone line, and thereby automatically scoring a driving license test.

The driver 's license test consists of a departmental test, a functional test, and a road test. Among these, the functional test is to evaluate the driver' s driving ability by installing a terrain similar to an actual road in a driver 's license test center.

The driver's license test system is equipped with a technical device for the driving license test system to detect the driver's immaturity and reflect it on the functional test score. Routes and baselines that the vehicle must follow for each task for the functional test are set. The driver's license test system buries a sensor (sensor) on the road surface of these paths or baselines, and when the test vehicle leaves the test vehicle and presses the sensor with the wheel, it is regarded as a driving mistake.

The conventional technique for performing the automatic grading in the driver's license test has been based on the basic structure in which the sensor is buried in the driving license test room and the test vehicle passes through it.

In this manner, multiple sensors should be embedded along the path at the bottom of the driver's license test site. For this purpose, from the time of civil engineering works for the construction of the driver 's license test site, the drawing for the sensor burial has to be designed and the sensor burial has to be done along with the civil engineering works. In the case of the existing driver's license test center, it was necessary to re-dig the floor of the test ground and lay down the sensor for automatic scoring while the driver's license test was stopped for all or a part of the area for a certain period of time. It took a considerable amount of time and money to construct a facility for automatic scoring.

Accordingly, there has been a demand for a technology capable of solving the problems of the prior art by reducing the cost and time for constructing the driving license test center in implementing the automatic license driving test.

1. Korean Patent Application No. 10-1996-0042147 "Driver's License Electronic Scoring System"

2. Korean Utility Model Registration Application No. 20-2001-0017959 entitled "Active Deviation Detection Device"

3. Korean Patent Application No. 10-2001-0051651 "Pressure Sensing System and Driver's License Test System Using It"

4. Korean Patent Application No. 10-2002-0009056 entitled " Road Running Scoring Device, Electronic Scoring System Including It, and Scoring Method "

5. Korean Patent Application No. 10-2004-002726 entitled " System and Method for Automatically Evaluating Vehicle Driving License Practice Using a State of the Art Satellite Positioning Apparatus "

6. Korean Patent Application No. 10-2005-0042906 "Apparatus for determining path deviation in driver's license test system and method thereof"

7. Utility Model Registration Application No. 20-2005-0014458 of the Republic of Korea "Apparatus for determining the departure of a vehicle from a driver's license test"

8. Korean Patent Application No. 10-2011-0139112 entitled " Road Test Test Scoring System and Road Test Test Scoring Method Using It "

9. Korean Patent Application No. 10-2012-0119083 "Electronic Scoring Apparatus and Method for Road Driving Test"

It is an object of the present invention to provide a technique for effectively scoring a driver's license test. In particular, it is an object of the present invention to provide a method and apparatus for acquiring a video image of a test vehicle through a camera, recognizing a zone on which the vehicle is proceeding through image processing, This is to provide a technique for judging whether or not to carry out the task of the driver's license test correctly without going out of the section line and automatically scoring the driver's license test.

According to another aspect of the present invention, there is provided a system for automatically scoring a driver's license using an overview image, comprising: an AVM image processor for generating an AVM image for a test vehicle by image- (100); And an automatic scoring controller (200) for storing identification information of a division check mark for the driver's license test course and analyzing the AVM image based on the identification information of the division check mark to perform automatic scoring of the driver's license test do.

At this time, the automatic scoring controller 200 includes an AVM receiver 210 for receiving an AVM image of a test vehicle from the AVM image processor; A check mark storage unit 230 for storing identification information of a division check mark including a section recognition mark and a course partition line for the driver's license test course; An interval marker identifying unit 240 for identifying the interval marker in the AVM image; A partition line identification unit 250 for identifying a course partition line in the AVM image; A scoring event processing unit (260) for identifying whether the test vehicle is out of the course boundary according to the driving license test course based on the course segment line and processing individual scoring corresponding thereto; And an automatic grading control unit (270) for identifying a test course based on the interval recognition markers and collecting individual grading results of the grading event processing unit and processing automatic grading for the identified test course.

The automatic grading control unit 200 further includes a vehicle sensor receiving unit 220 for receiving sensor inputs from a plurality of vehicle sensors installed on the test vehicle, And to reflect the result of comparing the requirement with the sensing input of the vehicle sensor to the test course score.

In addition, the driving license automatic grading system using the surround view image according to the present invention includes a grading monitor 300 for displaying an automatic grading result by the automatic grading controller; And a scoring management computer 400 of a control room for wirelessly receiving an automatic scoring result by the automatic scoring controller and real-time monitoring of the test progress status of the test vehicle.

Meanwhile, a method for automatically scoring a driving license using an approach view image according to the present invention includes: a first step of storing identification information of a division check mark for a driving license test course; A second step of generating an AVM image around the test vehicle by image processing an image captured from a plurality of cameras installed on the test vehicle; And a third step of performing automatic scoring of the driver's license test by analyzing the AVM image based on the identification information of the division check mark.

In this case, the division check mark includes a section recognition mark and a course partition line for the driving license test course, and the third step includes: identifying the examination course by identifying the section recognition mark on the AVM image; Identifying a course segment line in the AVM image to identify whether the test vehicle is out of the course boundary according to the driving license test course and processing the individual scoring accordingly; Collecting the individual scoring results and processing the automatic scoring for the identified test course.

In another aspect of the present invention, there is provided a method for automatically scoring a driver's license using an approach-view image, the method comprising: receiving a sensing input from a plurality of vehicle sensors installed on a test vehicle performed after a first step and before a third step; And the third step further comprises the step of identifying the position of the test vehicle in the test course from the AVM image and reflecting the result of the comparison with the requirement of the sensory input of the vehicle sensor in the test course scoring .

Meanwhile, a computer-readable recording medium according to the present invention records a program for executing a driver's license automatic scoring method using an overview image such as the above.

According to the present invention, there is an advantage that a driver's license automatic scoring system can be constructed quickly at a relatively low price in the driver's license test.

Particularly, according to the present invention, it is possible to obtain accurate test results through automatic scoring if only a simple camera device for obtaining an overview image of a vehicle is installed without burdening a large number of sensors on the floor of a driving license examination room.

In addition, according to the present invention, it is not necessary to install facilities (for example, sensors, tubes, etc.) for supporting automatic scoring on the floor of the driver's license test center, It has the advantage of being able to install quickly and quickly.

According to the present invention, even when an automatic scoring system is introduced to an existing driver's license test center, a separate floor facility is not required and system conversion is efficient since only some paintings need to be corrected. Test vehicles equipped with the camera equipment that generates the surround view image can be produced from outside, so the operation of the driving license test center can be continued normally even when the scoring system is switched.

1 is a view showing an example of a general overview image,
[Fig. 2] conceptually shows a structure for achieving automatic scoring of individual items of the driver's license test based on the surround view image according to the present invention,
3 is a block diagram showing an internal configuration of an automatic scoring controller in the present invention,
4 is a view conceptually showing a data collection and processing structure for automatic scoring in the driver's license automatic scoring system of the present invention,
5 is a diagram conceptually showing a course test process to which the driver's license automatic scoring system of the present invention is applied,
6 is a diagram showing that the driving license automatic scoring system of the present invention can be applied to various test courses,
7 is a flowchart showing an entire process for implementing driving license automatic scoring based on the surround view image according to the present invention;
FIG. 8 is a flowchart showing an embodiment of implementing driving license automatic scoring in the present invention. FIG.

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

FIG. 1 is a diagram showing an example of a general overview image (AVM) image.

Recently, AVM (Around View Monitor) system has been introduced to support parking. In the AVM system, a plurality of cameras are installed in a vehicle, and the images are photographed in front, rear, left and right sides of the vehicle, and then these photographed images are combined and displayed as a single image based on a change in the moving distance and azimuth of the vehicle.

Referring to FIG. 1, the AVM system continuously receives a plurality of photographed images 111 to 114 from a plurality of cameras 101 to 104 mounted on the front, rear, right, and left sides of the vehicle, Distortion correction, image registration, and synthesis to acquire the surround view image 115 of the vehicle.

FIG. 2 is a conceptual view showing a structure for achieving automatic grading of individual items of the driver's license test based on the surround view image according to the present invention. This is a conceptual diagram showing the course test process applied. 6 is a diagram showing that the driving license automatic scoring system of the present invention can be applied to various test courses.

5, for each test course of the driver's license test center, division check marks 131 to 135 for assisting automatic scoring according to the present invention are indicated, for example, by paints. At this time, it is preferable that the division check mark is realized by the interval recognition mark 131 and the course partition lines 132 to 135.

Referring to FIG. 2, the automatic scoring controller 200 performs automatic scoring of the driving license test based on the image processing and the object identification function for the AVM image 115 of the test vehicle. To this end, the auto-scoring controller 200 identifies which task (s) the test vehicle M should currently perform in the driving license test by identifying the interval recognition mark 131, e.g., "T1" The interval recognition tag 131 is used to distinguish the test course, and uses a predetermined mark previously set.

Subsequently, the automatic grading controller 200 identifies the course segment lines 132 to 135 for the test course by analyzing the AVM image 115 and determines whether the test vehicle M is out of the course boundary according to the driving license test course . FIG. 2 shows a case where the test vehicle invades the course boundary on the rear side. The automatic scoring controller 200 processes individual scoring points (for example, deduction points) according to a preset method every time the test vehicle M is identified as being out of the course boundary, and when the mission is completed, Handle automatic grading for the test course.

On the other hand, the course partition lines 132 to 135 are not limited to colors and patterns, but are preferably selected, for example, as yellow and black spots so as to facilitate identification even when the outdoor environment changes.

Whether or not the test vehicle M has entered the test course and started performing the task can be identified by the interval recognition mark 131 or by the course partition lines 132 to 135. [ 5, an entry boundary line is displayed in a part of the interval recognition mark 131 so that entry of the test course can be identified.

On the other hand, such a method of scoring a driver's license test based on an AVM image can be applied to various types of test courses. FIG. 6 is a diagram showing that the driving license automatic scoring system of the present invention can be applied to various test courses, and appropriately performs missions for test courses such as serial parking, parallel parking, angle parking, sudden braking, An evaluation of whether or not it has been completed can be automatically scored by analyzing the AVM image. 6, section recognition marks (T1, P1, A1, E1, and H1) are displayed differently for each test course, and a course partition line is appropriately displayed so that performance of the mission can be evaluated .

FIG. 3 is a block diagram showing an internal configuration of the automatic grading controller 200 in the present invention. FIG. 4 conceptually shows a data gathering and processing structure for automatic grading in the driver's license automatic grading system of the present invention. FIG.

Referring to FIG. 4, the DRM system according to the present invention includes an AVM image processor 100, an automatic scoring controller 200, a scoring monitor 300, and a scoring management computer 400 do.

The AVM image processor 100 generates an AVM image 115 for the test vehicle M by image processing the photographed images 111 to 114 obtained from the plurality of cameras 101 to 104 provided in the test vehicle M do.

Four 180 degree wide angle (fisheye) cameras 101 to 104 are mounted on the front, rear, left and right sides of the test vehicle M, and a plurality of photographed images 111 to 114 are obtained through these cameras. The images obtained through the wide-angle camera are not flat but distorted in a circular shape, which can be flattened and combined by image processing to obtain all of the images around the test vehicle M.

The image thus obtained is only the image of the surroundings of the vehicle, and the place where the actual vehicle should be present is empty. When the images of the vehicle are combined, as shown in FIG. 1, Can be implemented. In order to obtain a more sophisticated image without distortion, an additional image tuning process is required. After the tuning process, a good quality surrounding view image 115 for the test vehicle M can be obtained.

The automatic grading controller 200 stores the identification information of the division check marks 131 to 135 for the driver's license test course, and based on the identification information of the division check mark, the AVM image processor 100 real- (115) of the driver's license test.

The automatic grading controller 200 performs high-speed real-time image recognition processing on the AVM image 115 provided from the AVM image processor 100, for example, real-time image recognition of 5 frames per second or more, Recognizes whether or not the vehicle is in the lane, and carries out automatic scoring by combining the sensing inputs of various vehicle sensors 105 installed in the test vehicle (M).

To this end, the auto-scoring controller 200 stores the identification information of the division check marks 131 to 135 for the driver's license test course. As described above with reference to FIG. 5, the driver's license test site is provided with the interval recognition mark 131 and the course partition lines 132 to 135 as the division check marks for assisting the automatic scoring according to the present invention for each test course For example, paint. When a division check mark appears on the AVM image 115 photographed by the camera, the automatic grading controller 200 must identify the AVM image 115 and store identification information therefor.

On the other hand, an embodiment of the automatic grading controller 200 for performing the automatic grading of the driving license test by analyzing the AVM image 115 based on the identification information of the division check mark will be described later with reference to FIG.

The scoring monitor 300 displays the result of automatic scoring by the automatic scoring controller 200 on a monitor screen. Preferably, the scoring monitor 300 is installed inside the test vehicle M, and the examiner or the supervisor can check the automatic scoring result at that time. .

The scoring management computer 400 is a device for wirelessly receiving the automatic scoring result by the automatic scoring controller 200 and real-time monitoring the test progress of the test vehicle M by the control room of the driving license test center. In addition, the score management computer 400 may have a function of managing the result of the automatic license scoring as a database.

3, an automatic scoring controller 200 according to the present invention includes an AVM receiving unit 210, a vehicle sensor receiving unit 220, a check mark storing unit 230, an interval marker identifying unit 240, A division line identification unit 250, a scoring event processing unit 260, and an automatic scoring control unit 270.

The AVM receiving unit 210 receives the AVM image 115 for the test vehicle M from the AVM image processor 100.

The vehicle sensor receiving unit 220 receives sensor inputs from a plurality of vehicle sensors 105 installed on the test vehicle M. [ For the purpose of obtaining information for scoring the driver's license test results, the test vehicle M is provided with a vehicle sensor 105 for detecting the operation of the parking lever, the degree of acceleration, the operation of the steering wheel, the operation of the turn signal lamp, have. The vehicle sensor receiving section 220 receives sensor inputs from these vehicle sensors 105. [

The check mark storage unit 230 stores the identification information of the division check mark including the interval recognition mark 131 and the course partition lines 132 to 135 for the driving license test course. Preferably, basic information for identifying the division check marks 131 to 132 in the AVM image 115 is stored.

The section marker identifying unit 240 identifies the section marker 131 from the AVM image 115. As described above with reference to FIG. 5, the section recognition flag 131 is used to classify the test courses in the driving license test center. On the basis of this, the automatic grading controller 200 judges whether the test vehicle (M) You can now decide which tasks you need to perform.

The partition line identification unit 250 identifies the course partition lines 132 to 135 from the AVM image 115. As described above with reference to FIG. 5, the course partition lines 132 to 135 are for discriminating whether or not the test vehicle M is out of the course boundary according to the driving license examination course.

The scoring event processing unit 260 identifies whether the test vehicle M is out of the course boundary according to the driving license test course based on the course partition lines 132 to 135 and processes individual scoring corresponding thereto, for example, a scoring event.

The automatic grading control unit 270 identifies the test course based on the interval recognition mark 131 and collects the individual grading results of the grading event processing unit 260 when the mission execution for the corresponding test course is completed, Handles automatic grading for the course. In addition, the automatic grading control unit 270 identifies the position of the test vehicle M on the test course from the AVM image 115 and compares the result of the comparison with the requirement of the sensed input of the vehicle sensor 105, Reflect. For example, unlocking the parking lever and fastening the seatbelt must be performed at the beginning of the test course, and the turn signal lamp should also be operated at the corresponding point, so that the position of the test vehicle (M) There is a need.

7 is a flowchart showing the entire process of implementing driving license automatic scoring based on the surround view image 115 according to the present invention. ) In accordance with an embodiment of the present invention.

First, referring to FIG. 7, the overall process of automatic driver's license scoring according to the present invention will be described.

Step S110: The automatic grading controller 200 stores the identification information of the division check marks 131 to 135 for the driving license test course. The automatic grading controller 200 must recognize the segment recognition marks 131 and the course segment lines 132 to 135 on the AVM image 115 displayed on each test course of the driving license examination center so that the division check marks 131 to 135 ). ≪ / RTI >

Step S120: The AVM image processor 100 performs image processing of the photographed image obtained from the plurality of cameras 101 to 104 provided in the test vehicle M, thereby generating an AVM image 115). The AVM image processor 100 provides the AVM image 115 to the automatic scoring controller 200.

Step S130: The automatic grading controller 200 receives a sensing input from a plurality of vehicle sensors 105 installed on the test vehicle M. In order to assist the driver's license test, the test vehicle M is provided with a vehicle sensor 105 for detecting the operation of the parking lever, the degree of acceleration, the operation of the handle, the operation of the turn signal lamp, 200 are provided with sensing inputs from these vehicle sensors 105.

Step S 140: The automatic grading controller 200 performs the automatic grading of the driving license test reflecting the analysis result of the AVM image 115 based on the identification information of the division check marks 131 to 135 and the sensing input. A specific embodiment of step S140 will be described later with reference to FIG.

Next, an automatic scoring process performed by the automatic scoring controller 200 according to the present invention will be described with reference to FIG.

Step S141: The AVM video 115 identifies the section recognition mark 131 to identify the test course. 5, a section recognition flag 131, for example, "T1" is displayed in order to distinguish the test course from the driving license examination center. The automatic grading controller 200 receives the interval recognition mark 131). By identifying the interval recognition tag 131, the autograghing controller 200 can determine the corresponding test course and, preferably, identify that the test for that test course is initiated.

Step S142: The course partition lines 132 to 135 are identified in the AVM image 115 to identify whether or not the test vehicle M is out of the course boundary according to the driving license test course and process the score accordingly. 5, the course partition lines 132 to 135 are displayed in order to identify whether the test vehicle M is out of the course boundary while performing the mission according to the driving license test course, and the automatic score controller 200 Identifies the course partition lines 132 to 135 from the AVM image 115. [ It is possible to identify whether or not the test vehicle M has departed from the course boundary according to the identification result of the course partition lines 132 to 135 and to process individual scoring (e.g., scoring) events accordingly.

Step S143: The AVM image 115 is used to identify the position of the test vehicle M on the test course and to compare the requirement with the sensed input of the vehicle sensor 105 to determine whether there is a deduction or an emergency dropout Information.

Step S144: As the mission execution for the test course is completed, the automatic scoring controller 200 collects the individual scoring result and the contrast result to process the automatic scoring for the corresponding test course.

Meanwhile, the present invention can be embodied in computer-readable code on a computer-readable recording medium. The recording medium may be any type of recording device readable by a computer such as a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, ). ≪ / RTI > Also, the recording medium may be stored and executed in a distributed fashion on a network.

As described above, the embodiments of the present invention have been disclosed in the present specification and drawings, and although specific terms have been used, they have been used only in a general sense to easily describe the technical contents of the present invention and to facilitate understanding of the invention, It is not intended to limit the scope of the invention. It will be apparent to those skilled in the art that other modifications based on the technical idea of the present invention are possible in addition to the embodiments disclosed herein.

For example, although the technical idea of the present invention has been described in the specification of the present invention on the assumption of a general driver's license test, various embodiments of the present invention can be variously developed according to the development of the technology. For example, in the case of an unmanned automobile, which has recently attracted attention, a test for checking whether the running algorithm of an unmanned vehicle has been properly implemented can be configured in a form similar to a driver's license test. According to the present invention, an adequacy test of an unmanned vehicle can be implemented based on an AVM image. Thus, the present invention should be construed to be variously implementable without being limited to the form of the existing driver's license test.

100: AVM image processor
101 ~ 104: Camera
105: Vehicle sensor
111 ~ 114: Photographed video
115: AVM video
131: interval recognition mark
121, 132 to 135: course partition line
200: Automatic scoring controller
210: AVM receiver
220: vehicle sensor receiver
230: Checkmark storage unit
240: section marker identification unit
250:
260: Scoring event processor
270: automatic grading control section
300: Scoring monitor
400: Scoring management computer

Claims (8)

An AVM image processor (100) for generating an AVM image for a test vehicle by image processing a photographed image acquired from a plurality of cameras installed on a test vehicle;
An automatic scoring controller (200) for storing identification information of a division check mark for a driver's license test course and analyzing the AVM image based on the identification information of the division check mark to perform automatic scoring of a driver's license test;
A driver 's license auto - scoring system using an auroral view image.
The method according to claim 1,
The automatic grading controller (200)
An AVM receiver 210 for receiving an AVM image for a test vehicle from the AVM image processor;
A check mark storage unit 230 for storing identification information of a division check mark including a section recognition mark and a course partition line for the driver's license test course;
An interval marker identifying unit 240 for identifying the interval marker in the AVM image;
A partition line identification unit (250) for identifying the course partition line in the AVM image;
A scoring event processing unit (260) for identifying whether the test vehicle is out of the course boundary according to the driving license test course based on the course partition line and processing individual scoring corresponding thereto;
An automatic grading control unit (270) for identifying a test course based on the segment recognition mark and collecting individual grading results of the grading event processing unit to process an automatic grading for the identified test course;
And a driving license automatic scoring system using the surround view image.
The method of claim 2,
The automatic grading controller (200)
A vehicle sensor receiver 220 for receiving sensor inputs from a plurality of vehicle sensors installed on a test vehicle;
Further comprising:
The automatic grading control unit 270 identifies the position of the test vehicle in the test course from the AVM image and reflects the result of comparing the requirement with the sensing input of the vehicle sensor in the test course scoring Driver's License Scoring System Using Around View Image.
The method of claim 3,
A scoring monitor (300) for displaying an automatic scoring result by the automatic scoring controller;
A scoring management computer (400) of a control room for wirelessly receiving an automatic scoring result by the automatic scoring controller and real-time monitoring of a test progress status of the test vehicle;
And a driving license automatic scoring system using the surround view image.
A first step of storing identification information of a division check mark for a driver's license test course;
A second step of generating an AVM image around the test vehicle by image processing an image captured from a plurality of cameras installed on the test vehicle;
A third step of performing automatic scoring of the driver's license test by analyzing the AVM image based on the identification information of the division check mark;
A driver's license auto-scoring method using an arousal view image including a driver's license.
The method of claim 5,
The division check mark includes an interval recognition mark and a course partition line for the driving license test course,
In the third step,
Identifying a test course by identifying an interval recognition mark on the AVM image;
Identifying a course segment line in the AVM image to identify whether the test vehicle is out of the course boundary according to the driving license test course and processing the individual scoring corresponding thereto;
Collecting the individual scoring results to process the automatic scoring for the identified test course;
The method of claim 1 or 2,
The method of claim 6,
The method of claim 1, further comprising:
Receiving a sensing input from a plurality of vehicle sensors installed on a test vehicle;
Further comprising:
In the third step,
Identifying a location of the test vehicle on the test course from the AVM image and reflecting the result of comparing the requirement with the sensing input of the vehicle sensor in the test course scoring;
The method of claim 1, further comprising the steps of:
A computer-readable recording medium having recorded thereon a program for causing a computer to execute a driving license automatic scoring method using an approach view image according to any one of claims 5 to 7.

Images