KR102633425B1 - Apparatus for creating a traffic accident site reconstruction report - Google Patents

Abstract

A device for creating a traffic accident scene reconstruction report that ultimately generates a traffic accident scene reconstruction report using sequential images, GPS information, sensor information, and input data from a window editing unit is disclosed. The apparatus for creating a traffic accident scene reconstruction report of the present invention includes a data collection terminal that collects data including sequential images, GPS information, and sensor information at the moment of the accident; and an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal, wherein the accident restoration device includes: a window editing unit that inputs editing input values into the accident restoration device; A video generator that displays accident images edited from the sequential images; a target vehicle route generator that generates a route of the target vehicle from the GPS information; a background road creation unit that generates a background screen of the traffic accident scene through the window editing unit; and a moving object creation unit that creates a moving object from sequential images through the window editing unit, and generates a traffic accident scene reconstruction report including a video screen and a real-time top view. The report can help you precisely analyze the cause of the accident by expressing the scene of the traffic accident in detail.

Description

Apparatus for creating a traffic accident site reconstruction report}

This relates to a device that creates a traffic accident scene reconstruction report. More specifically, the accident restoration device is a traffic accident scene reconstruction device that ultimately generates a traffic accident scene reconstruction report through sequential images, GPS information, sensor information, and input data from the window editor. This relates to a device that creates a reconstruction report.

When a traffic accident occurs, there are many situations in which it is very ambiguous to determine the at-fault vehicle and the victim vehicle, and disputes continue to arise as a result.

Recently, as the spread of automobiles has increased and the number of casualties due to accidents has increased, black boxes that have been used in aviation are being used in vehicles as well. Vehicle black boxes allow the cause of traffic accidents to be clearly determined by analyzing data related to the vehicle's speed, direction, and brake operation.

These vehicle black boxes film the situation at the time of the accident with a camera installed on the front or rear of the vehicle, and capture all surrounding sounds with a microphone and store them in memory.

However, there is a problem in that it is difficult to accurately determine the path between the actual driving vehicle and the vehicle ahead at a glance. In other words, there is a need for a means to accurately determine the cause of the accident by restoring the overall path.

In Korean Registration No. 10-2433544, “Vehicle path restoration system through sequential image analysis and vehicle path restoration method using the same,” the applicant can restore the surrounding situation, such as the moment of a traffic accident, through sequential image analysis and express it as a top view. An invention is being disclosed.

In addition, when an actual accident occurs, there is a demand for devices and methods that can help precisely analyze the cause of the accident by providing a traffic accident scene reconstruction report and expressing the scene of the accident in detail.

Republic of Korea Patent No. 10-1342124 “Front vehicle recognition and tracking system using video and vehicle recognition and tracking method using the same” Republic of Korea Patent No. 10-1455835 “Lane recognition and tracking system using video, lane recognition and tracking method using the same” Republic of Korea Patent No. 10-1473866 “Vehicle image processing system and image processing method using the same” Republic of Korea Patent No. 10-2296520 “Curved lane detection method by path estimation using a monocular camera” Republic of Korea Patent No. 10-2318586 “Method for recognizing separation zones and predicting collision risk through video analysis” Republic of Korea Registration No. 10-2433544 “Vehicle path restoration system through sequential image analysis and vehicle path restoration method using the same”

The present invention was developed in consideration of the above-mentioned problems, and its purpose is to reconstruct the accident scene by using sequential images, GPS information, sensor information, and input data from the window editing unit to ultimately generate a traffic accident scene reconstruction report. It provides a device for creating reports.

The device for creating a traffic accident scene reconstruction report of the present invention for solving the above problems includes a data collection terminal that collects sequential images at the moment of the accident and data that is GPS information; and an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal, wherein the accident restoration device includes: a window editing unit that inputs editing input values into the accident restoration device; A video generator that displays accident images edited from the sequential images; a target vehicle route generator that generates a route of the target vehicle from the GPS information; a background road creation unit that generates a background screen of the traffic accident scene through the window editing unit; and a moving object creation unit that creates a moving object from sequential images through the window editing unit, and generates a traffic accident scene reconstruction report including a video screen and a real-time top view.

In the present invention, the data collection terminal further acquires sensor information, and the accident recovery device includes an impulse data generator that generates an impulse from the sensor information; A speed data generator that generates a speed change of the accident vehicle from the GPS information; And it may further include a rotation direction graph generator that generates the rotation direction of the accident vehicle from the sensor information.

In the present invention, the traffic accident reconstruction report displays a video screen, a real-time top view, a real-time still image, a speed graph, an impulse graph, and a rotation direction graph, so that the time and space at the moment of the accident can be easily understood.

In the present invention, the editing screen of the traffic accident scene reconstruction report performed in the window editing unit may display a video interface, a top view interface, a traffic accident scene editing screen, a specific location screen, and an original video.

The device for creating a traffic accident scene reconstruction report of the present invention for solving the above other problems includes a data collection terminal that collects data that is sequential images and GPS information at the moment of the accident; and an accident restoration device that creates a traffic accident scene reconstruction report from data information collected by the data collection terminal, wherein the accident restoration device includes: a video generator that displays accident images edited from the sequential images; and a target vehicle path generator that generates the path of the target vehicle from the GPS information, wherein the target vehicle path generator calculates the distance traveled in latitude and longitude per unit time using GPS latitude and longitude coordinates and moves per unit time. A vector is formed, the components of the movement vector per unit time are accumulated to obtain the actual path of the target vehicle, and the actual path of the target vehicle is rotated by the difference between the world coordinate system and the latitude and longitude coordinate system to generate the rotation path of the target vehicle.

The device for creating a traffic accident scene reconstruction report of the present invention to solve the above problem includes a data collection terminal that collects data that is sequential images and GPS information at the moment of the accident; and an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal, wherein the accident restoration device includes: a window editing unit that inputs editing input values into the accident restoration device; A video generator that displays accident images edited from the sequential images; and a background road creation unit that generates a background screen of the traffic accident scene through the window editing unit, wherein the background road creation unit meets at point C, depending on the type of the background road, and can each rotate around point C. An angle that forms two virtual center lines and sets the coordinates of the end points of each center line Obtain the coordinates rotated as much as the coordinates of the end points of each center line, and the rotation angle , width of the roadway Find the coordinates of the end points of 2n outer lanes using w, which is 1/2 of , and the coordinates and rotation angle of the starting point of each center line. , width of the roadway Obtain the coordinates of the starting points of the 2n outer lanes using w, which is 1/2 of The shape of the background road can be created by finding n intersections where adjacent outer lanes meet.

In the present invention, the area of the roadway is the number of lanes (N1, N2) and the width of the lane Since it changes depending on the can be obtained from the number of pixels.

The device for creating a traffic accident scene reconstruction report of the present invention to solve the above problem includes a data collection terminal that collects data that is sequential images and GPS information at the moment of the accident; and an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal, wherein the accident restoration device includes: a window editing unit that inputs editing input values into the accident restoration device; A video generator that displays accident images edited from the sequential images; and a moving object creation unit that creates a moving object from the sequential video through the window editing unit, wherein the moving object creation unit selects several key frames from n frames of the sequential video and moves the moving object on the traffic accident scene editing screen. Enter the coordinates of the object, the keyframes of the moving object are synchronized with sequential video frames, and the coordinates between keyframes are automatically created at regular intervals. The keyframes of the moving object are indicated with frame numbers on the editing screen. You can.

According to the present invention having the above-described configuration, the accident reconstruction device ultimately generates a traffic accident scene reconstruction report through sequential images, GPS information, sensor information, and input data from the window editing unit. In the scene reconstruction report, the video of the accident moment, top view video, speed graph, impact data graph, and gyro sensor graph are played in real time in one frame unit, and all data is displayed in synchronization. The report can help you precisely analyze the cause of the accident by expressing the scene of the traffic accident in detail.

Figure 1 shows the structure of an apparatus for creating a traffic accident scene reconstruction report according to an embodiment of the present invention.
Figure 2 shows a traffic accident scene reconstruction report according to an embodiment of the present invention.
Figure 3 shows an editing screen of an accident reconstruction device that creates a traffic accident reconstruction report according to an embodiment of the present invention.
Figure 4 is a flowchart explaining a method of creating a traffic accident scene reconstruction report according to an embodiment of the present invention.
Figure 5 is a flowchart showing a method of forming a traffic accident scene reconstruction report according to an embodiment of the present invention.
Figure 6 shows a world coordinate system representing the movement of a target vehicle according to an embodiment of the present invention.
Figure 7 is a diagram showing a movement vector of a target vehicle according to an embodiment of the present invention.
Figure 8 is a diagram showing a target vehicle path according to an embodiment of the present invention.
Figure 9 is a diagram showing a target vehicle rotation path according to an embodiment of the present invention.
Figure 10 is a diagram showing vehicle icons arranged on a target vehicle rotation path according to an embodiment of the present invention.
Figure 11 is an interface diagram for selecting the [Road] tab of the editing screen of Figure 3 according to an embodiment of the present invention.
Figure 12 shows the basic structure of a range according to an embodiment of the present invention.
FIG. 13 is a diagram illustrating a method of creating an outer lane of the left roadway of FIG. 11 according to an embodiment of the present invention.
Figure 14 is a diagram illustrating a method for finding an intersection in two exterior lanes according to an embodiment of the present invention.
Figure 15 is a diagram showing the intersection of intersections according to an embodiment of the present invention.
Figure 16 is a diagram showing an example of an intersection drawn in an accident restoration system according to an embodiment of the present invention.
17 to 19 are diagrams showing a method of creating a path for a moving object according to an embodiment of the present invention.
Figure 20 shows a data structure for defining properties of a moving object using keyframes according to an embodiment of the present invention.
Figure 21 is a diagram showing the three axes of an acceleration sensor installed in a terminal according to an embodiment of the present invention.
Figure 22 shows an example of a graph of the results of calculating the impact amount by applying Equation 12 according to an embodiment of the present invention.
Figure 23 geometrically shows the process of calculating the sum vector of three vectors according to an embodiment of the present invention.
Figure 24 is a diagram showing a graph of the average vector sum according to an embodiment of the present invention.

Hereinafter, an apparatus for creating a traffic accident scene reconstruction report according to a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

The present invention relates to a device that creates a traffic accident scene reconstruction report by analyzing various data obtained when a traffic accident occurs in a vehicle equipped with a terminal.

Figure 1 shows the structure of an apparatus for creating a traffic accident scene reconstruction report according to an embodiment of the present invention. Figure 2 shows a traffic accident scene reconstruction report according to an embodiment of the present invention. Figure 3 shows an editing screen of an accident reconstruction device that creates a traffic accident reconstruction report according to an embodiment of the present invention.

1 to 3, the device 10 for creating a traffic accident scene reconstruction report 300 includes a data collection terminal 100 that collects data at the moment of the accident and data information collected by the data collection terminal 100. It includes an accident restoration device (200) that creates a traffic accident scene reconstruction report (300) from. The data collection terminal 100 and the accident recovery device 200, and their respective components may be separated or formed integrally.

The data collection terminal 100 includes an image capture unit 110 for acquiring sequential images 111, a GPS 120 for acquiring GPS information 121, and a sensor module 130 for acquiring sensor information 131. Includes.

The image capture unit 110 acquires sequential images from the front camera installed on the subject vehicle, and the sequential image (111) information is the image obtained from the front camera, etc. to store the moment of the accident. It is used for this purpose. In the present invention, the target vehicle is an accident vehicle, and is usually displayed as a target vehicle until an accident occurs.

The GPS 120 acquires GPS information 121, and the GPS information 121 includes information such as time, latitude, and longitude. GPS information 121 is synchronized with the time of the sequential video 111.

The sensor module 130 is composed of an acceleration sensor, a gyro sensor, etc., and the sensor information 131 refers to a set of data obtained from the sensor module 130 and is synchronized with the time of the sequential image 111.

Meanwhile, the accident restoration device 200 processes the input data of the sequential images 111, GPS information 121, and sensor information 131 of the data collection terminal 100 to easily understand the time and space at the moment of the accident. Create a real-time top view (Top View, 320). (see Figure 2)

The accident restoration device 200 is a device that forms a traffic accident scene reconstruction report 300, and includes a window editing unit 220 that inputs editing input values to the accident restoration device 200 and an accident image edited from the sequential video 111. A background road is created to create a background screen of the traffic accident scene through a video generator 211 that shows, a target vehicle path generator 212 that generates the path of the target vehicle from GPS information 121, and a window editor 220. Unit 213, a moving object generator 214 that creates a moving object from the sequential image 111 through the window editing unit 220, an impulse data generator 215 that generates an impulse from the sensor information 131, and GPS A speed data generator 216 that generates the accident vehicle speed change from the information 121, a rotation direction graph generator 217 that generates the accident vehicle rotation direction from the sensor information 121, and a window editing screen 400. It includes a display unit 230 that displays the traffic accident scene reconstruction report 300 in the form of a top view.

The accident reconstruction device 200 finally generates a traffic accident scene reconstruction report 300 through sequential images 111, GPS information 121, sensor information 131, and input data from the window editing unit 220.

The traffic accident scene reconstruction report 300 displays the subject vehicle and the collision vehicle as icons in a real-time top view, moves in real time along a designated path, and shows the movement up to the collision situation. In the present invention, a collision vehicle is one type of moving object, and moving objects include light cars, sedans, SUVs, buses, trucks, bicycles, motorcycles, pedestrians, etc.

Additionally, the accident restoration device 200 generates a background of a real-time top view. The main background is the shape of the road, and there are general roads, three-way intersections, and intersections. In order to make the appearance of this road similar to the real one, the background road creation unit 213 creates the shape of the road through geometric calculations, which will be described later. Components of a road include outer lanes, center lines, crosswalks, detailed lanes, stop lines, and angles between lanes.

Additionally, the accident recovery device 200 can display additional elements for determining the situation of an accident on real-time video. Additional factors include traffic lights, signs, vehicle speed, relative distance, and time to collision (TTC).

Additionally, the accident recovery system 200 can calculate sensor data obtained from the data collection terminal 100 and display it in graphs.

Referring to Figure 2, as an embodiment of the traffic accident reconstruction report 300, a video screen 310, a real-time top view 320, a real-time still image 330, a speed graph 340, an impact graph 350, and a rotation direction graph 360 are displayed, and in addition, road conditions including the date and time of the accident, vehicle number, driver, accident location, lane and traffic lights can be displayed.

The video screen 310 refers to the accident moment generated by the video generator 211 and the original video before and after the accident moment. For example, a video with a total length of 10 seconds is a sequential video of 5 seconds before the accident and 5 seconds after the accident.

The top view 320 shows a real-time top view generated from the target vehicle path creation unit 212, the background road creation unit 213, and the moving object creation unit 214. In the embodiment, the target vehicle (yellow SUV) moves along a path on the background (intersection), and the collision vehicle (red compact car) also moves along the path. The video screen 310 and the real-time top view 320 are synchronized, and the collision time of the video screen 310 and the collision time of the real-time top view 320 coincide.

The real-time still image 330 refers to a sequential still image of the real-time top view 320. The real-time still screen 330 simplifies the movement of vehicles at the moment of the accident and shows, for example, data on vehicles from 5 seconds before the accident, separated by intervals of about 1 second.

The speed graph 340 shows the speed of the target vehicle obtained from the GPS 120 as a graph.

The impact amount graph 350 shows the impact amount of the target vehicle as a graph obtained by processing sensor information obtained using an acceleration sensor.

The rotation direction graph 360 shows the rotation direction of the target vehicle obtained using a gyro sensor as a graph.

Referring to Figure 3, as an embodiment of the editing screen 400 of the traffic accident scene reconstruction report performed in the window editing unit 220, the video interface 410, the top view interface 420, and the traffic accident scene editing screen 430 , the specific location screen 440 and the original video 450 are displayed.

The video interface 410 is a user interface that loads the original video screen 450 and the path of the target vehicle and controls the video.

The top view interface 420 creates various elements to be displayed on the top view and displays them as a preview.

The traffic accident scene editing screen 430 is a window screen for [Accident Scene Editing]. This is an editing window that displays all objects (roads, crosswalks, traffic lights, signs, cars, text, videos, etc.) to create a real-time top view. Here, the movement of vehicle icons can be synchronized with the original video (450). there is. In other words, it performs a similar function to a video editor.

The top view interface 420 has tabs that perform multiple functions. At the top of the top view interface 420, menu tabs for roads, crosswalks, traffic lights, signs, cars, text, and videos can be seen.

The [Road] and [Crosswalk] tabs create the appearance of the road that becomes the background of the top view. These types of roads include general roads, three-way intersections, and intersections. In the case of three-way intersections and intersections, the angle of each roadway can be adjusted, and center lines, detailed lanes, stop lines, and crosswalks can be displayed. Additionally, it is possible to express curves at locations where outer lanes of different roadways meet.

The [Traffic Light] tab is an important element of an accident situation, and it has several properties so you can quickly and accurately create the appearance of a traffic light. You can select and create driving traffic lights, pedestrian traffic lights, 2-way, 3-way, and 4-way traffic lights.

The [Signs] tab can be created by selecting the speed limit.

[Road], [Traffic light], and [Sign], which are the contents described above, can be created in the top view interface 420 of the Windows editing screen 400 and viewed as a preview, and the traffic accident scene can be edited by dragging the mouse along the path. It can be displayed at a specific location in the window 430. The unexplained reference numeral '440' is a specific location screen.

The [Car] tab sets the type and properties of the car icon and displays it as a preview. This preview icon can be displayed at a desired location on the traffic accident scene editing screen (430) by dragging it with the mouse. Types of car icons include compact cars, sedans, SUVs, buses, trucks, bicycles, motorcycles, and pedestrians.

In the [Text] tab, you can enter the desired text in the desired location. For example, accident time, vehicle information, driver's personal information, accident location information, etc.

Figure 4 is a flowchart explaining a method of creating a traffic accident scene reconstruction report according to an embodiment of the present invention.

Referring to FIG. 4, the data collection terminal 100 acquires sequential images 111, GPS information 121, and sensor information 131 (S101).

Next, the accident restoration device 200 finally generates a traffic accident scene reconstruction report 300 through sequential images 111, GPS information 121, sensor information 131, and window editing unit 220 (S102) )do.

Below, we will explain how to form a traffic accident scene reconstruction report.

Figure 5 is a flow chart showing a method of forming a traffic accident scene reconstruction report according to an embodiment of the present invention.

Referring to FIG. 5, the accident restoration device 200 generates the path of the target vehicle using the latitude and longitude of GPS (S201).

Next, the accident restoration device 200 generates the shape of the background road using a geometric method (S202).

Next, the accident restoration device 200 creates a moving object (S203).

Next, the accident restoration device 200 generates impact data of the accident vehicle (target vehicle) (S204).

Next, the accident restoration device 200 displays the path of the target vehicle, the background road, and the moving object as a top view and synchronizes it with the accident image to generate a traffic accident scene reconstruction report (S205). In addition, the traffic accident scene reconstruction report can display a graph of the speed change of the accident vehicle, a graph of the impact force of the accident vehicle, and a graph of the direction of rotation of the accident vehicle.

Below, we will explain a specific method of creating a traffic accident scene reconstruction report.

(How to create a route for a target vehicle using latitude and longitude)

This section explains the process of converting coordinates that can be located on a top view using the latitude and longitude obtained from the GPS of the target vehicle. (Reference: https://en.wikipedia.org/wiki/Longitude)

Latitude and longitude are angular values, and the units for two-dimensional top views are metric. The present invention deals with the path of a car at the moment of an accident and displays the distance traveled in a short period of time, such as 10 seconds.

Figure 6 shows a world coordinate system representing the movement of a target vehicle according to an embodiment of the present invention. Compared to the size of the Earth, the distance traveled by vehicles in a short period of time is very small, so even if it is assumed to be flat, the error is small. Therefore, it is assumed that the coordinate system of the latitude and longitude of the GPS and the coordinate system of the top view of the accident recovery device have the same directionality as shown in FIG. 6.

Referring to FIG. 6, N, S, E, W (North, South, East, West) refer to the direction of the Earth. X, Y refer to the axes of the world coordinate system. The Y axis means north and south, and the X axis means east and west.

If two different axes are mapped into one as shown in Figure 6, the movement of the vehicle can be easily expressed on a plane using latitude and longitude coordinates. In Figure 6, the X-axis refers to the longitude direction, and the Y-axis refers to the latitude direction. A and B represent the movement of the target vehicle location as a vector over time.

The length of longitude (east-west distance) depends on the radius of the circle of latitude. Therefore, the ratio varies between 0 and 1 depending on latitude. For a sphere with radius a, the latitude The radius is cos and the length of 1 degree of longitude ( ) is as shown below.

If the Earth is modeled as an ellipsoid, the arc length in Equation 1 changes as shown in Equation 2 below.

Here, the eccentricity e of the ellipsoid can be calculated by inputting the major axis a (radius of the equator) and b (radius of the poles) as shown in Equation 3.

Here, according to the WGS84 standard (World Geodetic System 1984), a = 6378137m and b = 6356752.314245m. is 1 at the equator and decreases to 0 at the poles. This means that the circle of latitude shrinks from the equator to the poles, and the length of the longitude decreases as well. This contrasts with the small increase (1%) in latitudinal length from equator to pole.

As an example, if the latitude value is 37.68972, applying Equation 2, can be obtained. This means that the actual length of 1 degree of longitude is about 88 km.

The following explains how to find the length of latitude.

Compared to longitude, the difference in latitude depending on the value of latitude is not large, but since the Earth is an ellipsoid, differences do exist. The formula for calculating the distance according to latitude is shown in Equation 4 below.

As an example, if the latitude value is 37.68972, applying equation 4, can be obtained. The length of 1 degree of latitude is approximately 111 km.

Result of Equation 2 is called the length of the arc of the unit angle of longitude. Result of Equation 4 is called the length of the arc of the unit angle of latitude.

The latitude and longitude data of the target vehicle are stored for the relevant time. means the latitude data for time i, means longitude data for time i. For example, if the time interval of latitude and longitude data is 1 second, is the latitude value that the target vehicle moved for 1 second, is the longitude value at which the target vehicle moved for 1 second. Using the above information, the movement vector of the target vehicle The equation for calculating the vector components is as shown in equation 5 below.

In equation 5: is the movement vector of the target vehicle It means the component of the X axis. is the movement vector of the target vehicle It means the component of the Y axis. means the distance the target vehicle moves on the X-axis during a certain period of time, means the distance the target vehicle moves on the Y axis during a certain period of time.

The above contents are explained by examples.

The difference between two adjacent latitudes is as follows.

37.689702 - 37.689697 = 4.9999999944816409e-06

In this calm By multiplying the value (110990.61299814848), the distance traveled in the latitude direction can be obtained as follows.

4.9999999944816409e-06 * 110990.61299814848 = 0.55495306437825631(m)

According to the above equation, the target vehicle moved approximately 0.55 m (55 cm) per second in the latitude direction.

The difference between two adjacent longitudes is as follows.

126.59813600000000- 126.59815500000001= -1.9000000008873030e-05

In this calm By multiplying the value (88201.262859806971), the distance traveled in the longitude direction can be obtained as follows.

-1.9000000008873030e-05 * 88201.262859806971= -1.6758239951189449(m)

According to the above equation, the target vehicle moved approximately -1.68 m (-168 cm) per second in the longitude direction.

Figure 7 is a diagram showing a movement vector of a target vehicle according to an embodiment of the present invention.

Referring to Figure 7, it shows that the vector of the target vehicle obtained in the above example is displayed in the latitude and longitude world coordinate system. In a two-dimensional coordinate system, the target vehicle moved from A(0, 0) to B(-1.68m, 0.55m) for 1 second.

The path coordinates of the target vehicle are the movement vector of the target vehicle. It can be calculated by accumulating the components. Equation 6 shows how to obtain path coordinates.

In equation 6, P is the path coordinate means a set of i means the number of the path coordinate, and is a value that increases sequentially over time. is the movement vector in Equation 5.

Figure 8 is a diagram showing a target vehicle path according to an embodiment of the present invention.

Referring to FIG. 8, the path coordinates of the target vehicle can be obtained using the method of Equation 6. This is the actual path of the target vehicle shown in Figure 2. In Figure 8, blue dots indicate the location where the target vehicle exists at each time zone.

Now, we will explain how to rotate and display the entire path of the target vehicle. Since the entire path rotates around the origin in FIG. 8, Equation 7 is applied.

In equation 7: is the path angle It means a set of path coordinates rotated by

Figure 9 is a diagram showing a target vehicle rotation path according to an embodiment of the present invention.

Referring to Figure 9, it shows the path of the target vehicle rotated using Equation 7. The blue dots connected by a solid red line are the rotation path coordinates. At this time, the latitude and longitude coordinate system (yellow arrow) and the world coordinate system (green arrow) are set by rotating around the origin.

Figure 10 is a diagram showing vehicle icons arranged on a target vehicle rotation path according to an embodiment of the present invention.

Referring to Figure 10, it shows a vehicle icon placed on the path in Figure 9.

Up to this point, we have explained how to convert GPS latitude and longitude data into world coordinates and generate the target vehicle's path coordinates on the top view.

(How to create the shape of a road using geometric methods)

This section explains how to create the shape of a road using geometric methods.

The top view 320 of FIG. 3 shows a crossroad shape similar to an actual map. This is a picture created directly by the accident restoration device 200 based on input values from the window editing unit 220. Input values include the type of background road, the angle of each roadway, the number of detailed lanes, whether there is a stop line, whether there is a crosswalk, etc., and the [Road] and [Crosswalk] tabs of the top view interface 420 of FIG. 3 described above. You can use to create the appearance of the road that becomes the background of the top view.

Figure 11 is an interface diagram for selecting the [Road] tab of the editing screen of Figure 3 according to an embodiment of the present invention.

Referring to Figure 11, the appearance of a road can be created by selecting the [Road] tab of the editing screen in Figure 3.

Types of roads include general roads, three-way intersections, and intersections. In the case of three-way intersections and intersections, the angle of each roadway can be adjusted, and center lines, detailed lanes, stop lines, and crosswalks can be displayed. Additionally, it is possible to express curves at locations where outer lanes of different roadways meet.

Hereinafter, as an example, a method of creating an intersection in the shape of a road will be described.

Figure 12 shows the basic structure of a range according to an embodiment of the present invention.

Referring to Figure 12, first, four virtual center lines representing the four roadways that make up the intersection are created. The four center lines represented by dotted lines in Figure 11 are called L1, L2, L3, and L4. These four center lines meet at point C, and each can rotate around point C. The initial coordinates of the end points of L1, L2, L3, and L4 are set to default values by the interface (Figure 11).

The equation for rotating the center line is as follows: For example, if the coordinates of the end point of L2 are (x,y), the angle set using Equation 8 You can get the coordinates rotated by that amount.

here, are the coordinates of point C. r is the distance between the coordinates of point C and the endpoint of L2.

The following explains how to create an outer lane. In Figure 8, the eight line segments (b1, b2, b3, b4, b5, b6, b7, b8) that make up the outer boundary of the intersection are called outer lanes. The location of these outer segments is determined by the width of each roadway and the angle of the center line.

below. This explains the process of finding the outer lane of the left roadway.

FIG. 13 is a diagram illustrating a method of creating an outer lane of the left roadway of FIG. 12 according to an embodiment of the present invention.

Referring to Figure 13, the end point of the center line L2 of the left roadway is at an angle with the origin C as the center. It rotated as much as possible, and the coordinates could be obtained using Equation 12. If the end point of L2 is (x,y), the coordinates of B4 Since meets the vertical line passing through the end point of L2, its coordinates can be obtained using Equation 9.

Here, w is a value of 1/2 of the area of the left roadway.

Coordinates of B3 Since exists in the opposite direction of B4, it can be obtained using Equation 10.

Through the above method, the end points of the outer lanes B3 and B4 of the left roadway were obtained. coordinates of is the origin instead of the endpoint (x,y) of L2 in Equation 13 You can get it by entering . coordinates of is the origin instead of the endpoint (x,y) of L2 in Equation 14 You can get it by entering . Now that we have found the starting and ending points of the line segments b3 and b4, we can also find the equation of the straight line. The above process is performed equally for all roadways at the intersection to obtain the coordinates of all outer lanes.

Figure 14 is a diagram illustrating a method for finding an intersection in two exterior lanes according to an embodiment of the present invention.

Referring to FIG. 14, the intersection point where two outer lanes meet is found in the line segment of the outer lane described above. The intersection point where line segments b4 and b5 meet is called J2. You can find the equation of the straight lines of b4 and b5, and use the formula to find the intersection of the equations of the straight lines to find the intersection point J2.

Figure 15 is a diagram showing the intersection of intersections according to an embodiment of the present invention.

Referring to Figure 15, it shows how all the intersection points (J1, J2, J3, J4) at the intersection were obtained.

Figure 16 is a diagram showing an example of an intersection drawn in an accident restoration system according to an embodiment of the present invention.

Referring to Figure 16, the width of the roadway is the number of lanes (N1, N2) and the width of the lane changes depending on You can find the area of the roadway using Equation 11.

For example, in Figure 16, in the left roadway, the number of lanes N1 is 2, N2 is 3, and the width of the lane is If this is 25 (pixels), then the width of the roadway is 125 (pixels). The width of the roadway obtained here is used when creating the outer lane described above. w in equations 9 and 10 is the area of the roadway It is 1/2 of .

The detailed lanes, center lines, and stop lines visible inside the roadway in FIG. 16 can be created by applying the method of creating outer lanes described above (see Equations 9 and 10 in FIG. 12). For example, the center line of the left roadway in FIG. 16 has a distance (w) of 12.5 (pixels) below the center line, so by entering this value into Equation 10, the location of the center line can be found.

(How to create a moving object)

Since the present invention shows the moment of a traffic accident as a real-time top view (specific location screen 440 in FIG. 3), it is necessary to control the movement of objects. The method of creating a route for the target vehicle was explained in the previous section. There are cases where it is difficult to automatically determine the exact location of the path of a crashed vehicle (moving object), and the path is created by estimating the path by looking at some aspects seen in the video.

17 to 19 are diagrams showing a method of creating a path for a moving object according to an embodiment of the present invention.

17 to 19, a moving object (eg, a car) is synchronized with a video frame. For example, if the number of sequential images in a video is 300, the number of object coordinates is also 300. The problem is that it is difficult to manually enter all 300 pieces of data. Therefore, you can set some important frames in the moving object and display them in the editing window. These important frames are called keyframes. Coordinates between keyframes can be automatically generated at regular intervals.

Referring to FIG. 17, a slider bar 411 exists inside the video interface 410 of the editing screen 400 of the traffic accident scene reconstruction report of FIG. 3. If the slider bar 411 shown in FIG. 17 is set to 126 frames and a car object is created, a blue color appears in the moving object editing screen 431 of FIG. 17, which is the upper part of the traffic accident scene editing screen 430 of FIG. 3. It points to the red car inside the circle. At this time, frame number 126 is displayed on the right side of the icon in the moving object editing screen 431.

And, reference numeral '451' in FIG. 17 is a video window, and shows the 126th sequential image of the original video.

The moving object editing screen 431 of FIG. 17 shows a portion of the traffic accident scene editing screen 430 of FIG. 3, and three key frames of the collision vehicle are displayed. Frames 126, 147, and 173 are set as keyframes, and here, the position of the vehicle icon, which is a keyframe, can be moved by dragging it with the mouse.

As mentioned above, the synchronization of a moving object with a video is shown in an example in FIGS. 17 to 19.

Referring to FIG. 18, when vehicle icon number 147 is selected with the mouse on the moving object editing screen 431, which is part of the traffic accident scene editing screen 430, the screen is switched as shown in FIG. 18. The position of the slider bar 411 is changed to 147, and the video window 451 can be seen displaying the 147th sequential image of the video as a blue circle.

Referring to FIG. 19, when vehicle icon number 173 is selected with the mouse on the moving object editing screen 431, which is part of the traffic accident scene editing screen 430, the screen is switched as shown in FIG. 19. The position of the slider bar 411 is changed to 173, and the video window 451 can be seen displaying the 173rd sequential image of the video as a blue circle.

Figure 20 shows a data structure for defining properties of a moving object using keyframes according to an embodiment of the present invention.

Referring to Figure 20, the top boxes ( , , ... , ) is a memory storing n keyframe values, and the bottom boxes are memories storing the properties of moving objects for all frames (m). Additionally, the data stored in the keyframe memory is a frame number, which has a structure where an arrow points to the corresponding location in the moving object memory. For example, if there are 4 data points in the keyframe memory, there are also 4 arrows. The bottom box stores the properties (e.g. position, angle, etc.) of the moving object in the frame. The properties of the keyframe are set first, and the data between keyframes is set later.

(How to create impact data of an accident vehicle)

This section explains how to create the impact force graph 350 of the accident vehicle in FIG. 2. Impact data is generated using data from an acceleration sensor installed in the data collection terminal 100 of the target vehicle. The data collection terminal 100 acquires acceleration data for three axes and stores them in memory. Since the data of these three axes has vector characteristics, the average size (norm) of the vector sum is obtained and this is defined as impulse data. The name of this method is called [average vector sum].

Generally, the amount of impact has its maximum value at the moment of a vehicle accident.

Figure 21 is a diagram showing the three axes of an acceleration sensor installed in a terminal according to an embodiment of the present invention.

Referring to FIG. 21, assuming the box is a vehicle, the arrows show the directions of the three axes of the acceleration sensor.

Assuming that the x-axis is the front direction of the vehicle, the y-axis is the lateral direction of the vehicle, and the z-axis is the vertical direction of the vehicle, the axes that greatly affect the amount of impact at the moment of an accident are x and y. According to analysis of actual accidents, the z-axis is the upward and downward movement of the vehicle, so when calculating the amount of impact at the moment of an accident, there is a tendency to add an error to the amount of impact. Therefore, in the present invention, when calculating the impulse, only data corresponding to the two axes (x, y) is considered.

The existing method of obtaining general impulse data uses a method of calculating the size (norm) of the vector as shown in Equation 12.

In Equation 12, x and y refer to the data of each axis of the acceleration sensor, is the size of the vector.

Figure 22 shows an example of a graph of the results of calculating the impact amount by applying Equation 12 according to an embodiment of the present invention.

Referring to FIG. 22, the x-axis is the time axis, the unit is frame, and is set to 30 frames/sec. The y-axis represents the amount of impulse, and the unit is gravitational acceleration ( )am.

The purpose of the present invention is to remove noise from this impulse graph and preserve the point of impact as is. For this purpose, we propose a method to calculate the average size of vector components.

Figure 23 geometrically shows the process of calculating the sum vector of three vectors according to an embodiment of the present invention.

Referring to FIG. 23, the sum of the vectors is the sum of the components, so it is equivalent to Equation 13 below.

In equation 13, is 3 vectors ( , , ) means the sum vector of here, , , am. And, the average size of this vector sum can be obtained using Equation 14.

When inputting multiple data, programming can be done using Equation 19 below.

Here, i refers to the index of acceleration data and is a value that increases sequentially with respect to time. According to Equation 15, it can be seen that it means the sum of the data on the left and right of index i. Although the formula for calculating the sum of these vector components seems simple, it can effectively reduce the noise of the acceleration vector.

For example, if the directions of surrounding vectors are significantly different, the size of the sum vector may be greatly reduced. Additionally, if the directions of neighboring vectors are similar, the size of the sum vector can be kept relatively constant. For this reason, the noise may be reduced, but the value of the actual impulse data may be maintained.

Equation 15 is and The method is to set the window size to 3, focusing on . If you want to further reduce noise, you can increase the window size in Equation 19. When the window size is n, the general formula for [average vector sum] can be expressed as Equation 16.

Figure 24 is a diagram showing a graph of the average vector sum according to an embodiment of the present invention.

Referring to Figure 24, compared to Figure 21, it can be seen that much of the noise has been removed and the maximum impact position (red vertical line) is maintained.

The present invention described above is not limited to the above-described drawings and detailed description, but can be modified and modified in various ways by those skilled in the art without departing from the spirit and scope of the present invention as set forth in the claims below. Of course, it is also included within the scope of the present invention.

10: Device for creating a traffic accident scene reconstruction report
100: data collection terminal 110: video recording unit
111: Sequential video 120: GPS
121: GPS information 130: Sensor module
131: Sensor module 200: Accident restoration device
211: video generation unit 212: target vehicle path creation unit
213: Background road creation unit 214: Moving object creation unit
215: impulse data generation unit 216: speed data generation unit
217: Rotation direction graph creation unit 220: Window editing unit
230: Display unit 300: Traffic accident scene reconstruction report
310: video screen 320: top view
330: Still image 340: Speed graph
350: Impulse graph 360: Rotation direction graph
400: Edit screen of traffic accident scene reconstruction report
410: Video Interface 411: Slider Bar
420: Top view interface 430: Traffic accident scene editing screen
440: Specific location screen 450: Original video

Claims (8)

delete delete delete delete A data collection terminal that collects data, which is sequential video and GPS information at the moment of the accident; and
Including an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal,
The accident restoration device is,
A video generator that displays accident images edited from the sequential images; and
It includes a target vehicle path generator that generates a path for the target vehicle from the GPS information,
The target vehicle route generator,
Using GPS latitude and longitude coordinates, the distance traveled in latitude and longitude per unit time is calculated to form a movement vector per unit time, the components of the movement vector per unit time are accumulated to obtain the actual path of the target vehicle, and the target vehicle's actual path is obtained. A device that creates a traffic accident scene reconstruction report, characterized by generating the rotation path of the target vehicle by rotating the actual path by the difference between the world coordinate system and the latitude and longitude coordinate system. A data collection terminal that collects data, which is sequential video and GPS information at the moment of the accident; and
Including an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal,
The accident restoration device is,
A window editing unit that inputs editing input values into the accident restoration device;
A video generator that displays accident images edited from the sequential images; and
It includes a background road creation unit that generates a background screen of the traffic accident scene through the window editing unit,
The background road generator,
Depending on the type of background road, it meets at point C, forming n virtual center lines that can each rotate around point C, and an angle that sets the coordinates of the end points of each center line. Obtain the coordinates rotated as much as the coordinates of the end points of each center line, and the rotation angle , width of the roadway Find the coordinates of the end points of 2n outer lanes using w, which is 1/2 of , and the coordinates and rotation angle of the starting point of each center line. , width of the roadway Find the coordinates of the starting points of the 2n outer lanes using w, which is 1/2 of A device that creates a traffic accident scene reconstruction report, characterized in that it generates the shape of the background road by finding n intersections where adjacent outer lanes meet. According to clause 6,
The width of the roadway is the number of lanes (N1, N2) and the width of the lane Since it changes depending on the A device that creates a traffic accident scene reconstruction report, which is obtained by calculating the number of pixels.
A data collection terminal that collects data, which is sequential video and GPS information at the moment of the accident; and
Including an accident restoration device that creates a traffic accident scene reconstruction report from the data information collected by the data collection terminal,
The accident restoration device is,
A window editing unit that inputs editing input values into the accident restoration device;
A video generator that displays accident images edited from the sequential images; and
It includes a moving object creation unit that creates a moving object from sequential images through the window editing unit,
The moving object creation unit,
Among the m frames of the sequential video, at least two n keyframes less than m are selected, the coordinates of the moving object are entered in the traffic accident scene editing screen, and the keyframes of the moving object are synchronized with the sequential video frame. A device for creating a traffic accident scene reconstruction report, wherein the coordinates between key frames are automatically created at regular intervals, and the key frames of the moving object are indicated with frame numbers on the editing screen.