KR102079291B1 - Forward vehicle collision warning apparatus and method thereof - Google Patents

Abstract

The present invention relates to a front vehicle collision warning device and method, and more particularly, both ends of the front vehicle at the rear of the front vehicle detected by the front image captured by the camera of the navigation device or the smart terminal. By extracting a plurality of feature points representative of, it is possible to calculate one or more of the relative speed and the relative distance with the front vehicle based on the change value of the distance between the feature points. The accuracy of the relative speed measurement and collision warning is very high even in poor driving conditions such as snow, rain, and bad weather, and the relative speed is calculated by periodically calculating the change in the interval between the feature points over time. Accurately calculate the instantaneous relative speeds for driving situations, Even in the same driving situation, the measurement accuracy of the relative speed and the relative distance is further increased, and the relative speed is calculated based on the change value of the interval between the feature points, so that the exact instantaneous relative speed can be calculated even when the distance between the vehicles is very close. The efficiency of collision warnings for forward vehicles present in the vehicle increases.

Description

FORWARD VEHICLE COLLISION WARNING APPARATUS AND METHOD THEREOF

The present invention relates to a front vehicle collision warning device and method, and more particularly, to photograph a vehicle traveling in front of the vehicle through a camera that photographs the front image while driving the vehicle and recognizes the front vehicle in the captured image and the relative to the front vehicle. A front vehicle collision warning device and method for warning a collision with a front vehicle based on speed or relative distance.

The conventional forward collision warning system (FCWS) oscillates an active signal in front of the vehicle, measures the return signal to measure the distance between the front and rear vehicles, and shares the real-time speed of the front vehicle by sharing the M2M. Mutual awareness was typical.

However, such a configuration requires the attachment of a separate expensive device to the vehicle and it is not practical because it requires subscription between members for M2M sharing.

In addition, some of the conventional front vehicle collision warning system has a configuration that warns of a collision for the front vehicle based on the image information of the front vehicle photographed through the camera.

However, such a configuration also requires continuous image analysis of the entire captured image including a road, lane, or background, so that the load of image analysis to obtain a relative speed or relative distance with a vehicle in front is large, and accurate instant relative Not only is it difficult to obtain speed quickly, but it is also very ineffective in bad weather such as difficult weather.

Korean Laid-open Patent No. 10-2013-0007243 [Name: Front collision warning method and system using vehicle camera]

An object of the embodiment of the present invention for improving the above-mentioned problems is a plurality of representative of both ends of the front vehicle at the rear of the front vehicle detected in the front image captured by the camera of the navigation device or the smart terminal A front vehicle collision warning device and method for extracting feature points to calculate at least one of a relative speed and a relative distance with a front vehicle based on a change value of a distance between the feature points.

Another object of the present invention is to provide a front vehicle collision warning device and method for accurately calculating the instantaneous relative speed suitable for the driving situation by calculating the relative speed by periodically calculating the change value of the interval between the feature points over the running time will be.

Another object of the present invention is to provide a front vehicle collision warning device and method for calculating the exact instantaneous relative speed even when the distance between vehicles is very close by calculating the relative speed based on the change value of the interval between the feature points.

Another object of the embodiment of the present invention is to adjust the speed change detection coefficient reflecting the difference in the distance change value between the feature points according to the distance change with the front vehicle through learning to reflect the adjusted speed change detection coefficient for any front vehicle It is to provide a front vehicle collision warning device and method for calculating the relative speed or relative distance.

Another object of the embodiment of the present invention is to adjust the speed change detection coefficients, which are divided according to the classification or type of the front vehicle, by learning by classifying one or more of the classification according to the size of the front vehicle or the detailed type of the front vehicle through learning. It is to provide a front vehicle collision warning device and method.

Another object of the embodiment of the present invention by providing a cruise function through the screen or the vehicle connection interface to maintain at least one of the relative speed and the relative distance with the vehicle ahead based on the calculated relative speed or relative distance It is to provide a front vehicle collision warning device and method for easily implementing a cruise function even a vehicle that is not equipped with a separate cruise function.

Front vehicle collision warning device according to an embodiment of the present invention for achieving the above object is a front vehicle detection unit for detecting the front vehicle in the front image photographed through the camera, the rear of the detected front vehicle (後 尾) A feature point recognizing unit for extracting a plurality of feature points representing both ends of the front vehicle, a vehicle width change calculating unit for calculating a space change value between a plurality of feature points over time, and a distance change value between the feature points And a speed change detector for calculating a relative speed or a relative distance between the vehicle and the vehicle ahead.

The front vehicle collision warning device according to another embodiment of the present invention further includes a collision warning unit that warns a collision with the front vehicle based on the calculated relative speed or relative distance.

 The collision warning unit preferably provides a collision warning on a screen or sound when the relative distance is within a preset safety distance according to a weather or weather condition.

 The collision warning unit may provide a change in the distance between the feature points on the screen or sound.

The front vehicle collision warning device according to another embodiment of the present invention may further include an augmented reality unit displaying one or more of the detected feature points, relative speeds, and relative distances of the front vehicle as augmented reality.

 The feature point recognizer may extract at least one of at least a portion of a headlight of the front vehicle, at least a portion of an outline, and at least a portion of a tire as a feature point.

 The speed change detector may periodically calculate a distance between the vehicle and the vehicle in front of the front image, and calculate a relative distance from the vehicle in front of the vehicle by comparing the distance change according to the change in the distance between the feature points.

In addition, the speed change detector may periodically calculate a relative speed between its own vehicle speed and the front vehicle, and calculate the acceleration or deceleration of the front vehicle according to the absolute speed change of the front vehicle.

On the other hand, the front vehicle collision warning device according to an embodiment of the present invention may be configured in one of a vehicle black box device, navigation device, front surveillance camera device, smart phone, smart terminal, tablet PC.

Front vehicle collision warning device according to an embodiment of the present invention, the front vehicle detection unit for detecting the front vehicle in the front image photographed through the camera, both ends of the front vehicle at the rear of the detected front vehicle A feature point recognizing unit for extracting a plurality of feature points representing a plurality of feature points, a vehicle width change calculator for calculating a distance change value between a plurality of feature points over time of travel, and periodically calculating a first distance between a vehicle and a vehicle in front of the vehicle Compute and compare the interval change value according to the first distance change, and the speed change for the front vehicle by reflecting the difference between the basic interval change value corresponding to the preset second distance change and the interval change value according to the first distance change. The distance change value between a plurality of feature points for any front vehicle by reflecting the speed change detection coefficient adjusted by the learning unit and the learning unit adjusting the detection coefficient. Calculation, and can on the basis of the calculated distance variation value can further include own vehicle and the preceding vehicle and the speed change detection unit for calculating the relative speed or relative distance.

 The learning unit compares the basic interval change value corresponding to the preset second distance change and the interval change value according to the first distance change to learn one or more of classification according to the size of the vehicle or detailed types of the vehicle, and classify the front vehicle. Alternatively, it is desirable to adjust the speed change detection coefficients which are divided according to the type by learning.

 The speed change detection unit may provide a cruise function through a screen or a vehicle connection interface that maintains one or more of the relative speed and the relative distance with the vehicle ahead based on one or more of the relative speed and relative distance calculated. Can be.

According to an embodiment of the present disclosure, a front vehicle collision warning method may include detecting a front vehicle from a front image captured by a camera through a front vehicle detection unit, and detecting the front vehicle through a feature point recognition unit at a rear of the detected front vehicle. Extracting a plurality of feature points representative of both ends of the front vehicle; calculating a distance change value between the plurality of feature points according to the passage of a traveling time through a vehicle width change calculator; and a speed change detector. Calculating a relative speed or a relative distance between the vehicle and the vehicle in front of the vehicle based on the value of the interval change between the feature points.

According to the present invention, the interval change between the feature points by extracting a plurality of feature points representing both ends of the front vehicle from the rear of the front vehicle detected from the front image captured by the navigation device or the camera of the smart terminal By calculating one or more of the relative speed and the relative distance with the vehicle ahead based on the value, it provides easy warning of collision with the vehicle ahead at low cost without the installation of a separate collision warning device, and poor conditions such as snow, rain and bad weather Even in driving conditions, relative speed measurement and collision warning accuracy are very effective.

In addition, by calculating the relative speed by periodically calculating the change value of the interval between the feature points as the running time elapses to accurately calculate the instantaneous relative speed suitable for the driving situation, the relative speed and relative distance even in the driving situation such as slow motion or delay There is an effect that the measurement accuracy is further increased.

In addition, by calculating the relative speed based on the change value of the distance between the feature points, even if the distance between the vehicles is very close to calculate the exact instantaneous relative speed to increase the efficiency of the collision warning for the vehicle in front of the near distance have.

Also, by adjusting the speed change detection coefficient reflecting the difference in the distance change value between the feature points according to the distance change with the front vehicle through learning, the relative speed or relative distance to any front vehicle is reflected by reflecting the adjusted speed change detection coefficient. It is effective to provide accurate collision warning regardless of the mounting type, mounting direction, mounting angle, etc. of the device.

In addition, by classifying one or more of the classification according to the size of the front vehicle or the detailed type of the front vehicle through learning to adjust the speed change detection coefficients, which are divided according to the classification or type of the front vehicle, by learning, the size of the front vehicle Alternatively, there is an effect of providing a more precise crash warning according to the type of the front vehicle.

In addition, a cruise function is provided through the screen or the vehicle connection interface to provide a cruise function for maintaining one or more of the relative speed and the relative distance with the vehicle ahead based on the calculated relative speed or relative distance. Even if the vehicle is not, it is possible to easily implement the cruise function to further increase the convenience of driving the vehicle.

1 is a block diagram of a front vehicle collision warning device according to an embodiment of the present invention.
2 is a block diagram of a front vehicle collision warning device according to another embodiment of the present invention.
3 is an exemplary view of feature point extraction according to an embodiment of the present invention.
4 is an exemplary view of measuring a change in gap between feature points according to an exemplary embodiment of the present invention.
5 is an exemplary view of providing a relative speed and a relative distance with a front vehicle according to an embodiment of the present invention.
6 is a flow chart of a front vehicle collision warning method according to an embodiment of the present invention.
7 is a flow chart of a front vehicle collision warning method according to another embodiment of the present invention.

The present invention as described above will be described in detail with reference to the accompanying drawings and embodiments. In the following description, a representative embodiment of the present invention for achieving the above-described technical problem will be described.

1 is a block diagram of a front vehicle collision warning device 100 according to an embodiment of the present invention.

The front vehicle collision warning device 100 according to an embodiment of the present invention has both ends of the front vehicle 1 at the rear of the front vehicle 1 detected in the front image captured by the camera. Extracting a plurality of feature points representative of and calculating at least one of a relative speed and a relative distance with the front vehicle 1 on the basis of a change value of the interval between the plurality of feature points as the driving time elapses. Warn of collision with the front vehicle 1 based on one or more.

As a preferred embodiment, the front vehicle collision warning device 100 according to an embodiment of the present invention is operated on the basis of augmented reality in a smart terminal, such as, for example, a smart phone, tablet PC without a separate device configuration in the vehicle Detect the change in speed of the front vehicle using image analysis by using the dedicated application to capture the front image captured by the camera, and use it to determine the distance from the front vehicle 1 or the relative speed with the front vehicle 1 The risk of collision can be calculated and calculated.

As another preferred embodiment, the front vehicle collision warning device 100 according to an embodiment of the present invention utilizes the front image recording function of the navigation device or the vehicle black box device possessed by the vehicle occupant or the augmented reality function of the smart terminal. By detecting the speed change of the front vehicle (1) by software can warn of the risk of collision or provide a safe distance service.

Referring to FIG. 1, the front vehicle collision warning device 100 according to an exemplary embodiment of the present disclosure may detect the front vehicle 1 from a front image photographed through a camera, and detect the front vehicle 1. A feature point recognizing unit 113 for extracting a plurality of feature points representing both ends of the front vehicle 1 from the rear of the front vehicle 1 and a change in the interval between the plurality of feature points as the traveling time elapses. A vehicle width change calculator for calculating a value and a speed change detector 114 for calculating a relative speed or a relative distance between the vehicle and the front vehicle 1 based on the interval change value between the plurality of feature points.

As a preferred embodiment described above, the front vehicle collision warning device 100 according to an embodiment of the present invention is the tail of the front vehicle 1 detected from the front image photographed through the camera of the navigation device or the smart terminal. Extracts a plurality of feature points representing both ends of the front vehicle 1, and calculates at least one of a relative speed and a relative distance with the front vehicle 1 based on the interval change value between the feature points.

This makes it easy to provide a collision warning with the front vehicle 1 at a low cost without installing a separate collision warning device, and to change the distance between the feature points of the front vehicle 1 even in poor driving conditions such as snow, rain and bad weather. All you need to do is track the relative speed and measure the relative speed and the accuracy of the collision warning and collision warning is very high.

In addition, the front vehicle collision warning device 100 according to an embodiment of the present invention calculates the relative speed by periodically calculating the change value of the interval between the feature points over the running time, the front vehicle 1 and the Relative speed and relative speed can be calculated accurately based on the change in the distance between the feature points without using relative distance or background change. Measurement accuracy over distance is further increased.

In addition, the configuration that calculates the relative speed based on the change value of the interval between the feature points, even if the distance between the vehicles is very close, it is possible to calculate the exact instantaneous relative speed or relative distance according to the change of the distance present in close proximity The efficiency of the collision warning for the front vehicle 1 is greatly increased.

The front vehicle collision warning device 100 according to an embodiment of the present invention further includes a collision warning unit 116 that warns the collision with the front vehicle 1 based on the calculated relative speed or relative distance.

According to a preferred embodiment, the collision warning unit 116 may provide a collision warning to a screen or sound when the relative distance is within a preset safety distance according to a weather or weather condition.

More preferably, the collision warning unit 116 may provide a change in the interval between the feature points on the screen or sound.

In addition, the front vehicle collision warning device 100 according to an embodiment of the present invention learns the relationship between the vehicle width of the front vehicle 1 and the distance between its own vehicle through the change of the distance between the feature points in augmented reality through learning. Afterwards, the distance between the vehicles can be predicted even with the vehicle width of the vehicle ahead 1 based on the information learned from the distance between the vehicle and the vehicle ahead 1, thereby providing a cruise control function. There is an advantage that can be provided.

As a preferred embodiment, the front vehicle collision warning device 100 according to an embodiment of the present invention according to the enlargement, reduction of the interval between the feature points of the front vehicle 1 obtained from the front image through the black box device installed in the vehicle Hazard signals such as sudden deceleration or rapid acceleration of the front vehicle 1 that is being driven may be measured, or relative distance, safety distance, or collision warning information according to the vehicle width may be provided based on information learned for cruise control.

2 is a block diagram of a front vehicle collision warning device 100 according to another embodiment of the present invention.

2, the front vehicle collision warning device 100 according to another embodiment of the present invention, through the camera unit 155, GPS (Global Positioning System) for taking a front image through the camera, GPS unit 150 for calculating the speed, a front vehicle detection unit 112 for detecting the front vehicle 1 through the image analysis from the front image, the rear image of the detected front vehicle 1 ), A feature point recognizing unit 113 for extracting a plurality of feature points representing both ends of the front vehicle 1, a vehicle width change calculation unit for calculating an interval value between a plurality of feature points varying with the running time, The speed change detection unit 114 that calculates a relative speed or a relative distance between the vehicle and the front vehicle 1 based on the degree of change of the interval value between the plurality of feature points calculated, and one of the calculated relative speed and the relative distance. Collision with the front vehicle 1 on the basis of the above And a collision warning unit 116 for generating warning information and an output unit 140 for outputting the generated collision warning information to a screen 141 (display unit 141) or a sound 142 (audio unit).

In another preferred embodiment, the front vehicle collision warning device 100 according to another embodiment of the present invention is to change the collision warning information generated by the collision warning unit 116 to a preset screen display format or sound providing format to provide Alert information providing unit 117 may be further included.

In addition, the front vehicle collision warning device 100 according to another embodiment of the present invention is an augmented reality unit 111 for displaying one or more of the detected feature point, relative speed, relative distance of the front vehicle 1 on the screen in augmented reality. ) May be further included.

The front vehicle collision warning device 100 according to another embodiment of the present invention may be configured in one of a vehicle black box device, a navigation device, a front surveillance camera device, a smart phone, a smart terminal, and a tablet PC.

The front vehicle collision warning device 100 according to another embodiment of the present invention includes a power supply unit 130 that supplies power or provides a power connection interface, an acceleration sensor and a gyroscope sensor sensing a speed or direction change of the vehicle. It may further include a sensing unit 131 having the same sensor, a recording unit 160 for recording a front image to be photographed, a storage unit for storing the front image to be recorded, and a user input unit 170 having a user input interface. have.

In a preferred embodiment, the feature point recognition unit 113 may extract at least one of at least a portion of a headlight of the front vehicle 1, at least a portion of an outline, and at least a portion of a tire as a feature point.

 The speed change detector 114 periodically calculates the distance between the vehicle and the vehicle 1 in the front image, and compares the distance change according to the change in the distance between the feature points and the relative distance with the vehicle 1 in front. To calculate.

According to a preferred embodiment, the speed change detecting unit 114 may provide a vehicle speed and a relative speed between the front vehicle 1 by dividing it into an acceleration state, a deceleration state, and a maintenance state.

For example, the speed change detector 114 periodically calculates a vehicle speed and a relative speed between the front vehicle 1 and the front vehicle 1 according to the absolute speed change of the front vehicle 1. The acceleration or deceleration can be calculated.

In this case, the speed change detector 114 may obtain one or more of a relative speed and a relative distance by obtaining a vehicle speed of the vehicle through a GPS unit 150 or a vehicle connection interface.

As another preferred embodiment of obtaining the own vehicle speed, the vehicle speed may be obtained through the sensing unit 131 or through the vehicle interface unit 106 (vehicle connection interface) connecting with the vehicle. The methods vary.

On the other hand, the front vehicle collision warning device 100 according to another embodiment of the present invention is the front vehicle detection unit 112 for detecting the front vehicle 1 in the front image photographed through the camera, the detected front vehicle (1) At the trailing edge, the feature point recognizing unit 113 extracts a plurality of feature points representing both ends of the front vehicle 1, and calculates a change value between the plurality of feature points over time of travel. The vehicle width change calculation unit periodically calculates a first distance between the vehicle and the vehicle 1 from the front image, compares the interval change value according to the first distance change, and corresponds to a preset second distance change. The speed change detection coefficient adjusted by the learner 118 and the learner 118 to adjust the speed change detection coefficient for the front vehicle 1 by reflecting the difference between the gap change value and the distance change value according to the first distance change. To reflect the arbitrary front car The speed change detection unit 114 calculates a distance change value between a plurality of feature points in (1) and calculates a relative speed or a relative distance between the vehicle and the front vehicle 1 based on the calculated gap change value. It includes.

The front vehicle collision warning device 100 according to another embodiment of the present disclosure adjusts the speed change detection coefficient reflecting the difference in the interval change value between the feature points according to the distance change with the front vehicle 1 through learning, It is a configuration that calculates the relative speed or relative distance to any front vehicle 1 by reflecting the speed change detection coefficient. It is possible to provide exactly the collision warning with the front vehicle 1.

Meanwhile, the front vehicle collision warning device 100 according to an embodiment of the present invention includes an augmented reality unit 111, a front vehicle detector 112, a feature point recognizer 113, a vehicle width change calculator, and a speed change detector 114. ), The collision warning unit 116, the warning information providing unit 117, the learning unit 118 may be configured as a front vehicle collision warning unit 110.

In addition, the front vehicle collision warning device 100 according to an embodiment of the present invention communicates with an external device or an external server by wire or wireless, for example, weather information, traffic conditions, traffic jams, accident information, road information, other It may further include a communication unit 105 for transmitting and receiving a variety of information utilized in the front vehicle collision warning, such as vehicle speed information.

In addition, the front vehicle collision warning device 100 according to an embodiment of the present invention is the communication unit 105, the vehicle interface unit 106, the front vehicle collision warning unit 110, the power supply unit 130, the sensing unit 131, The controller 120 may further include a control unit 120 that controls the storage unit, the output unit 140, the GPS unit 150, the camera unit 155, the recording unit 160, and the user input unit 170.

 The learning unit 118 compares the basic interval change value corresponding to the preset second distance change with the interval change value according to the first distance change, and learns at least one of a classification according to the size of the vehicle or a detailed type of the vehicle. For example, it is preferable to adjust the speed change detection coefficients which are divided according to the classification or type of the front vehicle 1 by learning.

In addition, by classifying one or more of the classification according to the size of the front vehicle 1 or the detailed type of the front vehicle 1 through the learning, learning the speed change detection coefficients respectively divided according to the classification or type of the front vehicle 1 It is effective to provide a more precise collision warning according to the size of the front vehicle 1 or the detailed type of the front vehicle 1 to adjust by.

 The speed change detector 114 displays a cruise function for maintaining a constant one or more of the relative speed and the relative distance with the front vehicle 1 based on one or more of the calculated relative speed and relative distance. Alternatively, the vehicle may be provided through a vehicle connection interface.

In addition, by providing a cruise function through the screen or the vehicle connection interface that maintains one or more of the relative speed and the relative distance with the front vehicle 1 on the basis of the calculated relative speed or relative distance, a separate cruise is provided. Even if a vehicle is not equipped with a function, the cruise function can be easily implemented to further increase the convenience of driving a vehicle.

3 is an exemplary view of feature point extraction according to an embodiment of the present invention.

Referring to FIG. 3, the front vehicle collision warning device 100 according to an exemplary embodiment of the present disclosure extracts the outline 3 of the vehicle, identifies the vehicle 4 at night through the rear headlights, or both bumpers or tire outlines. By extracting the feature points (5,6) can be specified.

According to a preferred embodiment, the front vehicle collision warning device 100 according to an embodiment of the present invention is a front image captured by the camera through various image analysis methods such as an outline extraction technique or a histogram analysis technique based on brightness differences. After extracting the front vehicle 1, after extracting the candidate point of the feature point from the image of the front vehicle 1, it is possible to designate the two end points representing the vehicle width of the front vehicle 1 of the candidate points as the feature point.

As a preferred embodiment, referring to FIG. 3A, the front vehicle collision warning device 100 according to an embodiment of the present invention extracts the outline 3 of the front vehicle 1 from the front image to be photographed, and then displays the vehicle width among the outlines. Points 5 and 6 representing both ends of the vehicle may be recognized and designated as feature points for measuring the relative speed of the front vehicle 1.

Or, as another preferred embodiment, referring to Figure 3b, the front vehicle collision warning device 100 according to an embodiment of the present invention is not easy to extract the outline of the front vehicle 1, such as at night in the front image to be photographed In this case, after recognizing the difference in brightness according to the light of the headlight, the relative speed of the vehicle 1 in front of the point (5,6) that can represent both ends of the vehicle among the boundary (4) of the area recognized as the difference in brightness The feature point can be recognized and specified as a measurement point. In addition, the feature point can be designated by various image analysis methods.

4 is an exemplary view of measuring a change in gap between feature points according to an exemplary embodiment of the present invention.

Referring to FIG. 4, the front vehicle collision warning device 100 according to an exemplary embodiment of the present disclosure may move forward according to the distance change w1 and w2 between the feature points based on the plurality of feature points of the front vehicle 1 recognized in FIG. 3. The traveling speed of the vehicle 1 or the distance to the front vehicle 1 can be calculated.

As a preferred embodiment, as shown in FIG. 4, the front vehicle collision warning device 100 according to the present invention analyzes a feature point of the front vehicle 1 in real time on the basis of augmented reality in a front image obtained and performs the front vehicle ( The speed and distance of the vehicle in front of the vehicle are measured by recognizing the difference that the width of the object recognized as 1) increases or decreases.

More preferably, the relationship between the distance change between the feature points of the front vehicle 1 and the distance between the front vehicle 1 is learned, and the relative distance or the relative speed with the front vehicle 1 based on the learned information. The cruise control function may be provided to keep the constant.

The front vehicle collision warning device 100 according to the embodiment of the present invention has a large distance w1 between the feature points of the front vehicle 1 as shown in FIG. 4A, as shown in FIG. 4B. When the distance w2 between the feature points is narrowed, the relative speed or the absolute speed of the front vehicle 1 can be calculated in correspondence to the interval change value between the feature points compared to the absolute speed of the own vehicle.

Furthermore, the front vehicle collision warning device 100 according to an embodiment of the present invention is measured through augmented reality when the distance to the front vehicle 1 is determined through learning or a predetermined image analysis method at a predetermined time point. The change in the relative distance or the absolute distance (d1-> d2) between the own vehicle 50 and the front vehicle 1 may be calculated by changing the interval between the feature points of the front vehicle 1 as time passes. .

5 is an exemplary diagram of providing a relative speed and a relative distance with the front vehicle 1 according to an embodiment of the present invention.

The front vehicle collision warning device 100 according to an exemplary embodiment of the present disclosure may set, in advance, a feature point interval according to a safety distance between front and rear vehicles based on a vehicle width of each vehicle type.

For example, in order to measure the traveling speed of the vehicle in front of the vehicle, the speed of the front vehicle 1 according to the change value 20 of the interval of the feature points 5 and 6 is added to the vehicle speed measured by the GPS. The traveling speed 25 of the front vehicle 1 can be calculated.

As a preferred embodiment, referring to FIG. 5A, when the front vehicle collision warning device 100 according to an embodiment of the present invention is enlarged 20 through a feature point comparison analysis by time, the preset feature point interval of the front vehicle 1 is enlarged 20. In addition, the vehicle 1 may determine that the vehicle 1 is in a relatively decelerated state relative to the vehicle speed 40, and calculate and display the traveling speed of the vehicle 1 in the manner described above on the screen. The risk of collision may be warned by measuring the instantaneous reduction distance according to the present invention.

As another preferred embodiment, referring to FIG. 5B, the front vehicle collision warning device 100 according to an embodiment of the present invention reduces the predetermined feature point interval of the front vehicle 1 through a feature point comparison analysis for each time 20. When the front vehicle 1 is accelerated relative to its vehicle speed 40, the driving speed 25 of the front vehicle 1 may be calculated and displayed on the screen. The collision risk warning 10 may be omitted by determining that there is no risk of collision by measuring the instantaneous reduction distance according to the distance between vehicles.

As another preferred embodiment, referring to FIG. 5C, when the front vehicle collision warning device 100 according to the embodiment of the present invention has a predetermined feature point interval of the front vehicle 1 through a time-specific feature point comparison analysis, It is possible to confirm that the front vehicle 1 is at a constant speed relative to its vehicle speed 40, and calculate and display the traveling speed 20 of the front vehicle 1 in the above-described manner, and display the distance between the vehicles. The collision risk warning 10 may be provided or omitted by determining whether there is a collision risk by measuring the instantaneous reduction distance according to the interval.

Referring to FIG. 5A, in a front vehicle 1 in a front image displayed on an augmented reality screen, for example, the feature point interval of the current front vehicle 1 is 1 second before the present embodiment. When it is reduced in comparison with the feature point interval of the vehicle 1, the relative speed with the front vehicle 1 according to the feature point interval change value may be calculated according to the change in the interval in response to the relative speed reduction between the vehicles.

In the example of FIG. 5A, if it is calculated that the relative acceleration of the front vehicle 1 corresponding to the feature point spacing reduction is reduced from 64 km / h to 60 km / h at 4 km / h per second, the relative distance between the vehicles is reduced to 1.11 m per second. Can be provided.

In this case, preferably, the front vehicle collision warning device 100 according to an embodiment of the present invention may recognize a vehicle speed by interlocking a GPS or a vehicle instrument panel, and the driving speed of the front vehicle 1 or the front vehicle. The distance from (1) can be calculated by applying the coefficient corrected by learning in advance or by forward feature point correction.

In addition, the front vehicle collision warning device 100 according to an embodiment of the present invention reflects a predetermined weather conditions 45, such as snow, rain, fog (safety road: 1,5 times the usual, ice road or Attention: 3 times as usual) and the collision warning can be provided if the safety distance is within the safety distance according to the calculated distance to the front vehicle 1 (35).

In addition, as shown in Figure 5a, 5b, 5c, the front vehicle collision warning device 100 according to an embodiment of the present invention is generated by the collision warning unit 116 in the warning information providing unit 117 The collision warning information may be provided by changing a traffic light, a feature point variation measurer, a distance between vehicles according to a safety distance, etc., into a preset screen display format or a sound providing format on an augmented reality screen on which the front vehicle 1 is displayed. .

The front vehicle collision warning device 100 according to an embodiment of the present invention is a technology based on the degree of change in the speed of the front driving vehicle according to the change of the feature point of augmented reality. It can be used appropriately for distance determination, and it is very easy to present the criterion for avoiding a collision situation because it is possible to quickly determine the relative deceleration and acceleration of the front vehicle 1 by calculating the momentary feature point interval change even in a slow section. Done.

6 is a flowchart of a front vehicle collision warning method according to an embodiment of the present invention.

6, the front vehicle collision warning method according to an embodiment of the present invention, the front vehicle collision warning method according to another embodiment of the present invention through the front vehicle detecting unit 112, the front to shoot through the camera Step (S10, S15) of detecting the front vehicle 1 in the image, both ends of the front vehicle 1 at the rear of the detected front vehicle 1 through the feature point recognition unit 113 Extracting a plurality of feature points representing a) (S20), a step of calculating a distance change value between the plurality of feature points according to the passage of time through the vehicle width change calculator (S25), through the learning unit 118, Computing the distance between the vehicle and the front vehicle 1 periodically from the image, and comparing the interval change value according to the distance change (S30), through the learning unit 118, corresponding to the preset distance change The distance change compared with the default gap change value Adjusting the speed change detection coefficient for the front vehicle 1 by reflecting the difference in the interval change value according to the step (S35) and the speed change detection unit 114, the front vehicle 1 detection step to the interval change comparison step Iteratively calculate the interval change value between a plurality of feature points for any front vehicle 1 by reflecting the adjusted speed change detection coefficient, and based on the calculated interval change value, the vehicle and the front vehicle 1 and Calculating relative speeds or relative distances (S45, S50, S60).

 The speed change detection coefficient adjusting step may be performed by comparing at least one of a classification according to the size of the vehicle or a detailed type of the vehicle by comparing the basic interval change value corresponding to the preset second distance change with the interval change value according to the first distance change. In addition, it is preferable to adjust (S40) the speed change detection coefficients, which are divided according to the classification or type of the front vehicle 1, by learning.

The front vehicle collision warning method according to another embodiment of the present invention, based on one or more of the relative speed, relative distance calculated by the speed change detection unit 114, relative speed with the front vehicle 1, relative distance The method may further include providing a cruise function through a screen or a vehicle connection interface that maintains one or more of the same.

7 is a flowchart illustrating a front vehicle collision warning method according to another embodiment of the present invention.

Referring to FIG. 7, the front vehicle collision warning method according to another embodiment of the present disclosure detects the front vehicle 1 from a front image photographed through a camera through the front vehicle detection unit 112 (S61). Extracting a plurality of feature points representing both ends of the front vehicle 1 from the detected rear end of the front vehicle 1 through the feature point recognition unit 113 (step S62). Computing the interval change value between the plurality of feature points through the change operation unit (S63) and the speed change detection unit 114, the vehicle and the front of the vehicle based on the change value of the plurality of feature points through the speed change unit 114 Computing the relative speed or the relative distance with the vehicle 1 (S64).

The front vehicle collision warning method according to an embodiment of the present disclosure further includes a step S65 of warning a collision with the front vehicle 1 based on the calculated relative speed or relative distance through the collision warning unit 116. Include.

 The feature point extracting step S62 may extract at least one of at least a portion of a headlight of the front vehicle 1, at least a portion of an outline, and at least a portion of a tire as a feature point.

The vehicle width change calculation step (S64) periodically calculates the distance between the vehicle and the vehicle 1 in front of the vehicle in the front image, and compares the distance change according to the change in the distance between the feature points and the relative distance with the vehicle in front of the vehicle 1. Can be calculated.

In the above described and illustrated with respect to preferred embodiments according to the present invention. However, the present invention is not limited to the above-described embodiments, and various modifications can be made by those skilled in the art without departing from the scope of the present invention, which is appended in the claims. .

The present invention extracts a plurality of feature points representing both ends of a front vehicle from a rear image of a front vehicle detected by a front image photographed by a camera of a navigation device or a smart terminal, thereby determining a change value between feature points. By calculating one or more of the relative speed and relative distance with the vehicle in front of you, you can easily provide a collision warning with the vehicle in front of you at low cost without installing a separate collision warning device, and in difficult driving conditions such as snow, rain and bad weather In addition, the accuracy of the measurement of relative speed and the collision warning is very high, so it can be widely used in the field of front vehicle collision warning system (FCWS), traffic information provision, navigation device field, smart terminal field, etc.

1: vehicle ahead 3: outline
4: headlight 5,6: feature points
50: own vehicle 100: front vehicle collision warning device
105: communication unit 106: vehicle interface unit
110: front vehicle collision warning unit 111: augmented reality
112: front vehicle detection unit 113: feature point recognition unit
114: speed change detection unit 116: collision warning unit
117: warning information provider 118: learning unit
120: control unit 130: power unit
131: sensing unit 132: storage unit
140: output unit 141: display unit
142: sound unit 150: GPS unit
155: camera unit 160: recording unit
170: user input unit

Claims (11)

A front vehicle detecting unit detecting a front vehicle in a front image photographed through a camera;
A feature point recognizing unit configured to extract a plurality of feature points representing both ends of the front vehicle from the sensed tail of the front vehicle;
A vehicle width change calculator configured to calculate an interval change value between the plurality of feature points over time of travel; and
A speed change detector configured to calculate a relative speed or a relative distance between the vehicle and the front vehicle based on the interval change value between the plurality of feature points; And
The first distance between the vehicle and the vehicle in front of the vehicle is periodically calculated in the front image so that the relative speed or the distance between the vehicle in front of the vehicle is calculated based on the change in the position and the direction in which the vehicle image is captured. Compare the interval change value according to the distance change, and adjust the speed change detection coefficient for the front vehicle by reflecting the difference between the basic interval change value corresponding to the preset second distance change and the interval change value according to the first distance change. A front vehicle collision warning device comprising a learning unit. The method of claim 1,
And a collision warning unit warning a collision with the front vehicle based on the relative speed or the relative distance calculated. The method of claim 2, wherein the collision warning unit
A front vehicle collision warning device for providing a change in the interval between the feature points on the screen or sound. The method of claim 1,
And augmented reality unit displaying one or more of the detected feature points, relative speeds, and relative distances of the front vehicle as augmented reality on a screen. The method of claim 1, wherein the feature point recognizer
And a front vehicle collision warning device for extracting one or more of a portion of a tail light, a portion of an outline, and a portion of a tire as the feature point. The method of claim 1, wherein the speed change detection unit
And calculating a distance between the vehicle and the vehicle ahead of the vehicle from the front image periodically, and calculating the relative distance between the vehicle and the vehicle by comparing the distance change according to the change of the distance between the feature points. The method of claim 1, wherein the front vehicle collision warning device
A front vehicle collision warning device configured in one of a vehicle black box device, a navigation device, a front surveillance camera device, a smart phone, a smart terminal, and a tablet PC. The method of claim 1,
The distance change value between the plurality of feature points is calculated for any front vehicle by reflecting the speed change detection coefficient adjusted by the learning unit, and the vehicle between the vehicle and the front vehicle based on the calculated space change value. And a speed change detector configured to calculate a relative speed or a relative distance. The method of claim 1, wherein the learning unit
Compares the basic interval change value corresponding to the preset second distance change with the interval change value according to the first distance change to learn one or more of classification according to the size of the vehicle or detailed types of the vehicle, and classify the front vehicle. Or the vehicle collision warning device for adjusting the speed change detection coefficients, which are divided according to a type, by the learning. 10. The method of claim 9, wherein the speed change detection unit
A forward vehicle collision providing a cruise function through a screen or a vehicle connection interface based on one or more of the relative speed and relative distance to be calculated, maintaining a constant one or more of the relative speed and the relative distance with the front vehicle. Warning device. Detecting, by the front vehicle detecting unit, the front vehicle in the front image photographed through the camera;
Extracting a plurality of feature points representing both ends of the front vehicle from the detected rear of the front vehicle through a feature point recognizing unit;
Calculating, by a vehicle width change calculator, an interval change value between the plurality of feature points over time;
Calculating a relative speed or a relative distance between the vehicle and the vehicle ahead of the vehicle based on a change value of the distance between the plurality of feature points through a speed change detector; And
The learner periodically calculates a first distance between the vehicle and the vehicle in front of the vehicle so that the relative speed or the distance between the vehicle in front of the vehicle may be calculated based on a change in the position and the direction in which the vehicle image is captured. Comparing the interval change values according to the first distance change and reflecting the difference between the basic interval change value corresponding to the preset second distance change and the interval change value according to the first distance change, and then changing the speed of the front vehicle. And adjusting the detection coefficient.

Images