KR101803893B1 - Apparatus for preventing collision of vehicle and method for preventing collision thereof - Google Patents

Abstract

The present invention relates to an image processing apparatus, comprising: a ROI setting unit that sets at least two ROIs from a ROI to a ROI for at least one of a current target vehicle and a current non-target vehicle; The current minutia information corresponding to the external pattern of the current target is detected with respect to the current target among the at least two regions of interest set by the region of interest setting section and the outline pattern of the current target is detected based on the current minutia information, (Cell), and detecting the current image information of the other vehicle; A control unit receiving at least one of the current feature point information detected by the detection unit and the image information of the current vehicle, and including a preset reference feature point information, an image information of the reference target, and an access information range with the reference target; Determining whether at least one of the current feature point information detected by the detection unit and the image information of the current vehicle is at least one of the reference feature point information and the image information of the reference vehicle under the control of the control unit, A judging unit for judging whether the vehicle is in a range of access information with the vehicle; And an identification unit for identifying the collision situation with the current vehicle if the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle. .

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a vehicle collision avoidance apparatus,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001]

In general, since the conventional vehicle can not recognize the access information with the current vehicle at the time of traveling, it has been necessary to judge the access situation with the current vehicle based on the experience value of the driver.

Particularly, in the conventional vehicle, there is a limit in recognizing the access information with the current vehicle according to the meteorological conditions such as rain, snow, etc., There was a limit to judging the access situation.

Therefore, in recent years, there has been proposed an improved vehicle collision avoidance device capable of inducing driver's attention while preventing the occurrence of a traffic accident by identifying the collision situation with the current vehicle regardless of the weather condition at the time of driving, Research on prevention methods has been continuously carried out.

SUMMARY OF THE INVENTION An object of the present invention is to provide a vehicle collision avoidance apparatus and a collision avoidance method thereof capable of preventing the occurrence of a traffic accident in advance.

It is another object of the present invention to provide a vehicle collision avoidance apparatus and a collision avoidance method that can induce driver's attention driving.

In order to achieve the above object, the present invention provides a vehicle navigation system comprising: a ROI setting unit that sets at least two ROIs from a ROI to a ROI for a current vehicle and a current non-vehicle in a forward situation photographed by a photographing unit; The current feature point information corresponding to the external pattern of the current vehicle is detected for the current and non-current at least two regions of interest set by the region-of-interest setting unit, and based on the current feature point information, A unit cell, and detecting the image information of the current vehicle; A control unit receiving at least one of the current feature point information detected by the detection unit and the image information of the current vehicle, and including a preset reference feature point information, an image information of the reference target, and an access information range with the reference target; The control unit determines whether at least one of the current feature point information detected by the detection unit and the image information of the current target is at least one of the reference feature point information and the image information of the reference target, A determination unit for determining whether or not the range of access information with the other vehicle is within a range; And a first identification unit for identifying, when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, a collision situation with the current vehicle.

According to another aspect of the present invention, the ROI setting unit sets at least two ROIs from at least one of a current ROI and a current ROI to a ROI from a ROI to a ROI in a cascade manner, Window).

According to another aspect of the present invention, the detecting unit applies a horizontal / vertical edge mask to at least two of the ROIs currently set in the ROI setting unit, Calculates a size and an angle, and detects the current feature point information corresponding to the contour pattern of the current target including the size and angle of the calculated edge by a SVM (Support Vector Machine) method.

According to another aspect of the present invention, the detecting unit divides the external pattern of the current vehicle into the unit cells based on the current feature point information, calculates a histogram of the unit cells, And detects the image information of the current vehicle including the histogram of the unit cell.

According to another aspect of the present invention, the detecting unit divides the contour pattern of the current vehicle into a unit cell based on the current feature point information and detects the current contiguous vehicle as the image information of the current vehicle It estimates the lost part of the current track tracked for a certain period of time while tracking, and removes the noise component of the current car using Kalman filter, and detects the current track estimated as the lost part as image information .

According to another aspect of the present invention, the control unit further includes a reference speed adjustment information range that is already set; The control unit may further include a speed control unit for controlling the speed of the present vehicle in accordance with the reference speed control information range under the control of the control unit when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle do.

According to another aspect of the present invention, when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, the controller adjusts the current vehicle speed in accordance with the reference speed control information range And a second identification unit for identifying the second identification unit.

The present invention is also directed to an attention area setting step of setting at least two ROIs in a range from a ROI to a ROI with respect to a current target vehicle and a current non- ; The current minutia information corresponding to the contour pattern of the current target is detected from the at least two regions of interest set by the region of interest setting unit and detected by the detecting unit, A detecting step of dividing the outer shape pattern into unit cells and detecting the image information of the current other vehicle by the detecting unit; The determination unit determines whether at least one of the current feature point information detected by the detection unit and the image information of the current vehicle is at least one of the reference feature point information set in the control unit and the image information of the reference target, A determination step of determining, by the control unit, whether the access information with respect to the reference vehicle is within the access information range set in the control unit; And a first discriminating step of discriminating, by the first discriminating unit, the collision situation with the current car if the judging unit judges that the access information with the current car is the access information range with the reference car.

According to another aspect of the present invention, the ROI setting unit sets at least two ROIs from the ROI to the ROI for at least one of the current vehicle and the current non- (Scan Window) according to the method of the present invention.

According to another aspect of the present invention, in the detecting step, a horizontal / vertical edge mask is applied to the current vehicle and the current non-vehicle among at least two ROIs set by the ROI setting unit, And detecting the current feature point information corresponding to the outline pattern of the current vehicle including the size and angle of the edge calculated by the detection unit by the SVM (Support Vector Machine) method .

According to another aspect of the present invention, in the detecting step, the outer shape pattern of the current vehicle is divided into unit cells on the basis of the current feature point information in the detection unit, and a histogram of the unit cells is detected in the detection unit. And detecting the image information of the current vehicle including the histogram of the unit cell calculated by the detecting unit.

According to still another aspect of the present invention, in the detecting step, when the detecting unit divides the external pattern of the current vehicle into the unit cells based on the current feature point information and detects the current pattern as the image information of the current vehicle, It estimates the lost part of the current track tracked for a certain time while tracing the present current track and removes the noise component of the current track using Kalman filter in the detection part, And detecting the other vehicle based on the image information.

According to another aspect of the present invention, when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle after the determination, And a speed adjustment step of adjusting the speed of the present vehicle in accordance with the adjustment information range.

According to another aspect of the present invention, when the determination unit determines that the access information with the current vehicle is the access information range with respect to the reference vehicle after the determination step, And a second identification step of identifying, by the second identification unit, that the status is a condition for controlling the speed.

According to the vehicle collision avoidance apparatus and the collision avoidance method of the present invention as described above, the following effects can be obtained.

First, there is an effect that the occurrence of a traffic accident can be prevented in advance.

Second, there are other effects that can drive driving attention to the driver.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an apparatus for preventing collision of a vehicle according to a first embodiment of the present invention; FIG.
FIG. 2 is a front view illustrating a process of setting a region of interest through the ROI setting unit shown in FIG. 1; FIG.
FIG. 3 and FIG. 4 show current feature point information corresponding to the external pattern of the current vehicle with respect to the current vehicle and the current non-vehicle among the ROIs through the ROI setting unit shown in FIG. 1, It shows the process of detecting by the image information of the vehicle.
FIG. 5 is a flowchart illustrating a process of applying a horizontal / vertical edge mask to a current vehicle and a current non-vehicle through the ROI setting unit shown in FIG. 1 to obtain current minutia information corresponding to an external pattern of the current vehicle A front view to show the process of detection.
6 is a flowchart illustrating a process of detecting a current vehicle estimated as a lost part as image information while removing a noise component of a current vehicle using a Kalman filter through the detection unit shown in FIG. Front view.
FIG. 7 is a front view showing the reference minutia information set in the control unit shown in FIG. 1 and the image information of the reference target.
FIG. 8 is a front view illustrating a process of detecting the current minutia information among the image information of the current target through the determination unit shown in FIG. 1 and measuring access information with the current target.
9 is a flowchart showing a collision avoiding method of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.
10 is a flowchart showing an example of a method for preventing collision of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.
11 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.
12 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.
13 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.
14 is a block diagram showing a vehicle collision avoidance apparatus according to a second embodiment of the present invention;
15 is a flowchart showing a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.
16 is a flowchart showing an example of a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.
17 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.
18 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.
19 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.
20 is a block diagram showing a vehicle collision avoidance apparatus according to a third embodiment of the present invention.
FIG. 21 is a flowchart showing a collision avoidance method of a vehicle collision avoidance apparatus according to a third embodiment of the present invention. FIG.
22 is a flowchart showing an example of a collision avoidance method of a vehicle collision avoidance apparatus according to a third embodiment of the present invention.
23 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a third embodiment of the present invention.
24 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a third embodiment of the present invention.
FIG. 25 is a flowchart showing still another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a third embodiment of the present invention. FIG.

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

≪ Embodiment 1 >

FIG. 1 is a block diagram illustrating a vehicle collision avoidance apparatus according to a first embodiment of the present invention. FIG. 2 is a front view illustrating a process of setting a region of interest through the region of interest setting unit shown in FIG.

FIG. 3 and FIG. 4 show current feature point information corresponding to the external pattern of the current vehicle with respect to the current vehicle and the current non-vehicle among the ROIs through the ROI setting unit shown in FIG. 1, FIG. 2 is a front view showing a process of detecting a vehicle image information. FIG.

FIG. 5 is a flowchart illustrating a process of applying a horizontal / vertical edge mask to a current vehicle and a current non-vehicle through the ROI setting unit shown in FIG. 1 to obtain current minutia information corresponding to an external pattern of the current vehicle And FIG.

6 is a flowchart illustrating a process of detecting a current vehicle estimated as a lost part as image information while removing a noise component of a current vehicle using a Kalman filter through the detection unit shown in FIG. Front view.

FIG. 7 is a front view showing the reference minutia information set in the control unit shown in FIG. 1 and the image information of the reference target. FIG. 8 is a view showing the current minutia information in the image information of the current car through the determination unit shown in FIG. It is a front view for showing the process of measuring access information with current vehicles.

1 to 8, a vehicle collision avoidance apparatus 100 according to a first embodiment of the present invention includes a region of interest setting unit 102, a detection unit 104, a control unit 106, and a determination unit 108. [ And a first identification unit (110).

As shown in FIGS. 1 and 2, the ROI setting unit 102 sets a ROI from the ROI (A) to the ROI (ROI) Is provided to set up at least two regions of interest (A, B) up to the region B.

For example, the ROI setting unit 102 may determine at least two ROIs (A and B) from the ROI (A) to the ROI (B) And may be provided to be set in a scan window in a cascade manner.

As shown in FIGS. 1, 3 and 4, the detection unit 104 detects current and non-current at least two of the at least two regions of interest A and B set by the region-of-interest setting unit 102, P2, P3, P4, and P5 corresponding to the outline pattern of the other vehicle, and based on the current feature point information P1, P2, P3, P4, and P5, Is divided into unit cells (Cells) and is provided to detect the image information of the current vehicle.

For example, as shown in FIG. 1 and FIG. 5, the detection unit 104 detects a current position of a current non-vehicle among the at least two regions of interest (A, B) set by the region- The edge size and angle are calculated by applying a vertical edge mask and the current feature point information P6, which corresponds to the contour pattern of the current rider including the calculated size and angle of the edge, P7) by a SVM (Support Vector Machine) method.

Here, the method of applying a horizontal / vertical edge mask is a criterion for determining the complexity of the complex part and the simple part of the image, and the mask width for the boundary matching can be adjusted.

In this case, in the method of applying the horizontal / vertical edge mask, the edge code block can be discriminated by using the edge threshold information and the edge mask information of the code block unit in four directions (east, west, south, north) .

1 and 5, the detecting unit 104 divides the external pattern of the current vehicle into unit cells based on the current feature point information P6 and P7, ), And can be provided to detect the image information of the current target including the histogram of the calculated unit cell (Cell).

As shown in FIGS. 1 and 6, the detecting unit 104 divides the current external pattern of the current vehicle into unit cells based on the current feature point information P8, P9, P10, and P11 (P12) of the current track that has been tracked for a certain time while tracking the currently displayed current track while detecting the current image information of the current track, and then, using the Kalman filter, It can be provided to detect the current vehicle estimated as the lost portion P12 as the image information while removing the noise component P13 of the vehicle.

As shown in FIGS. 1 to 7, the control unit 106 detects the current minutia information (P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11) detected by the detecting unit 104 (S1a, S1b, S2a, S2b, S3a, S3b, S4a, S4b, S5a, S5b, S6a, and S6b) and the image information of the reference target vehicle (S1, S2, S3, S4, S5) and a reference information area.

The determination unit 108 determines whether at least one of the current minutia information P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11 detected by the detection unit 104 and the video information of the current target vehicle S2a, S2b, S3a, S3b, S4a, S4b, S5a, S5b, S6a, and S6b and the reference image information S1, S2, S3, S4, According to the control of the control unit 106, and is provided to judge whether the access information with the current vehicle is in the access information range with the reference vehicle.

For example, as shown in FIGS. 1 and 8, the determination unit 108 determines whether or not the current access point information 52.7 m (P10, P10) is present based on the current minutia information P9, P10, and P11 detected by the detection unit 104 , 35.3m, 31.6m) can be provided to judge whether it is the access information range with the reference other vehicle.

Although not shown, the control unit 106 and the determination unit 108 may be implemented by a general ECU (Electric Control Unit) (not shown) that controls and determines the overall operation of the vehicle by a main computer applied to the vehicle, And a general microcontroller (MCU) (not shown) that controls and judges the overall operation by having an input / output device. However, the present invention is not limited to this, and all of the judgment control means It is possible.

At this time, the control unit 106 and the determination unit 108 may be provided integrally with an ECU (Electric Control Unit) (not shown) or an MCU (Micro Control Unit, not shown) Can be provided.

If the determination unit 108 determines that the access information with the current vehicle is in the range of the access information with the reference vehicle, the first identification unit 110 is provided to identify the current collision situation with the other vehicle.

Although not shown, the first identification unit 110 may include a speaker (not shown) and a light emitting member (not shown), which are provided for the driver to identify the information or state of the vehicle, and a user, (Not shown) including at least one of an HMI (Human Machine Interface) module (not shown) and a HUD (Head-Up Display) module (not shown) HUD (Human-Machine Interface) message display operation of an HMI (Human Machine Interface) module (not shown) and a HUD (Head-UP Display) message of a HUD It is possible to identify the collision situation with the current vehicle through at least one of the display operations.

A method for preventing collision of the vehicle collision avoidance apparatus 100 according to the first embodiment of the present invention will now be described with reference to FIGS. 9 to 13. FIG.

FIG. 9 is a flowchart showing a method of preventing a collision of a vehicle collision avoidance apparatus according to a first embodiment of the present invention, and FIG. 10 is an example of a collision avoidance method of a collision avoidance apparatus of a vehicle according to the first embodiment of the present invention Fig.

FIG. 11 is a flow chart showing another example of a collision avoidance method of a collision avoidance apparatus for a vehicle according to the first embodiment of the present invention, FIG. 12 is a flowchart illustrating a collision avoidance method of a collision avoidance apparatus of a vehicle according to the first embodiment of the present invention As another example.

13 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to the first embodiment of the present invention.

9 to 13, the collision avoidance method 900, 1000, 1100, 1200, 1300 of the vehicle collision avoidance apparatus (100 of FIG. 1) according to the first embodiment of the present invention includes: S902, S1002), a detection step (S904, S1104, S1204, S1304), a determination step (S906), and a first identification step (S908).

First, the ROI setting step (S902) is a step of setting a ROI (ROI) from a near region of interest (A in FIG. 2) to a current region of interest 2) to the region of interest (102 in FIG. 1) of at least two regions of interest (A, B in FIG. 2)

For example, as shown in FIG. 10, the ROI setting step S1002 is a step of setting a ROI (at 102 in FIG. 1) to a ROI (at least one of current and non- (A and B in FIG. 2) from the first to the second interest regions (FIG. 2B) to the long-distance interest region (FIG. 2B) by a scan window in a cascade manner.

Thereafter, the detecting step S904 is performed to determine whether or not the present and other non-vehicles of the current one of the at least two areas of interest (A and B in Figs. 2 and 3) set by the region- P1, P2, P3, P4, and P5 in Fig. 4) of the current feature point information (P1, P2, P3, P4, and P5 in Fig. 4) , The external pattern of the current vehicle is divided into unit cells and detected by the detection unit 104 of FIG. 1 as image information of the current vehicle.

For example, as shown in Fig. 11, the detecting step S1104 may include detecting the current one of the at least two regions of interest (A and B in Fig. 2 and Fig. 3) set in the ROI setting unit 102 A horizontal / vertical edge mask is applied to the current non-vehicle by a detection unit 104 in FIG. 1 to calculate the size and angle of the edge, and the edge calculated at the detection unit (104 in FIG. 1) (P6, P7 in Fig. 5) corresponding to the contour pattern of the current target, including the size and angle of the current vehicle, by using the SVM (Support Vector Machine) method.

12, the detecting step (S1204) may detect the contour pattern of the current vehicle on the basis of the current minutia information (P6, P7 in Fig. 5) in the detecting unit (104 in Fig. 1) The histogram of the unit cell in the detection unit 104 in FIG. 1 is calculated and the histogram of the unit cell calculated in the detection unit 104 in FIG. And detecting the image information of the current vehicle including the current vehicle.

13, the detecting step (S1304) divides the external pattern of the current vehicle into unit cells (Cells) based on the current minutia information in the detecting unit (104 of FIG. 1) (P12 in Fig. 6) of the current track tracked for a predetermined time while tracking the current track continuously appearing in the detection unit (104 in Fig. 1) (P12 in Fig. 6) while removing the noise component (P13 in Fig. 6) of the current rider by using a Kalman filter in the current rider Step.

P0, P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11 in Fig. 1 to Fig. 7) detected by the detection unit S1a, S1b, S2a, S2b, S3a, S3b, S4a, S4b, S5a, S5b, S6a, and S5b in the control unit (106 in Fig. 1) (108 in Fig. 1) in accordance with the control of the control unit (106 in Fig. 1) to determine whether at least one of the image data (S6a, S6b) of the reference vehicle and the image information (S1, S2, S3, S4, , The determination unit (108 in FIG. 1) determines whether the access information with the current vehicle is the access information range with the reference vehicle set in the control unit (106 in FIG. 1) under the control of the control unit .

For example, in the determination step S906, the determination unit (108 in FIG. 1) determines whether or not the current minutia information is present in the current minutia based on the current minutia information (P9, P10, P11 in FIG. 8) It is possible to perform the step of determining whether the access information (52.7m, 35.3m, 31.6m in Fig. 8) is in the access information range with the reference vehicle.

1), the first identification step (S908) determines that the access information with the current vehicle is the access information range with the reference vehicle in the determination unit (108 in FIG. 1) And performs a step of identifying the first identification unit (110 of FIG. 1).

The collision avoidance apparatus 100 and the collision avoidance methods 900, 1000, 1100, 1200, and 1300 of the vehicle according to the first embodiment of the present invention may include the area-of-interest setting unit 102, the detection unit 104, (S902, S1002), a detection step (S904, S1104, S1204, S1304), a determination step (S906), and a determination step (S906), including the control unit 106, the determination unit 108, and the first identification unit 110 1 identification step S908.

Therefore, the collision avoidance apparatus 100 and the collision avoidance methods 900, 1000, 1100, 1200, and 1300 of the vehicle according to the first embodiment of the present invention are capable of preventing a collision situation with a current vehicle It is possible to prevent the occurrence of a traffic accident beforehand.

≪ Embodiment 2 >

14 is a block diagram showing a vehicle collision avoidance apparatus according to a second embodiment of the present invention.

Referring to FIG. 14, the vehicle collision avoidance apparatus 1400 according to the second embodiment of the present invention includes a region-of-interest setting unit 102 and a detection unit 104, similar to the vehicle collision avoidance apparatus 100 according to the first embodiment. A controller 104, a controller 1406, a determination unit 108, and a first identification unit 110.

The function of each component corresponding to the vehicle collision avoidance device 1400 according to the second embodiment of the present invention and the organic connection relation between them are the same as those of the collision avoidance device of the vehicle according to the first embodiment 100), and the organic connection relation between them, each of which will be omitted below.

Here, the vehicle collision avoidance apparatus 1400 according to the second embodiment of the present invention further includes a speed adjusting section 1412, and the control section 1406 further includes a predetermined reference speed adjusting information range.

That is, when the determination unit 108 determines that the access information with the current vehicle is the access information range with respect to the reference vehicle, the speed control unit 1412 sets the reference speed control information range To adjust the speed of the current vehicle.

A method of preventing collision of the vehicle collision avoidance apparatus 1400 according to the second embodiment of the present invention will now be described with reference to FIGS. 15 to 19. FIG.

FIG. 15 is a flowchart showing a method of preventing collision of a vehicle collision avoidance apparatus according to a second embodiment of the present invention, and FIG. 16 is an example of a collision avoidance method of a collision avoidance apparatus of a vehicle according to the second embodiment of the present invention Fig.

FIG. 17 is a flow chart showing another example of a collision avoidance method of a collision avoidance apparatus for a vehicle according to a second embodiment of the present invention, FIG. 18 is a flowchart illustrating a collision avoidance method of a collision avoidance apparatus for a vehicle according to a second embodiment of the present invention As another example.

FIG. 19 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to a second embodiment of the present invention.

15 to 19, the collision avoidance method 1500, 1600, 1700, 1800, 1900 of the vehicle collision avoidance apparatus (1400 of FIG. 14) according to the second embodiment of the present invention is similar to the first embodiment (S902, S1002) and a detection step (S902) in the same manner as the collision avoidance method (900, 1000, 1100, 1200, 1300 in Figs. 9 to 13) , S1104, S1204, and S1304), a determination step (S906), and a first identification step (S908).

The function for each of the steps corresponding to the anti-collision method (1500, 1600, 1700, 1800, 1900) of the vehicle collision avoidance apparatus (1400 in FIG. 14) according to the second embodiment of the present invention, The organic connection relationship is the function for each step corresponding to the collision avoidance method (900, 1000, 1100, 1200, 1300 in Figs. 9 to 13) of the vehicle collision avoidance apparatus 100 according to the first embodiment And the organic connection relationship therebetween, and therefore, respective additional explanations thereof will be omitted below.

Here, the collision avoidance method (1500, 1600, 1700, 1800, 1900) of the vehicle collision avoidance apparatus (1400 of FIG. 14) according to the second embodiment of the present invention, after the judgment step S906, (S1510, S1610, S1710, S1810, S1910).

For example, the anti-collision method 1500, 1600, 1700, 1800, 1900 of the vehicle collision avoidance apparatus 1400 (FIG. 14) according to the second embodiment of the present invention, after the first identification step S908, (Steps S1510, S1610, S1710, S1810, and S1910).

14) of the vehicle collision avoidance apparatus (1400 in Fig. 14) according to the second embodiment of the present invention is not shown, but may include a first discriminating step (not shown) after the judging step (Not shown) in synchronism with the speed control step (not shown).

14) of the vehicle collision avoidance apparatus (1400 in FIG. 14) according to the second embodiment of the present invention is not shown, but after the determination step (not shown) and the first identification step (Not shown), a speed adjustment step (not shown) may be further performed.

That is, when the determination unit (108 in FIG. 14) determines that the access information with the current vehicle is the access information range with the reference vehicle, the speed adjustment steps S1510, S1610, S1710, S1810, (1406 in FIG. 14) to adjust the speed of the present vehicle in accordance with the reference speed control information range in the speed controller (1412 in FIG. 14).

The vehicle collision avoidance apparatus 1400 and the collision avoidance methods 1500, 1600, 1700, 1800, and 1900 according to the second embodiment of the present invention may include a region of interest setting unit 102, a detection unit 104, (S902, S1002), a detection step (S904, S1104, S1204, S1304) including the control unit 106, the determination unit 108, the first identification unit 110 and the speed adjustment unit 1412, A determination step S906, a first identification step S908, and a speed adjustment step S1510, S1610, S1710, S1810, and S1910 are performed.

Therefore, the vehicle collision avoidance apparatus 1400 and the collision avoidance methods 1500, 1600, 1700, 1800, and 1900 according to the second embodiment of the present invention are capable of preventing a collision situation with a current vehicle It is possible to prevent the occurrence of a traffic accident beforehand.

The collision avoidance apparatus 1400 and the collision avoidance methods 1500, 1600, 1700, 1800, and 1900 of the vehicle according to the second embodiment of the present invention may be configured such that the access information with respect to the currently- In the information range, the speed of the present vehicle can be adjusted in accordance with the reference speed control information range, so that the occurrence of a traffic accident can be prevented in advance.

≪ Third Embodiment >

20 is a block diagram showing a vehicle collision avoidance apparatus according to a third embodiment of the present invention.

Referring to FIG. 20, the vehicle collision avoidance apparatus 2000 according to the third embodiment of the present invention includes a region of interest setting unit 102 and a detection unit 130, similar to the vehicle collision avoidance apparatus 1400 according to the second embodiment. A control unit 1406, a determination unit 108, a first identification unit 110, and a speed control unit 1412.

The function of each component corresponding to the vehicle collision avoidance apparatus 2000 according to the third embodiment of the present invention and the organic connection relation between them are identical to those of the collision avoidance apparatus of the vehicle according to the second embodiment 1400), and the organic connection relation between them, each of which will be omitted below.

Here, the vehicle collision avoidance apparatus 2000 according to the third embodiment of the present invention further includes a second identification portion 2014.

That is, when the determination unit 108 determines that the access information with the current vehicle is the access information range with respect to the reference vehicle, the second identification unit 2014 sets the speed of the current vehicle To identify the situation in which it is controlled.

Although not shown, the second identification unit 2014 may include a speaker (not shown) and a light emitting member (not shown), which are provided for the driver to identify the information or status of the vehicle, and a user interface with the user, (Not shown) including at least one of an HMI (Human Machine Interface) module (not shown) and a HUD (Head-Up Display) module (not shown) HUD (Human-Machine Interface) message display operation of an HMI (Human Machine Interface) module (not shown) and a HUD (Head-UP Display) message of a HUD It is possible to identify the situation in which the current vehicle speed is adjusted in accordance with the reference speed control information range through at least one operation of the display operation.

A method of preventing collision of the vehicle collision avoidance apparatus 2000 according to the third embodiment of the present invention will be described with reference to FIGS. 21 to 25. FIG.

FIG. 21 is a flowchart showing a collision avoidance method of a collision avoidance apparatus of a vehicle according to a third embodiment of the present invention, and FIG. 22 is an example of a collision avoidance method of a collision avoidance apparatus of a vehicle according to a third embodiment of the present invention Fig.

23 is a flowchart showing another example of the collision avoidance method of the collision avoidance apparatus of the vehicle according to the third embodiment of the present invention, and Fig. 24 is a flowchart illustrating the collision avoidance method of the collision avoidance apparatus of the vehicle according to the third embodiment of the present invention As another example.

25 is a flowchart showing another example of a collision avoidance method of a vehicle collision avoidance apparatus according to the third embodiment of the present invention.

21 to 25, the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention are similar to those of the second embodiment (Step S902, S1004, S1104, S1204, S1304, and S1004) in the same manner as the collision avoidance method (1500, 1600, 1700, 1800, 1900) , A determination step S906, a first identification step S908, and a speed adjustment step S1510, S1610, S1710, S1810, and S1910.

The function for each step corresponding to the anti-collision method (2100, 2200, 2300, 2400, 2500) of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention, The organic connection relationship is a function for each step corresponding to the anti-collision method 1500, 1600, 1700, 1800, 1900 of the vehicle collision avoidance apparatus 1400 (FIG. 14) according to the second embodiment, Since the organic connection relationship is the same, each of the further explanations thereof will be omitted below.

Here, the anti-collision methods 2100, 2200, 2300, 2400, and 2500 of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention, after the determination step S906, Steps S2109, S2209, S2309, S2409, and S2509 are further performed.

For example, the anti-collision methods 2100, 2200, 2300, 2400, and 2500 of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention may include, after the first identification step S908, The second identification step S2109, S2209, S2309, S2409, and S2509 may be performed before the steps S1510, S1610, S1710, S1810, and S1910.

As another example, although the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention are not shown, The second identification step (not shown) can be further performed.

As another example, the anti-collision methods 2100, 2200, 2300, 2400, and 2500 of the vehicle collision avoidance apparatus (2000 in FIG. 20) according to the third embodiment of the present invention are not shown, The second identification step (not shown) can be further performed.

That is, when the determination unit (108 in FIG. 20) determines that the access information with the current vehicle is in the access information range with the reference vehicle, the second identification step S2109, S2209, S2309, S2409, (Step 2014 in FIG. 20) that the current state of the subject vehicle is adjusted in accordance with the speed control information range.

The vehicle collision avoidance apparatus 2000 and the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 according to the third embodiment of the present invention can be applied to the collision avoidance apparatus 2000 and the detection unit 104, (S902, S1002) and a detection step (S904), including the control unit 106, the determination unit 108, the first identification unit 110, the speed adjustment unit 1412, and the second identification unit 2014, S1510, S1710, S1810, S1810, S1510, S1510, S1510, S1510, S1610, S1610, S1910).

Therefore, the vehicle collision avoidance apparatus 2000 and the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 according to the third embodiment of the present invention are capable of preventing a collision situation with a current vehicle It is possible to prevent the occurrence of a traffic accident beforehand.

In addition, the vehicle collision avoidance apparatus 2000 and the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 according to the third embodiment of the present invention may be configured such that the access information with respect to the present- In the information range, the speed of the present vehicle can be adjusted in accordance with the reference speed control information range, so that the occurrence of a traffic accident can be prevented in advance.

Further, the vehicle collision avoidance apparatus 2000 and the collision avoidance methods 2100, 2200, 2300, 2400, and 2500 according to the third embodiment of the present invention may be configured such that the access information with respect to the current vehicle is close to the reference vehicle In the case of the information range, it is possible to identify the situation in which the speed of the current vehicle is adjusted in accordance with the reference speed control information range.

It will be understood by those skilled in the art that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. Therefore, it should be understood that the above-described embodiments are to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than the foregoing description, It is intended that all changes and modifications derived from the equivalent concept be included within the scope of the present invention.

Claims (14)

A ROI setting unit that sets at least two ROIs from a ROI to a ROI for a current vehicle and a current non-vehicle in a forward situation photographed by a photographing unit;
Detecting current feature point information corresponding to an appearance pattern of the current target vehicle with respect to the current target vehicle and the current non-target vehicle among at least two ROIs set by the ROI setting unit, A detector for dividing the external pattern of the current vehicle into unit cells and detecting the external pattern as image information of the current vehicle;
A controller receiving at least one of the current feature point information detected by the detection unit and the image information of the current vehicle, and including a preset reference feature point information, an image information of the reference target, and an access information range with the reference target;
Determining whether at least one of the current minutia information detected by the detecting unit and the video information of the current target is at least one of the reference minutia information and the video information of the reference target under the control of the controller, A determination unit for determining whether the access information with the vehicle is in the access information range with the reference vehicle; And
And a first identification unit for identifying, when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, the collision state with the current vehicle, .
The method according to claim 1,
Wherein the ROI setting unit comprises:
Wherein at least two ROIs from at least one ROI to at least one ROI are set in a scan window in a cascade manner for at least one of the current target vehicle and the current non- Of the vehicle.
The method according to claim 1,
Wherein:
Calculating a size and an angle of an edge by applying a horizontal / vertical edge mask to the current riding vehicle and the current non-riding vehicle among the at least two ROIs set by the ROI setting unit, Wherein the current feature point information corresponding to the external pattern of the current vehicle including the size and angle of the edge is detected by SVM (Support Vector Machine).
The method of claim 3,
Wherein:
A histogram of the unit cell is calculated based on the current feature point information, the outline pattern of the current target is divided into unit cells, a histogram of the unit cell is calculated, and detecting the image information of the current vehicle including the histogram.
The method according to claim 1,
Wherein:
Wherein when the external pattern of the current vehicle is divided into unit cells and the current pattern is detected as the image information of the current vehicle based on the current feature point information, Estimating a lost part of the current riding vehicle and detecting a current riding vehicle estimated as the lost part as image information while removing a noise component of the current riding vehicle using a Kalman filter A device for preventing collision of a vehicle.
The method according to claim 1,
Wherein the control unit further includes a reference speed adjustment information range that is already set;
The control unit controls the current speed of the current vehicle in accordance with the reference speed control information range when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, Further comprising a speed adjusting section (20).
The method according to claim 6,
And a control unit for controlling the speed of the current vehicle according to the reference speed control information range when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, 2 identification unit.
A ROI setting step of setting at least two ROIs from a ROI to a ROI for a current non-current vehicle and a current non-ROI in a forward situation photographed by a photographing unit;
Detecting a current feature point information corresponding to a contour pattern of the current target relative to the current target vehicle and the current non-target vehicle among at least two ROIs set by the ROI setting unit, A detecting step of dividing the external pattern of the current vehicle into unit cells and detecting the image information of the current vehicle by the detecting unit;
The determination unit determines whether at least one of the current feature point information detected by the detection unit and the image information of the current target is at least one of reference feature point information set in the control unit and image information of the reference target, A determining step of determining, by the control unit, whether the access information with respect to the current vehicle is an access information range with respect to the reference vehicle set in the control unit; And
And a first identification step of identifying, by the first identification unit, that the collision situation with the current vehicle is detected when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle Of the vehicle.
9. The method of claim 8,
The method of claim 1,
Wherein the ROI setting unit sets at least two ROIs in at least one of the current ROI and the current ROI from the ROI to the ROI in a cascade manner, Wherein the step of setting the collision avoidance flag is performed.
9. The method of claim 8,
Wherein the detecting step comprises:
Calculating a size and an angle of an edge by applying a horizontal / vertical edge mask to the current vehicle and the current non-vehicle among the at least two ROIs set by the ROI setting unit; And detecting the current feature point information corresponding to the contour pattern of the current vehicle including the size and angle of the edge calculated by the detection unit by SVM (Support Vector Machine) method. How to prevent.
11. The method of claim 10,
Wherein the detecting step comprises:
Wherein the detecting unit divides the external pattern of the current target into unit cells based on the current feature point information, calculates a histogram of the unit cells in the detection unit, Wherein the step of detecting a collision of the vehicle comprises the step of detecting the image information of the current vehicle including the histogram of the unit cell.
9. The method of claim 8,
Wherein the detecting step comprises:
Wherein the detection unit divides the external pattern of the current vehicle into unit cells based on the current feature point information to detect the current vehicle continuously appearing in the detection unit as the image information of the current vehicle, Estimates a lost portion of the current track tracked for a predetermined time, and removes a noise component of the current track from the detected portion using a Kalman filter to detect a current track estimated as the lost portion Wherein the step of detecting the collision of the vehicle comprises the step of detecting the collision of the vehicle with the image information.
9. The method of claim 8,
After the determining step,
When the determination unit determines that the access information with the current vehicle is the access information range with respect to the reference vehicle, the speed controller adjusts the speed of the current vehicle according to the reference speed control information range under the control of the controller, And a speed adjustment step of adjusting the speed of the vehicle.
14. The method of claim 13,
After the determining step,
Wherein the control unit controls the speed of the current vehicle according to the reference speed control information range when the determination unit determines that the access information with the current vehicle is the access information range with the reference vehicle, And a second identification step of identifying the vehicle in the vehicle.

Images