KR102065337B1 - Apparatus and method for measuring movement information of an object using a cross-ratio - Google Patents

Abstract

The present invention provides an apparatus and a method for measuring movement information of a target by using a cross-ratio which can quickly and accurately calculate movement information of a target by using a cross-ratio. The apparatus comprises: a camera to acquire an image frame of a moving target; a moving information extraction unit to receive a plurality of image frames from the camera, and extract a position change of a reference point of the target from each image frame; a moving distance measurement unit to use a cross-ratio of the target and the position change of the reference point between image frames photographed by the camera in a first environment to measure a first distance on pixels by which the target has moved; a first moving distance calculation unit to correct the first distance to calculate a second distance which is the real distance by which the target has moved in a three-dimensional space in the first environment; a reference data storage unit storing a corresponding relationship of the first distance and the second distance as reference data; and a second moving distance calculation unit to compare a position change of the reference point between image frames photographed by the camera in a second environment with a lower identifying capability than the first environment and the reference data to calculate a third distance which is the real moving distance of the target in the second environment.

Description

Apparatus and method for measuring movement information of an object using a cross-ratio}

The present invention relates to an apparatus and a method, and more particularly, to an apparatus and method for measuring movement information of a moving object using a non-harmonic ratio.

When a camera photographs an object moving in three dimensions, a two-dimensional image is generated. The captured two-dimensional image cannot reflect three-dimensional information, and a difference in information occurs according to the position or direction of the camera.

In addition, the captured image is affected by mechanical parts inside the camera, such as the lens used, the distance between the lens and the image sensor, and the angle between the lens and the image sensor. Therefore, when calculating the position where the 3D points are projected on the image or conversely restoring the 3D spatial coordinates from the coordinates in the image, it is necessary to know the parameters of the internal structure of the camera so that accurate calculation is possible. That is, camera calibration is required to obtain 3D actual information from the 2D image.

The grid is often used to calibrate the camera. However, this calibration method has a difficulty in that the position of the camera must be fixed and the gratings must be set according to various imaging angles and positions of the camera.

In the case of CCTV or black box images used for emotions, such as forensic science, it is difficult to obtain measurement data by extracting correction parameters from a camera. In general, the method of calculating the average speed in the coordinate interval after selecting fixed coordinates in the image is used, but it is difficult to analyze the instantaneous speed change or viewpoint, and it is difficult to extract the fixed coordinates from the image, for example, dark environment, There is a limit that it is not applicable to a dirt road or a wide flat land.

The present invention provides an apparatus and method capable of quickly and accurately calculating movement information of an object using a cross-ratio.

According to an aspect of the present invention, a camera for acquiring an image frame of a moving object, a movement information extracting unit for receiving a plurality of image frames from the camera, and extracting a position change of a reference point of the object in each image frame; And a moving distance measuring a first distance on a pixel to which the object is moved by using a change in position of the reference point and a cross-ratio of the object between image frames captured by the camera in a first environment. A first moving distance calculator configured to correct the first distance to calculate a second distance that is an actual distance of the object in a three-dimensional space in the first environment, the first distance and the first distance; A reference data storage unit storing the correspondence relationship of two distances as reference data, and the image captured by the camera in a second environment having a lower discrimination power than the first environment. An object using a non-harmonic ratio including a second moving distance calculator configured to compare a position change of the reference point and the reference data between unknown frames and calculate a third distance, which is an actual moving distance of the object in the second environment; It provides a device for measuring the movement information of.

The moving distance measuring unit may distinguish whether or not the reference sections, which are distances between the first reference point and the second reference point of the object, overlap each other in the image frames to be compared.

The movement distance measuring unit may calculate the first distance according to the following equation when the reference intervals overlap in the image frames to be compared.

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

The moving distance measuring unit may calculate the first distance according to the following equation when the reference section is continuous in the image frames to be compared.

D1 = L

D1: first distance, L: distance between the first reference point and the second reference point (reference section)

The moving distance measuring unit may calculate the first distance according to the following equation when the reference section is spaced apart from the image frames to be compared.

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

The first moving distance calculator may calculate the second distance from a previously stored correction function for the first distance.

The apparatus may further include a speed calculator configured to calculate an actual moving speed of the object by using at least one of the second distance and the third distance.

In addition, the actual moving speed of the object may be calculated according to the following equation.

V _{i, j} is the average speed of the object from the i th image frame to the j th image frame, D2 _{i, j} is the second or third distance of the object from the i th image frame to the j th image frame, and FPS (Frames per second) is the number of image frames per second, and | ij | is the number of frames between the i th image frame and the j th image frame.

In addition, the movement information extracting unit sets the reference point of the object in the first environment to the rotation center position of the front wheel of the vehicle and the rotation center position of the rear wheel, and the reference point of the object in the second environment Can be set to position.

The apparatus may further include an object information storage configured to store shape information for each object, and the movement information extractor may specify an object included in the image frame using the shape information of the object information storage.

According to another aspect of the present invention, there is provided a method of obtaining a plurality of image frames of an object moving from a camera, comparing the plurality of image frames, extracting a change in position of a reference point of the object, and Calculating a first distance that the object is moved in the image frame to be compared by using a positional change of the reference point and a cross-ratio of the object in the image frame photographed by the camera; Correcting a first distance to a second distance, which is an actual distance traveled by the object in a three-dimensional space of the first environment, storing a corresponding relationship between the first distance and the second distance as reference data, and The positional change of the reference point and the reference data in the image frame photographed by the camera in a second environment having a lower discrimination power than the first environment Comparing the data, and calculating a third distance, which is an actual distance of the object moving in the three-dimensional space of the second environment, in the image frames to be compared. to provide.

The calculating of the first distance may determine whether the reference sections, which are distances between the first reference point and the second reference point of the object, overlap each other in the image frames to be compared.

In the calculating of the first distance, when the reference intervals overlap in the image frames to be compared, the first distance may be calculated according to the following equation.

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

In the calculating of the first distance, when the reference section is continuous in the image frames to be compared, the first distance may be calculated according to the following equation.

D1 = L

D1: first distance, L: distance between the first reference point and the second reference point (reference section)

In the calculating of the first distance, when the reference section is spaced apart from the image frames to be compared, the first distance may be calculated according to the following equation.

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

The method may further include calculating an actual moving speed of the object by using at least one of the second distance and the third distance.

In the calculating of the actual moving speed of the object, the actual moving speed may be calculated according to the following equation.

V _{i, j} is the average velocity of the object from the i th image frame to the j th image frame, D2 _{i, j} is the second or third distance of the object from the i th image frame to the j th image frame, and FPS (Frames per second) is the number of image frames per second, and | ij | is the number of frames between the i th image frame and the j th image frame.

The extracting of the change of the position of the reference point of the object may include setting the reference point of the object in the first environment as the rotation center position of the front wheel of the vehicle and the rotation center position of the rear wheel, and in the second environment, the reference point of the object. May be set to one preset position of the vehicle.

An apparatus and method for measuring movement information of an object using a non-harmonic ratio according to the present invention may accurately and simply measure movement information of an object in an image frame photographed at night based on information obtained at daytime. Can be. In the first environment where the object is highly distinguishable, the moving information in three dimensions is accurately calculated from the image captured in two dimensions using non-harmonization and stored as reference data, and the image is captured in the second environment where the object is low in recognition power By comparing the positional change of the reference point on the pixel with the reference data, the actual moving distance and speed can be calculated accurately.

The apparatus and method for measuring movement information of an object using a non-harmonic ratio according to an embodiment of the present invention do not require additional work for calibrating a camera, and convert the reference point to coordinates without measuring a distance within an image to quickly move information. Can be calculated.

The apparatus and method for measuring movement information of an object using a non-harmonic ratio according to an embodiment of the present invention can accurately analyze a moving distance and a speed of a vehicle without being influenced by the weather or the surrounding environment, and the object at each moment. Speed change can be used to effectively predict and reproduce traffic conditions.

The apparatus and method for measuring movement information of an object using a non-harmonic ratio according to an embodiment of the present invention may store reference data about the shape of each object, and quickly calculate the movement information of the object quickly and accurately by using the same. . Of course, the scope of the present invention is not limited by these effects.

1 is a structural diagram illustrating an apparatus for measuring movement information of an object using a harmonic ratio according to an embodiment of the present invention.
FIG. 2 is a structural diagram illustrating a movement information extractor of the object of FIG. 1.
3A to 3C are diagrams illustrating a moving distance according to a position of an object.
4 is a graph showing a relationship between a first distance and a second distance.
5 is a diagram illustrating a moving distance of an object in a second environment.
6 is a flowchart illustrating a method of measuring movement information of an object using a harmonic ratio according to another embodiment of the present invention.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. Effects and features of the present invention, and methods of achieving them will be apparent with reference to the embodiments described below in detail together with the drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various forms.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, and the same or corresponding components will be denoted by the same reference numerals, and redundant description thereof will be omitted. .

In the following examples, the singular forms "a", "an" and "the" include plural forms unless the context clearly indicates otherwise.

In the following examples, the terms including or having have meant that there is a feature or component described in the specification and does not preclude the possibility of adding one or more other features or components.

In the case where an embodiment may be implemented differently, a specific process order may be performed out of the described order. For example, two processes described in succession may be performed substantially simultaneously or in the reverse order of the described order.

In the drawings, components may be exaggerated or reduced in size for convenience of description. For example, the size and thickness of each component shown in the drawings are arbitrarily shown for convenience of description, and therefore the following embodiments are not necessarily limited to those shown.

1 is a structural diagram illustrating an apparatus 1 for measuring movement information of an object using a harmonic ratio according to an embodiment of the present invention.

Referring to FIG. 1, an apparatus 1 for measuring movement information of an object includes a camera 10, a reference data storage 20, a controller 30, and a display 40, and the controller 30 moves. The information extracting unit 31, the moving distance measuring unit 32, the first moving distance calculating unit 33, the second moving distance calculating unit 34, the reference data storing unit 35 and the speed calculating unit 36 It can be provided.

The apparatus 1 for measuring movement information of an object may measure position information, velocity information, acceleration information, and the like of a moving object. In particular, the apparatus 1 for measuring the movement information of the object extracts and processes the position information of the object from the image frame acquired by the camera 10, and thus, the position information, velocity information, or acceleration information of the object in a real three-dimensional space. Etc. can be measured. That is, the apparatus 1 for measuring the movement information of the object may extract position information from a two-dimensional image image to accurately obtain information about a position or a speed at which the object actually moves in three dimensions.

The object may be various objects that are movable. For example, the object may be various objects that may be captured by the camera 10 while moving, such as a person, an animal, a vehicle, an aircraft, a ship, and the like. However, hereinafter, the description will be given based on the case where the object is a vehicle for convenience of description.

Hereinafter, the first environment is defined as an environment in which the resolution of the image frame generated by the camera 10 is high, for example, an environment in which the object can be clearly identified in the captured image, such as a bright environment and a clear weather. . However, hereinafter, the first environment will be described based on the case where the first environment is a bright environment for the convenience of description.

Hereinafter, the second environment is an environment in which the image frame generated in the camera 10 has a lower resolution than the image frame picked up in the first environment. For example, the second environment is defined as an environment in which the object is not clearly identified in the captured image or only a part of the object can be identified, such as an environment in which the main surface is dark, a weather is cloudy or rainy. However, hereinafter, the description will be given based on the case where the second environment is a dark environment at night for convenience of description.

The camera 10 may acquire an image frame by capturing a moving object. The camera 10 may acquire a plurality of image frames by continuously photographing an object.

The camera 10 may generate an image frame in a first environment, and may generate an image frame in a second environment in which the object of identification of the object is lower than that of the first environment.

The reference data storage 20 may store reference data of each object, for example, shape information. For example, shape information of various types of vehicles running on a road can be stored. In this case, the shape information may be set in various ways such as information about the appearance of the vehicle, information about the distance between the front wheel and the rear wheel of the vehicle, information about the position of the headlight.

The controller 30 may receive a plurality of image frames from the camera 10, process the same, and calculate movement information of a moving object. The controller 30 includes a movement information extracting unit 31, a moving distance measuring unit 32, a first moving distance calculating unit 33, a second moving distance calculating unit 34, a reference data storage unit 35, and a speed. The calculator 36 may be provided.

The movement information extractor 31 may receive a plurality of image frames from the camera 10 and extract a change in the position of the reference point of the object from each image frame. In addition, the movement information extractor 31 may track the coordinates of the reference point of the vehicle in each image frame.

The reference point of the object is used to define a non-harmonization ratio of the object in the first environment, and may have a pair of first reference point and second reference point. The first reference point and the second reference point are not limited to a specific position and may be variously set according to the object. In the first environment (weekly), the reference points of the objects may be specified in pairs. In the daytime, since the position information of the pixel can be clearly obtained from the image frame acquired through the camera, a pair of reference points are used to accurately define the non-harmonization ratio of the object. For example, the position of the front bumper of the vehicle and the position of the rear bumper of the vehicle may be variously set. However, hereinafter, the first reference point is the position of the rear wheel of the vehicle, and the second reference point will be described with reference to the case of the position of the front wheel of the vehicle for convenience of description.

Also, the reference point of the object may be used to specify the positional movement of the vehicle in the second environment. In the second environment, the reference point of the object may be specified as one. At night, since there is a limit in accurately obtaining position information of a plurality of pixels from an image frame acquired through a camera, one reference point is used. For example, the reference point may be specified as a head lamp of the vehicle, a bumper of the vehicle, or the like.

FIG. 2 is a structural diagram illustrating the movement information extracting unit 31 of the object of FIG. 1.

Referring to FIG. 2, the movement information extractor 31 may include a reference information acquirer 311 and a reference section calculator 312 of the object.

The reference information acquisition unit 311 is connected to the object information storage unit 20 and may receive information for specifying the object. The reference information acquisition unit 311 may specify an object captured in the image frame by using shape information of the object information storage unit 20. For example, the shape of the vehicle may be extracted from the image frame and compared with the reference data to specify that the moving object is a vehicle of a specific model.

The reference section calculator 312 may specify a first reference point and a second reference point in the image frame acquired in the first environment. The first reference point is the rotation center of the rear wheel, and the second reference point is the rotation center of the front wheel. In addition, the distance between the first reference point and the second reference point may be defined as a reference section.

The reference section calculator 312 may calculate a reference section by specifying a first reference point and a second reference point for each of the plurality of image frames received from the camera 10.

The movement distance measuring unit 32 may calculate the first distance that the object moves between image frames by using the position change of the reference point and the cross-ratio of the object in the first environment.

Perspective Transformation is a linear transformation that maintains the linearity of a straight line from one plane to another. In projective geometry, two different straight lines meet at one point. In projective geometry, the nonharmonic ratios of four points on the same line are constant as projective invariants. Therefore, both dimensions from a straight line connecting four points (A, B, C, D) in a real three-dimensional space, and a straight line connecting four points (A ', B', C ', D') in an image frame. It may have an incompatibility ratio of the liver. Accordingly, the present invention can obtain a non-harmonic ratio in an image frame, and calculate the distance and speed at which the object has moved in actual three dimensions based on this.

Hereinafter, the non-harmonization ratio of the object may be calculated using the coordinate values of the pixels through digital image processing. The non-harmonization ratio is calculated through image processing in advance before calculating the first distance.

3A to 3B are diagrams illustrating a moving distance according to the position of the object.

3A to 3B, three cases of an image frame continuously captured according to time occur according to whether or not a distance between a first reference point and a second reference point of an object (vehicle) overlaps.

The movement distance measuring unit 32 distinguishes whether reference intervals, which are distances between the first reference point and the second reference point of the object, overlap each other in the image frames to be compared, and calculates a first distance according to each case. can do.

Referring to FIG. 3A, reference intervals overlap in image frames to be compared. In the first frame T1, the first reference point is (A ') and the second reference point is (B'). Thereafter, in the frame T2, the first reference point changes to (C ′), and the second reference point changes to (D ′). In this case, C 'is present in a section between A' and B '.

In this case, the non-harmonic ratio R may be defined as follows, and the first distance may be calculated according to Equation 1 below.

[Equation 1]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

Referring to FIG. 3B, reference intervals are continued in the image frames to be compared. In the first frame T1, the first reference point is (A ') and the second reference point is (B'). Thereafter, in the frame T2, the first reference point changes to (C ′), and the second reference point changes to (D ′). At this time, the position of B 'and the position of C' is the same.

In this case, although the non-harmonic ratio R is not specified, the first distance may be calculated according to Equation 2 below.

[Equation 2]

D1 = L

D1: first distance, L: distance between the first reference point and the second reference point (reference section)

Referring to FIG. 3C, reference intervals are spaced apart from image frames to be compared. In the first frame T1, the first reference point is (A ') and the second reference point is (B'). Thereafter, in the frame T2, the first reference point changes to (C ′), and the second reference point changes to (D ′). At this time, C 'is spaced apart from B'.

In this case, the non-harmonic ratio R may be defined as follows, and the first distance may be calculated according to Equation 3 below.

[Equation 3]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object

The movement distance measuring unit 32 may calculate D1 in which the object is moved on the pixel in the first environment in the image frame to be compared by using different equations and the non-harmonization ratio of the object according to each case.

The first moving distance calculator 33 corrects the first distance D1, which is the distance the object moves on the image frame, so that the second distance D2 is the actual distance that the object moves in the 3D space in the first environment. Can be calculated.

The first moving distance calculator 33 may calculate the second distance from the first distance by using various interpolation methods. For example, the first moving distance calculator 33 may calculate the second distance D2 from a correction function for the previously stored first distance D1.

4 is a graph showing the relationship between the first distance D1 and the second distance D2.

Referring to FIG. 4, the reference data storage unit 35 may store the correspondence relationship between the first distance D1 and the second distance D2 as reference data. The reference data storage unit 35 functions as a function the correspondence relationship between the first distance D1 calculated by the moving distance measuring unit 32 and the second distance D2 calculated by the first moving distance calculating unit 33, respectively. Can be stored.

The graph of FIG. 4 illustrates a correction function between D1, which is a position change in a pixel of an object in an image frame, and D2, which is a distance traveled by an object in an actual three-dimensional space. As shown in FIG. 4, reference data having a relationship of D2, which is an actual three-dimensional moving distance corresponding to D1, which is a distance between pixels, may be stored in the reference data storage 35 as a function. The reference data may be derived by measuring and calculating D1 and D2 a plurality of times in the first environment.

5 is a diagram illustrating a moving distance of an object in a second environment.

Referring to FIG. 5, the second moving distance calculator 34 changes a position between an image frame of a reference point in a second environment and a reference distance to a third distance by using the reference data. It can calculate by (D3).

The second moving distance calculator 34 is used to calculate the third distance D3 which is the actual moving distance of the object in the second environment. In the second environment, it is difficult for the camera 10 to capture a clear image frame. In particular, it is difficult to specify the reference point of the object in the image in a dark environment, the accuracy is relatively low.

The second moving distance calculator 34 may calculate the third distance D3 simply and accurately using the reference data acquired in the first environment. In the daytime, the first distance D1 and the second distance D2 calculated by the movement distance measuring unit 32 and the first movement distance calculation unit 33 have high accuracy, and the reference data also has high accuracy. Therefore, if only the moving distance of the object on the image frame can be accurately specified through the data in the second environment, the actual moving distance D3 may be calculated using the reference data.

In detail, the movement information extractor 31 acquires image frames T1 and T2 of the object moving in the second environment from the camera 10. The movement information extractor 31 sets the reference point E of the object to one preset position of the object in the second environment. For example, if the object is a vehicle, the object may be set as a head lamp or a bumper of the vehicle. Hereinafter, for convenience of description, a case in which the head lamp is set as a reference point will be described.

That is, the movement information extracting unit 31 may extract the movement trajectories TL of each headlamp from the plurality of image frames transmitted from the camera 10 in the second environment. The movement information extractor 31 may extract a movement distance between the first image frame T1 and the second image frame T2 through the movement trajectory TL on the pixel of the headlamp.

The second movement distance calculator 34 may calculate the movement distance D3 in the actual second environment by comparing the movement distance of the reference point E with the reference data in the movement information extractor 31. Since the movement trajectory TL of the reference point E corresponds to D1, the third distance D3 of the movement of the actual object may be calculated using a function of reference data. An actual moving speed of the object may be calculated using at least one of the distance D2 and the third distance D3. The speed calculator 36 may calculate an average speed of the object actually moving in three dimensions between the image frames to be compared.

The actual moving speed of the object may be calculated according to Equation 4 below.

[Equation 4]

V _{i, j} is the average speed of the object from the i th image frame to the j th image frame, D2 _{i, j} is the second or third distance of the object from the i th image frame to the j th image frame, and FPS (Frames per second) is the number of image frames per second, and | ij | is the number of frames between the i th image frame and the j th image frame.

The display unit 40 is connected to the controller 30 and may display movement information of the object calculated by the controller 30. For example, the display unit 40 may display an actual moving distance of the object, an actual moving speed of the object, or an acceleration of the object.

The display unit 40 may be a variety of display devices or terminals that can communicate with the controller 30 by wire / wireless. For example, the display unit 40 may be a monitor connected to the controller 30 by wire. In another embodiment, the display unit 40 may be a portable terminal connected to the controller 30 through a wireless network.

6 is a flowchart illustrating a method of measuring movement information of an object using a harmonic ratio according to another embodiment of the present invention.

Referring to FIG. 6, the method of measuring movement information of an object may include obtaining a plurality of image frames of an object moving from a camera (S10), comparing the plurality of image frames, and comparing a preset reference point of the object. Extracting the position change (S20), and using the position change of the reference point and the cross-ratio of the object in a first environment, a first distance moved by the object between the image frames; Calculating (S30), correcting the first distance to a second distance that is an actual distance that the object moves in a three-dimensional space in a first environment (S40), and the first distance and the second distance Storing the corresponding relation of the reference data as reference data (S50), comparing the positional change of the reference point with the reference data in a second environment, and comparing the object in the image frame to be compared Computing a third distance which is the actual distance moved in the three-dimensional space of the second environment (S60), and using the at least one of the second distance and the third distance, calculating the actual moving speed of the object Step S70 may be included.

In operation S10 of obtaining a plurality of image frames of a moving object from the camera, the plurality of image frames may be obtained by continuously photographing the moving object of the camera 10.

Comparing the plurality of image frames, and extracting a change in the position of the reference point of the predetermined object (S20) by comparing a plurality of image frames, each of the reference point of the object may be extracted as coordinates. The movement information extractor 31 may extract a reference point from each of the two-dimensional comparing image frames.

At this time, in the first environment, the reference point of the object may be set to the rotation center position of the front wheel and the rear wheel rotation center position of the vehicle, and in the second environment, the reference point of the object may be set to one preset position of the vehicle. The reference point at night may be set, for example, as a head lamp or a bumper of a vehicle.

Comparing the plurality of image frames generated in the first environment, and extracting the change in the position of the reference point of the predetermined object (S30) in the image frame compared by the moving distance measuring unit 32 overlap the reference intervals The first distance D1 may be calculated according to each case.

 In operation S40, the first distance is corrected by a second distance, which is an actual distance of the object in a three-dimensional space of a first environment, by using a correction function in the first movement distance calculator 33. (D1) can be calculated as the second distance D2. That is, the distance traveled by the actual object in the first environment may be calculated by calculating the second distance D2, which is the actual moving distance on the 3D, from the first distance D1, which is the moving distance on the 2D.

In operation S50, the corresponding relationship between the first distance and the second distance may be stored as reference data, and the reference data storage 35 may function as a function of the correspondence between the first distance D1 and the second distance D2, respectively. Can be stored as Reference data having a relationship of D2, which is an actual three-dimensional moving distance corresponding to D1, which is a distance between pixels in an image frame, may be stored in the reference data storage 35 as a function. The reference data may be derived by measuring and calculating D1 and D2 a plurality of times in the first environment.

Comparing the position change of the reference point and the reference data in a second environment, calculating a third distance which is the actual distance that the object moved in the three-dimensional space of the second environment in the image frame to be compared (S60) The moving distance of the object in the second environment may be calculated.

The second moving distance calculator 34 may calculate the third distance D3 simply and accurately using the reference data acquired in the first environment. In the first environment, the first distance D1 and the second distance D2 calculated by the movement distance measuring unit 32 and the first movement distance calculation unit 33 have high accuracy, and the reference data has high accuracy. Therefore, if only the moving distance of the object on the image frame can be accurately specified through the data in the second environment, the actual moving distance D3 may be calculated using the reference data.

The second movement distance calculator 34 may calculate the movement distance D3 in the actual second environment by comparing the movement distance of the reference point (for example, the headlamp) and the reference data in the movement information extractor 31. . Since the moving distance of the reference point corresponds to D1, the moving distance of the actual object may be calculated through reference data.

The calculating of the actual moving speed of the object by using at least one of the second distance and the third distance (S70) is performed by the speed calculator 36 using the second distance and the FPS information. The velocity moved in the dimensional space can be calculated.

An apparatus and method for measuring movement information of an object using a non-harmonic ratio according to the present invention, based on the information obtained in the first environment, accurately and simply the movement information of the object in the image frame captured in the second environment It can be measured. In the first environment, non-harmonization is used to accurately calculate movement information in three dimensions from an image captured in two dimensions and store it as reference data.In the second environment, the positional change of a reference point on a pixel and reference data in the captured image. By comparing this, the actual moving distance and speed can be calculated accurately.

The apparatus and method for measuring movement information of an object by using a non-harmonic ratio according to the present invention have no additional work for calibrating a camera, and can quickly calculate movement information by converting only a reference point to coordinates without measuring a distance in an image. Can be.

The apparatus and method for measuring movement information of an object using a non-harmonic ratio according to the present invention can accurately analyze a moving distance and a speed of a vehicle without being affected by the weather or the surrounding environment, and change the speed of the object at each moment. It can be effectively used for predicting and reproducing traffic conditions.

The apparatus and method for measuring movement information of an object using a non-harmonic ratio according to the present invention may store reference data about the shape of each object, and quickly and accurately calculate movement information of the object by using the same.

The spirit of the present invention should not be limited to the embodiments described above, and the scope of the appended claims, as well as all equivalents or equivalents thereto, are defined within the scope of the spirit of the present invention. Will belong.

1: Device for measuring movement information of an object using a non-harmonic ratio
10: camera
20: reference data storage
30: controller
40: display unit

Claims (18)

A camera for obtaining an image frame of a moving object;
A movement information extracting unit receiving a plurality of image frames from the camera and extracting a change in position of a reference point of the object in each image frame;
Moving distance measurement for measuring a first distance on a pixel to which the object is moved using a change in position of the reference point and a cross-ratio of the object between image frames captured by the camera in a first environment part;
A first moving distance calculator configured to correct the first distance and calculate a second distance that is an actual distance of the object in a three-dimensional space in the first environment;
A reference data storage unit storing a correspondence relationship between the first distance and the second distance as reference data; And
Comparing the positional change of the reference point and the reference data between image frames captured by the camera in a second environment having a lower discrimination power than the first environment, a third distance, which is an actual moving distance of the object in the second environment, is compared. And a second movement distance calculator configured to calculate movement information of the object using a non-harmonic ratio. According to claim 1,
The moving distance measuring unit
The apparatus for measuring movement information of an object by using a non-harmonic ratio in the image frame to be compared, to distinguish whether the reference interval, which is a distance between the first reference point and the second reference point of the object overlap each other. The method of claim 2,
The moving distance measuring unit
And calculating the first distance according to Equation 1 below when the reference periods overlap in the image frames to be compared.
[Equation 1]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object The method of claim 2,
The moving distance measuring unit
And calculating the first distance according to Equation 2 below, when the reference section is continuous in the image frames to be compared, using the non-harmonic ratio.
[Equation 2]
D1 = L
D1: first distance, L: distance between the first reference point and the second reference point (reference section) The method of claim 2,
The moving distance measuring unit
And calculating the first distance according to Equation 3 below when the reference interval is spaced from the image frames to be compared, using the non-harmonic ratio.
[Equation 3]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object According to claim 1,
The first moving distance calculator
And calculating the second distance from the previously stored correction function for the first distance, using the non-harmonic ratio. According to claim 1,
And a speed calculator configured to calculate an actual movement speed of the object by using at least one of the second distance and the third distance. The method of claim 7, wherein
The apparatus for measuring the movement information of the object using the non-harmonic ratio, the actual moving speed of the object is calculated according to Equation 4 below.
[Equation 4]

V _{i, j} is the average speed of the object from the i th image frame to the j th image frame, D2 _{i, j} is the second or third distance of the object from the i th image frame to the j th image frame, frames per second ) Is the number of image frames per second, and | ij | is the number of frames between the i th image frame and the j th image frame. According to claim 1,
The movement information extraction unit
In the first environment, the reference point of the object is set to the rotation center position of the front wheel of the vehicle and the rotation center position of the rear wheel,
The apparatus for measuring movement information of an object using a non-harmonic ratio, which sets the reference point of the object to a preset position of the vehicle in the second environment. According to claim 1,
Further comprising; object information storage unit for storing the shape information for each object,
The movement information extraction unit
An apparatus for measuring movement information of an object using a non-harmonic ratio that specifies an object included in the image frame by using the shape information of the object information storage unit. Obtaining a plurality of image frames of the object moving from the camera;
Comparing a plurality of image frames to extract a change in position of a reference point of the object;
Computing a first distance that the object is moved in the image frame to be compared using the change of the position of the reference point and the cross-ratio of the object in the image frame photographed by the camera in a first environment step;
Correcting the first distance to a second distance that is an actual distance that the object moves in a three-dimensional space of the first environment;
Storing a correspondence relationship between the first distance and the second distance as reference data; And
The object is three-dimensional space of the second environment in the image frame comparing the position data of the reference point and the reference data in the image frame photographed by the camera in a second environment having a lower discrimination power than the first environment. And calculating a third distance, which is a real distance moved in, the movement information of the object using the non-harmonic ratio. The method of claim 11, wherein
Calculating the first distance
And determining whether the reference sections, which are distances between the first reference point and the second reference point of the object, overlap each other in the image frames to be compared. The method of claim 12,
Calculating the first distance
When the reference section overlaps in the image frames to be compared, the first distance is calculated according to Equation 1 below.
[Equation 1]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object The method of claim 12,
Calculating the first distance
And calculating the first distance according to Equation 2 below when the reference section is continuous in the image frames to be compared.
[Equation 2]
D1 = L
D1: first distance, L: distance between the first reference point and the second reference point (reference section) The method of claim 12,
Calculating the first distance
And calculating the first distance according to Equation 3 below when the reference section is spaced apart from the image frames to be compared.
[Equation 3]

D1: first distance, L: distance between the first reference point and the second reference point (reference section), R: non-harmonization ratio of the object The method of claim 11, wherein
And calculating an actual moving speed of the object by using at least one of the second distance and the third distance. The method of claim 16,
Calculating the actual moving speed of the object
A method of measuring movement information of an object using a non-harmonic ratio, which calculates the actual movement speed according to Equation 4 below.
[Equation 4]

V _{i, j} is the average speed of the object from the i th image frame to the j th image frame, D2 _{i, j} is the second or third distance of the object from the i th image frame to the j th image frame, frames per second ) Is the number of image frames per second, and | ij | is the number of frames between the i th image frame and the j th image frame. The method of claim 11, wherein
Extracting a change in the position of the reference point of the object
In the first environment, the reference point of the object is set to the rotation center position of the front wheel of the vehicle and the rotation center position of the rear wheel,
And setting the reference point of the object to one preset position of the vehicle in the second environment.

Images