KR101484170B1 - Assessment system and method for image projected from head up display - Google Patents

Abstract

A coordinate setting unit for setting a coordinate system for each object, which is an individual element included in the virtual image, and a reference value for determining whether or not the virtual image is deformed, A template storing unit for storing information about each object included in the reference image, and a dual image is determined according to whether the edges of the objects are overlapped with each other by scanning each object. In case of a double image, And a distortion determining unit for determining whether the virtual image is distorted or distorted by comparing the object of the reference image drawn from the template storing unit with the object of the virtual image. As a result, it is possible to prevent distortions and double images through evaluation, so that it is possible to generate a clear virtual image of the image transmitted from the HUD system, thereby improving not only the quality of the HUD system but also the user's satisfaction have.

Description

[0001] HUD IMAGE EVALUATION SYSTEM AND METHOD AND APPARATUS FOR MEASURING THE SAME [0002]

The present invention relates to an HUD image evaluation system and an evaluation method thereof, and more particularly, to a HUD image evaluation system and an evaluation method thereof that enable evaluation of whether a virtual image is generated without distortion by generating a quantitative verification tool .

The Head Up Display (HUD) system was originally introduced on airplanes and started by displaying images related to flight information on the windows and helmets of an airplane for safety during flight.

Recently, a HUD system for a vehicle applied to a vehicle has been developed, as shown in many patents including domestic patent publication No. 1020070049990 "Vehicle Driving Aid Device and Vehicle Having Vehicle Driving Aid Device" Has been used. The vehicle HUD system reflects various information related to the driving of the vehicle on the windshield of the vehicle. The information provided by the vehicle HUD system includes a vehicle speedometer, vehicle speed, navigation, cruise control, Road information, back side warnings, and the like.

1, the HUD system 100 for a vehicle includes an HUD main body 101 for transmitting an image to a dashboard of a vehicle, and an image transmitted from the HUD main body 101 is transmitted through a reflector 102 It is reflected by the windshield window. The actual image is formed on the windshield window, but the driver appears to have the image floating a certain distance from the windshield window and is called a virtual image. For windshields, the virtual image is clearly displayed by inserting an oxide coating (TiO2 / SiO2 composite layer) or a wedge-shaped PVB film with a reflectance different from that of glass.

In this HUD system 100, the driver's side focus is determined according to the thickness and refractive index of the windshield, and when the driver's eyes are positioned at the driver's side, the image is focused and an accurate virtual image can be seen. At this time, since the driver's side focus must be formed within a certain range that allows the driver's seat to move, the focus can be adjusted by moving the reflector 102 and adjusting the vertical distance and position at which the virtual image is displayed.

In the HUD system 100, since the virtual image is generated at a certain distance from the front of the vehicle after the image is reflected on the windshield, even if the hardware for reflecting the image and the software for generating the image have a minor problem, Can be deformed. In addition, deformation of the virtual image may occur due to vehicle speed, vibration, hardware or software malfunction, etc. when the vehicle is stationary and while driving. When the deformed image is displayed in front of the vehicle, the driver may be disturbed by the driver, rather than being assisted by the driver due to an unclear virtual image.

Accordingly, it is necessary to evaluate whether or not the virtual image generated in the HUD system 100 is clearly generated without deformation. By evaluating the virtual image generated in the HUD system 100, it is possible to check the quality of the HUD system 100 when the vehicle is stopped and running, and suggest the hardware and software development of the HUD system 100.

It is an object of the present invention to provide a HUD image evaluation system and an evaluation method thereof, which can evaluate whether a virtual image is clearly generated without deformation by generating a quantitative verification tool.

It is another object of the present invention to provide a HUD image evaluation system and a method of evaluating the HUD image that can provide a clear image by proposing a position of a camera capable of eliminating distortion, .

The object is achieved by a camera for photographing a virtual image formed from a HUD system mounted on a vehicle; A coordinate setting unit for setting a coordinate system for each object that is an individual element included in the virtual image; A template storage unit for storing information on each object included in a reference image that is a reference for determining whether the virtual image is deformed; A dual phase determiner for determining a double phase according to whether the edges of the objects are overlapped or not by scanning each of the objects and for measuring the degree of displacement using the coordinate system set by the coordinate setting unit when the double phase is detected; And a distortion determination unit for determining whether the virtual image is distorted or distorted by comparing an object of the reference image drawn from the template storage unit with an object of the virtual image, Lt; / RTI >

The above object may be achieved by a method comprising: receiving a virtual image formed from a HUD system mounted on a vehicle; A coordinate setting step of setting a coordinate system for each object that is an individual element included in the virtual image; A dual image determining step of scanning each of the objects and determining a double image of the virtual image according to whether the edges of the objects overlap each other; A displacement measuring step of measuring a degree of displacement of the virtual image using the coordinate system when the virtual image is a double phase; And a distortion judging step of judging whether the virtual image is distorted or distorted by comparing each object included in the reference image, which is a reference for judging whether or not the virtual image is deformed, with the object of the virtual image, The HUD image evaluation method according to the present invention can be also achieved by the HUD image evaluation method.

According to the present invention, it is possible to set a position of a camera that can prevent distortion and a double image by providing a tool for measuring the distortion, the double phase, the rotation, and the displacement amount of an image transmitted from the HUD system. In addition, by displaying the results of the rotations externally, the operator can give an opportunity to explore various ways to eliminate the deformation of the virtual image. Therefore, a clear virtual image can be generated for the image transmitted from the HUD system, so that the quality of the HUD system can be improved and the user can be satisfied thereby.

BRIEF DESCRIPTION OF DRAWINGS FIG. 1 is a schematic block diagram of a vehicular HUD system;
FIG. 2 is a block diagram of a system for evaluating an HUD image according to the present invention.
3 (a) and 3 (b) are diagrams showing an example of a virtual image in which an ROI is formed,
4 (a) is an illustration of an object of a reference image for distortion determination,
4 (b) is an illustration of an object of an evaluation target image for distortion determination,
Fig. 4 (c) is an example of an overlapping state between the object of Fig. 4 (a) and the object of Fig. 4 (b)
5 is an illustration of an evaluation target image for double-phase determination,
6 is a flowchart illustrating a process of generating a template in the HUD image evaluation system of the present invention.
FIG. 7 is a flowchart illustrating a process of calculating distances by determining distortions and dual images in the HUD image evaluation system of the present invention.

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

2 is a block diagram of a system for evaluating an HUD image according to the present invention.

The HUD image evaluation system 1 according to the present invention can determine whether a virtual image generated after being transmitted from the HUD system and reflected on the windshield of the vehicle has been transformed. The HUD image evaluation system 1 includes a camera 10, an ROI setting unit 15, a coordinate setting unit 30, a binary image processing unit 20, a color image processing unit 25, a rotation determination unit 35, A distortion determining unit 55, a dual image determining unit 45, a coordinate transforming unit 50, a distance calculating unit 60, and a template storing unit 65.

The camera 10 photographs a virtual image formed by being transmitted from the HUD system, and can transmit the photographed image to the coordinate setting unit 30. [ The camera 10 should be installed in a position corresponding to the eyes of the driver to photograph the virtual image as seen by the driver. For example, the camera 10 may be installed on the ceiling so as to protrude toward the driver's head, or may be installed so that the camera 10 protrudes forward from the headrest of the driver's seat.

The ROI setting unit 15 can receive a virtual image photographed from the camera 10 or an image directly from the HUD system and can form a region of interest (ROI) for a plurality of objects included in the image.

Here, an object is an element constituting a HUD image such as a shape, a character, and a number included in an image transmitted from the HUD system, and can be used to judge deformation of the HUD image. The object may include a vehicle speedometer, vehicle speed, navigation, cruise control, lane departure warning, road information, backward warning, and the like. The ROI is formed in a rectangular frame shape surrounding each object and is a reference for forming coordinate axes and center coordinates necessary for judging deformation of the HUD image.

The ROI setting unit 15 can set an ROI for two types of images.

One is the reference information to be compared so that the deformation of the virtual image of the HUD system can be measured, and the ROI for the unmodified reference image can be set. Information about each object included in the reference image, the coordinate axis of the ROI, and the center coordinate are stored in the template storage unit 65 as template information. In this case, in order to form a template, an unmodified virtual image is required. The virtual image can be obtained by photographing the virtual unimorphized image of the image transmitted from the HUD system with the camera 10, It can also be used.

The other is an image taken from a HUD system and photographed with a virtual image formed on a windshield of a vehicle, and is evaluated as to whether or not there is deformation. Hereinafter, it is referred to as an evaluation target image.

The ROI setting unit 15 forms an ROI for each of the reference image and the evaluation target image, and the ROIs of the reference image and the ROIs of the evaluation target image are matched and matched one to one, It can be judged. That is, each ROI is a region to be compared in order to judge whether or not the virtual image is deformed.

FIG. 3A shows a reference image, FIG. 3B shows an image to be evaluated, and FIG. 3A and FIG. 3B show an operation of the ROI setting unit 15 The following is an explanation.

In order to form the ROI of the reference image and the evaluation target image, the ROI setting unit 15 may first define the shape and character of each object. For example, as shown in FIG. 3A, when the object is formed by connecting a circular shape and an elliptical shape, the ROI setting unit 15 sets the ROI using the shape of the object including the original shape and the color information of the object You can define an object, or you can define an object by saving the image of the object as a file. As another example, when the object is a number having the same speed as the vehicle speed, the coordinate setting unit 30 can define the object as a number. In addition, the object including the font size, font type, color information, . The color information may include red, green, blue, Hue, Saturation, Intensity, Value, Luminance, and the like. Here, the reason for extracting information on various colors such as red, green, and blue in the color image is that each object of the evaluation target image may have a different color from each other. Therefore, by extracting information on various colors, So that all of them can be detected. In addition, each object can have different colors as well as different saturation, brightness, hue, brightness, etc., and all such color information can be used later to determine each ROI. The contents defining the respective objects of the reference image can be stored in the template storage unit 65.

When the object is defined as described above, the ROI setting unit 15 can form an ROI for each object. In order to form an ROI, the ROI setting unit 15 detects an object using an edge detection algorithm. The corner detection algorithms include Laplacian algorithm, Differentiation algorithm, Prewitt algorithm, Sobel algorithm, etc. Any of these algorithms can be used. The edge detection algorithm is currently widely used and can be implemented without a detailed explanation by those skilled in the art. Therefore, a detailed description of the edge detection algorithm will be omitted in the present specification.

When the edge of each object is detected by the edge detection algorithm, the ROI setting unit 15 can form an ROI that is a rectangular frame surrounding the outer edge of the object based on the edge of each object.

When the ROI is set in the ROI setting unit 15, the ROI setting unit 15 transmits the reference image including the ROI to the coordinate setting unit 30, and the evaluation image is stored in the binary image processing unit 20, (25).

In order to convert an evaluation target image into a binary image, the binary image processing unit 20 first selects an ROI region in the evaluation target image, and converts each ROI region into a gray scale image. Then, the binary image processing unit 20 converts the ROI region of the gray scale image into a binary image. The binary image processing unit 20 can transmit the evaluation target image converted into the binary image to the coordinate setting unit 30. [

In the color image processing unit 25, color information can be extracted for each object of the color evaluation target image. The information extracted by the color image processing unit 25 may include red, green, blue, hue, saturation, brightness, color value, brightness, and the like. When the color information for each object is extracted, the color image processing unit 25 can transmit the evaluation target image to the coordinate setting unit 30. [

The coordinate setting unit 30 can set coordinate axes and center coordinates for each object of the reference image and the evaluation object image. The process of setting the coordinate axis and the center coordinate for the reference image object in the coordinate setting unit 30 and the process of setting the coordinate axis and the center coordinate for the object of the evaluation target image are the same, . However, the coordinate setting unit 30 processes information on coordinate axes and center coordinates set for each object differently depending on whether it is a reference image or an evaluation target image.

First, a process of setting coordinate axes and center coordinates for each object of the reference image and the evaluation target image in the coordinate setting unit 30 is as follows.

When the reference image and the evaluation target image are provided, the coordinate setting unit 30 can form a coordinate system using the longitudinal sides and the lateral sides of each ROI. That is, the coordinate setting unit 30 can form a coordinate system having a vertical axis and a horizontal axis that are parallel to the longitudinal side and the transverse side of each ROI. Methods for setting the coordinate system can be selected in various ways. Among them, a method using a representative object and a method using a reference image or a coordinate system for the entire image to be evaluated will be described.

In the method of using the representative object, one of the objects is selected as the representative object, the coordinate system is set for the representative object, and the coordinate system of the other objects is set by using the position relative to the coordinate system of the representative object. In the method of using the representative coordinate system for the reference image or the whole image to be evaluated, a coordinate system for the entire image is set first, a coordinate system is set as a representative coordinate system, and each object is set to a coordinate system using a position relative to the representative coordinate system will be.

Whichever way you set the coordinate system of each object, you can set the center of each coordinate system as the center coordinate. Thus, each object may have its own coordinate system with center coordinates.

When the coordinate system and the center coordinates of each object for the reference image are formed, the coordinate setting unit 30 transmits the information to the template storage unit 65 and stores the information together with the definition of each object. The coordinate system and the center coordinate of each object stored in the template storage unit 65 are compared with the coordinate system and the center coordinate of each object of the evaluation target image as the reference coordinate system and the reference center coordinate so as to evaluate whether or not the evaluation target image is transformed .

On the other hand, when the coordinate system and the center coordinates of each object are formed with respect to the evaluation target image, the coordinate setting unit 30 transmits information about the coordinate system and the center coordinates of each object to the rotation determination unit 35 or the object detection unit 40 . The coordinate target of the evaluation object image includes a color image and a binary image. The coordinate setting unit 30 sets the coordinate system and the center coordinate of each object with respect to the binary image from the binary image processing unit 20, together with the binary image, The coordinate system and the center coordinates of each object with respect to the color image from the color image processing unit 25 are transmitted to the object detection unit 40 together with the color image.

In the template storage unit 65, the image and the definition of each object for each object of the reference image set by the ROI setting unit 15, and the information about the coordinate system and the coordinates of each object set by the coordinate setting unit 30, And is stored as information.

The rotation determination unit 35 receives the coordinate system and the coordinates of the center of each object of the binary image as the evaluation target image from the coordinate setting unit 30. [ The rotation determination unit 35 can determine whether each object is rotated using the coordinate system and the center coordinates of each object. For this, the rotation determination unit 35 extracts the coordinate system and / or the center coordinate of each object with respect to the basic image from the template storage unit 65, and calculates the coordinate system and / or the center coordinate of each object of the basic image, Compare the coordinate system and center coordinates of each object. At this time, the rotation determining unit 35 may compare the vertical axis and the horizontal axis for each object of the basic image and the binary image, or may compare the center coordinates of each object of the basic image and the binary image. Whether the vertical axis and the horizontal axis are compared or the center coordinates are compared, the rotation determination unit 35 can obtain the rotation angle of the binary image.

For example, in the case of the object of the basic image shown in Fig. 3A, the object of the evaluation object image shown in Fig. 3B is not rotated but the object of the evaluation object is rotated to one side. The rotation determination unit 35 can easily determine whether the binary image is rotated by comparing the coordinate system or the center coordinate of each object of the basic image and the binary image.

3 (a) and FIG. 3 (b), it can be seen that the object to be evaluated has moved to one side with respect to the object of the basic image, not just with the object of the evaluation object rotated. This is because the image to be evaluated is not rotated about the object but rotated around another axis of rotation. In this case, the rotation determining unit 35 must find an object of the basic image matching the object of the evaluation target image. To this end, the rotation determination unit 35 retrieves and retrieves an object of the basic image having the same or very similar object definition as the object definition of the evaluation target image from the template storage unit 65. [ Then, the rotation determination unit 35 can obtain the rotation angle of the evaluation object image by comparing the coordinate system or the center coordinate of the object of the extracted basic image with the coordinate system or the center coordinate of the object of the evaluation object image. The rotation angle for each object of the evaluation object image obtained by the rotation determination unit 35 is provided to the distance calculation unit 60. [

The object detection unit 40 receives a color evaluation target image from the coordinate setting unit 30, detects an object of each ROI using an object detection algorithm (algorithm), and outputs the detected result to a dual- (45) and the distortion determining unit (55).

In the process of the object detection algorithm, image processing is facilitated by first removing an image noise using an image contour processing algorithm (Image Smoothing Algorithm). As an image contour processing algorithm, a low pass filter, a local average filter, a Gaussian filter, a Median filter, or the like can be used. Then, by using the edge detection algorithm mentioned above, it is possible to grasp the object by detecting the edge of each object.

Meanwhile, the object detecting unit 40 may use a color segmentation algorithm (Color Segmentation Algorithm). The color separation algorithm is an algorithm for detecting only a specific color in an evaluation target image. The object detection unit 40 sets a threshold value for each color information for each object of the evaluation target image using a color separation algorithm, . At this time, the reason for detecting multiple colors is to increase the detection probability of each object having various colors. The object detection unit 40 compares the color information of each object detected by the color image processing unit 25 with the color information of each object of the basic image stored in the template storage unit 65, Such as illuminance, saturation, lightness, etc., of the object.

When each object is identified, the distortion determining unit 55 can determine the distortion of the image to be evaluated by comparing the object detected by the object detecting unit 40 with an object of the basic image.

The distortion determining unit 55 can determine the amount of distortion and distortion of the evaluation target image by comparing the object image of the basic image with the object image of the evaluation target image by using an edge-based geometric pattern matching algorithm (Geometric Pattern Matching) algorithm.

FIG. 4A shows an object of the basic image stored in the template storage unit 65, and FIG. 4B shows an object of the image to be evaluated. The distortion determining unit 55 compares or overlaps the object of the evaluation target image of FIG. 4B with the object of the basic image.

At this time, when the object of the evaluation target image is distorted, the object of the basic image and the object of the evaluation target image do not match as shown in Fig. 4 (c). In Figure 4 (c), the gray part is the part where the object of the base image matches the object of the evaluation object, the green part is the remaining part of the basic image, and the red part is the remaining part of the object to be evaluated. 4 (c), it can be seen that the object of the evaluation target image is distorted. That is, it can be seen that the evaluation target image is distorted. The distortion determining unit 55 can calculate the amount of distortion using the area or width of the green portion and the red portion.

The double phase determination unit 45 can determine the double phase and calculate the degree of displacement by using the horizontal corner, vertical corner, circular corner, and concentric corner of each object detected by the object detection unit 40.

5, when each object is a polygon such as a triangle, a rectangle, etc., the double-top determination unit 45 draws a line on the horizontal edge and the vertical edge of each object, And a vertical line 501 is formed. Then, the double phase determination unit 45 scans the object in the vertical direction with respect to the horizontal line 502 from the center of the object, and when the edge is detected, the coordinates of the corresponding position are recorded. At this time, when the double phase is formed, two edges are detected, and the double phase determination unit 45 also records the coordinates of the second detected edge. As the edge is detected twice, the double phase determination unit 45 determines that the double phase is formed in the vertical direction, and the degree of the double phase in the vertical direction can be measured using the difference value of the coordinates.

Likewise, the double-phase determination unit 45 scans the center of the object in the horizontal direction with respect to the vertical line 501 as a reference, and records coordinates of the corresponding position when an edge is detected. When the corner is detected twice, the double phase determination unit 45 can determine that the double phase is formed in the horizontal direction. Further, the double phase determination unit 45 can measure the degree of occurrence of the double phase in the horizontal direction by using the difference between the coordinate values of the two vertical edges.

Since the dual phase determination unit 45 scans both the horizontal direction and the vertical direction, it can be determined whether the dual phase is formed in only one direction or both the horizontal direction and the vertical direction.

5, the double phase determination unit 45 detects a circular edge and / or a concentric circle edge 503 with respect to a circular or concentric object, Scan along the direction going outwards with respect to the center to sense the circular or concentric corners 503 and record the coordinates of the detected corners. In this case, when two or more circular edges are detected in the case of a circular shape, and a circular edge is detected in a number larger than the number of concentric circles in the case of a concentric circle, the double phase determination unit 45 may determine that a double phase has been formed. On the other hand, in the case of circular or concentric circles, coordinates of a plurality of positions are recorded at regular intervals along the circumferential direction. This is because, in the case of a circle or a concentric circle, the distance between the circular edge formed by the double phase and the original circular edge of the object is different along the circumferential direction.

The double phase determination unit 45 determines the distance by using the coordinates of the original circular edge of the object and the circular edge of the double phase, and determines whether the displacement of the double phase is formed in the vertical direction or the horizontal direction, It is possible to judge whether or not they are formed on both sides.

On the other hand, the coordinates recorded by the scan of the corners in the double-phase determination unit 45 are pixel coordinates. The coordinate conversion unit 50 receives the pixel coordinates from the double-phase determination unit 45 and converts them into a general coordinate system. And provides the coordinate system to the double-phase determination unit 45 so that the double-phase determination unit 45 can calculate the difference between the coordinate values.

The distance calculating unit 60 may determine the position of the camera 10 using the degree of distortion of the image to be evaluated and the degree of displacement of the double images and estimate the position of the virtual image according to the position of the camera 10. [

The distance calculation unit 60 defines the distance from the camera 10 to the virtual image as an arbitrary distance d and defines the distance from the windshield to the camera 10 as s. The normal distance calculation unit 60 defines the position of the camera 10 of the reference image as d + x. If the degree of distortion of the image is defined as w and the displacement of the double image is defined as t, the distance calculation unit 60 calculates the distance displacement x of the camera 10 and the amount of distortion w of the image to be proportional to each other , And the positional displacement amount x of the camera 10 can be calculated on the assumption that the positional displacement amount x of the camera 10 and the double-phase displacement amount t are proportional to each other. That is, the distance calculation unit 60 can determine the distance (d + x) between the camera 10 and the windshield, thereby estimating the distance from the camera 10 to the virtual image.

A process of forming a template from a reference image, which is a criterion for evaluating an evaluation target image in the HUD image evaluation system according to such a configuration, will be described with reference to FIG.

If a reference image having no distortion such as distortion or dual phase is input from the camera 10 or the HUD system in step S600, the ROI setting unit 15 defines an object using the shape and characters of each object in step S610, An edge detection algorithm is used to detect each object in the reference image. When the edge of each object is sensed, an ROI that surrounds each object is generated (S620). In step S630, the coordinate setting unit 30 forms a coordinate axis for the ROI of each object, thereby forming an XY coordinate system and a center coordinate for each object. At this time, if the coordinate system of the object is inclined with respect to the vertical axis and the horizontal axis of the entire screen of the basic image, the coordinate setting unit 30 can form each coordinate system using the rotation angle of the object.

The definition of each object set by this process, the ROI for each object, and information about the coordinate system are stored in the template storage unit 65 (S640).

When the template for HUD image evaluation is prepared, the HUD image evaluation system 1 evaluates the evaluation target image to determine the position of the camera 10 which can prevent the image distortion and the double image, as shown in FIG. 7 . This process will be described in detail as follows.

First, when the evaluation target image is input through the camera 10 (S700), the ROI setting unit 15 detects each object in the evaluation target image using the edge detection algorithm (S705) (Step S710).

The evaluation object image having the ROI is transmitted to the binary image processing unit 20, the color image processing unit 25, and the coordinate setting unit 30, respectively.

In step S715, the binary image processing unit 20 converts each ROI area of the evaluated object image into a gray scale image, and converts the gray scale image into a binary image again in step S720. Then, the binary image processing unit 20 provides the binary image to the coordinate setting unit 30. [ The coordinate setting unit 30 forms the coordinate system and the center coordinate by setting the vertical and horizontal coordinate axes for each ROI of the binary image (S725). At this time, the coordinate system set by the coordinate setting unit 30 is a pixel coordinate system.

The coordinate setting unit 30 provides the rotation determination unit 35 with information such as a coordinate system and a center coordinate for each object of the binary image. The rotation determination unit 35 extracts the coordinate system and / or the center coordinates of each object with respect to the base image from the template storage unit 65 and acquires the coordinate system and / or the center coordinate of each object of the base image and each object of the binary image (Step S730). The rotation angle is calculated by comparing the coordinate system and / The rotation angle of each object measured by the rotation determination unit 35 is a rotation angle of the evaluation target image, and this information is displayed externally so that the operator of the present HUD image evaluation system 1 can recognize the rotation angle.

The color image processing unit 25 acquires color information for each ROI of the evaluation target image (S735). The color information processing unit 25 may include color information for each ROI from the distortion determining unit 55 and the color information for each ROI, (45).

The coordinate setting unit 30 sets a coordinate axis and a center coordinate for each object of the color evaluation target image (S740), and transmits information about the coordinate axes and the center coordinates of each object to the object detection unit 40. [

The object detection unit 40 detects an object for each ROI of the evaluation object image using the object detection algorithm, and extracts information about the object from the template storage unit 65 (S745). Then, the distortion determining unit 55 not only can determine whether the evaluation target image is distorted by comparing or overlaying the object of the basic image with the object of the evaluation target image by using the geometric pattern matching algorithm, The amount of distortion can be grasped (S750). The distortion amount is provided to the normal distance calculation unit 60.

The double-phase determination unit 45 can determine the double-phase state and the degree of displacement by scanning the object in each ROI of the evaluation object image in the horizontal direction, the vertical direction, and / or the concentric direction (S755). For example, when the object is a polygon, the double-phase determination unit 45 scans the ROI in the horizontal direction from the center of the ROI, scans the ROI in the vertical direction with respect to the horizontal axis, Record the coordinates of the location. If the edge is detected more than once, the double phase determination unit 45 determines that a double phase has occurred. The double phase determination unit 45 transfers the coordinates of the respective corners to the coordinate transformation unit 50, and the coordinate transformation unit 50 converts the pixel coordinates into general coordinates and provides them to the double phase determination unit 45 again. In the double phase determination unit 45, the degree of displacement of the double phase can be measured by calculating the coordinate difference value of the general coordinate system. The dual-phase displacement value measured by the double-phase determination unit 45 is provided to the distance calculation unit 60.

The actual distance between the camera 10 and the windshield is calculated using the distortion amount and the dual-phase displacement value, which is determined by the average distance calculating unit 60, to be proportional to the amount of change in the distance between the camera 10 and the windshield window (S760) . Thus, the installation position of the camera 10 can be determined.

As described above, since the distortion, the double phase, the rotation, and the displacement amount of the image transmitted from the HUD system can be measured, it is possible to set the position of the camera 10 which can prevent the distortion and the double image. The result of the rotation is displayed outside, thereby providing an opportunity for the operator to search for countermeasures to prevent rotation as well as distortion and double phase. Therefore, a clear virtual image can be generated for the image transmitted from the HUD system, so that not only the quality of the HUD system can be improved, but also the satisfaction of the user can be achieved.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, Of the right.

1: HUD image evaluation system 10: camera
15: ROI setting unit 20: Binary image processing unit
25: Color image processing unit 30: Coordinate setting unit
35: rotation determination unit 40: object detection unit
45: double phase judging unit 50: distortion judging unit
60: Normal distance calculating unit 65: Template storing unit

Claims (24)

A camera for photographing a virtual image formed from a HUD system mounted on a vehicle;
A coordinate setting unit for setting a coordinate system for each object that is an individual element included in the virtual image;
A template storage unit for storing information on each object included in a reference image that is a reference for determining whether the virtual image is deformed;
A dual phase determiner for determining a double phase according to whether the edges of the objects are overlapped or not by scanning each of the objects and for measuring the degree of displacement using the coordinate system set by the coordinate setting unit when the double phase is detected; And
And a distortion determination unit for determining whether the virtual image is distorted or distorted by comparing an object of the reference image drawn from the template storage unit with an object of the virtual image. The method according to claim 1,
And an ROI setting unit that receives the reference image and the virtual image, detects each of the objects using a corner detection algorithm, and sets a ROI (Region of Interest), which is a rectangular frame surrounding the outline of each object HUD image evaluation system. 3. The method of claim 2,
Wherein the ROI setting unit defines each object using information including at least one of a type, a shape and a color of each object. 3. The method of claim 2,
Further comprising a binary image processing unit for converting the gray image into a binary image by receiving the virtual image in which the ROI is set and converting the gray image into a gray image. 3. The method of claim 2,
Further comprising a color image processing unit for receiving the virtual image in which the ROI is set and extracting color information including at least one of a hue, illuminance, brightness, saturation, and brightness along with the color. 5. The method of claim 4,
Wherein the coordinate setting unit sets a coordinate system for each object included in the binary image provided from the binary image processing unit. The method according to claim 6,
Further comprising a rotation determination unit for determining a rotation state and a rotation angle of each object by comparing a coordinate system of each object of each of the binary images with a coordinate system of a corresponding object stored in the template storage unit. 3. The method of claim 2,
Further comprising an object detection unit for detecting an object in each ROI by applying an object detection algorithm to the virtual image in which the ROI is set. 9. The method of claim 8,
Wherein the distortion judging unit compares an object of the basic image corresponding to the object detected by the object detecting unit with an object of the virtual image using an edge based geometric pattern matching algorithm to calculate a distortion and a distortion amount And the HUD image evaluation system. The method according to claim 1,
Wherein the double phase determination unit scans each object along a direction orthogonal to a vertical edge, a horizontal edge, a circular edge, and a concentric corner of each object to record coordinates for each detected edge, and when each object is a polygon or a circle If the number of detected edges is two or more, it is determined to be a dual phase, and when each of the objects is a concentric circle, if the detected edge is greater than the number of concentric circles, the dual image is determined. 11. The method of claim 10,
And a coordinate converter for converting the pixel coordinates recorded for the edges detected by the double-phase determination unit into general coordinates. 12. The method of claim 11,
Wherein the double-phase determination unit calculates a dual-phase displacement value by calculating a difference between coordinate values of the corners in the general coordinate system. The method according to claim 1,
And a controller for receiving the double-phase displacement value calculated by the double-phase determination unit and the distortion amount measured by the distortion determination unit to calculate a distance between the camera and the windshield of the vehicle, Wherein the HUD image evaluation system further comprises a calculation unit. Receiving a virtual image formed from a HUD system mounted on a vehicle;
A coordinate setting step of setting a coordinate system for each object that is an individual element included in the virtual image;
A dual image determining step of scanning each of the objects and determining a double image of the virtual image according to whether the edges of the objects overlap each other;
A displacement measuring step of measuring a degree of displacement of the virtual image using the coordinate system when the virtual image is a double phase; And
And a distortion judging step of judging whether the virtual image is distorted or distorted by comparing each object included in the reference image, which is a reference for judging whether or not the virtual image is deformed, with the object of the virtual image, To evaluate the HUD image. 15. The method of claim 14,
When the reference image or the virtual image is input,
Sensing each object using an edge detection algorithm;
Further comprising setting a region of interest (ROI) that is a rectangular frame surrounding the perimeter of each object. 15. The method of claim 14,
When the reference image or the virtual image is input,
Further comprising defining each of the objects using information including at least one of a type, a shape, and a color of each of the objects. 16. The method of claim 15,
Further comprising the step of converting the virtual image having the ROI set to a gray scale image if the ROI is set in the virtual image, and converting the gray scale image into a binary image. 16. The method of claim 15,
Further comprising extracting color information including at least one of hue, illuminance, lightness, saturation, and brightness with the color by receiving the virtual image in which the ROI is set. 18. The method of claim 17,
And setting a coordinate system for each object included in the binary image. 20. The method of claim 19,
Further comprising the step of comparing the coordinate system of each object of the binary image with the coordinate system of each object of the reference image to determine the rotation and the rotation angle of each object. 16. The method of claim 15,
Further comprising detecting an object in each ROI by applying an object detection algorithm to the virtual image in which the ROI is set. 15. The method of claim 14,
The double-
Scanning each of the objects along a direction perpendicular to a vertical edge, a horizontal edge, a circular edge, and a concentric corner of each of the objects to record coordinates for each detected edge;
If the detected object is a polygon or a circle, it is determined that the object is a double-face if there are two or more sensed edges, and if the object is a concentric circle, if the sensed edge is greater than the number of concentric circles, HUD image evaluation method. 16. The method of claim 15,
The double-
Converting the coordinates of the edge from pixel coordinates to general coordinates,
And calculating a difference value of coordinates between the corners in the general coordinate system to calculate a displacement value of the double-phase image. 15. The method of claim 14,
Further comprising the step of calculating a distance between the camera and the windshield of the vehicle capable of eliminating the double phase and the distortion by using the double-phase displacement value and the distortion amount.

Images