KR20150076884A

KR20150076884A - Device and method for alarming start of preceding-vehicle to rewarding-vehicle

Info

Publication number: KR20150076884A
Application number: KR1020130165563A
Authority: KR
Inventors: 최주희
Original assignee: 최주희
Priority date: 2013-12-27
Filing date: 2013-12-27
Publication date: 2015-07-07

Abstract

An alarm device for notifying a departure fact of a forward vehicle of the present invention comprises an image collection module installed in a rear vehicle for collecting image data of a preceding vehicle, a taillight pattern detected by processing the collected image data, And an event occurrence detection module for detecting whether the event is generated by detecting the absence of the detected taillight. According to the present invention as described above, it is very convenient for the elderly who are somewhat lacking in instantaneous power.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to an alarm device,

The present invention relates to an alarm apparatus and method for informing a driver of a rear vehicle of a departure of a preceding vehicle, and more particularly, And the low-luminance lighting state is changed. In the stopped rear vehicle, the event occurrence is automatically recognized through the image-based or sensor-based pattern detection and the distance value calculation to inform the driver visually or audibly, The present invention relates to an alarm device and a method for notifying a departure state of a front vehicle that does not need to pay attention to a starting state of a vehicle.

In general, the vehicle is provided with a direction indicator for informing the progress direction of the vehicle by a blinking operation, such as a brake indicating that the braking device is operated while the vehicle is running and a brake for informing the outside, And the emergency stop light flashing, the vehicle width indicating the width of the vehicle, and the like.

Among them, a brake or the like is turned on by the brake switch when the driver of the front vehicle depresses the brake pedal, and is turned off by the brake switch when the brake pedal is released from the brake pedal. So that it can be operated safely.

In particular, the driver of the rear-stopping vehicle can see the fact that the lighting state of the brake or the like of the front vehicle is changed to the light-off state during the daytime or the nighttime state is changed from the high-luminance lighting state to the low-luminance lighting state.

However, when the driver is driving the vehicle for a long time, he / she can not recognize the brake of the preceding vehicle due to the fatigue. Therefore, the driver often does not know the departure and hears the horn sound of the trailing vehicle. Especially, when the signal change period is short at the intersection, it affects the vehicle flow.

Furthermore, elderly people with poor concentration have difficulty in recognizing the departure of a forward vehicle, which may interfere with the overall flow of traffic compared to young people.

On the other hand, if the car is stalled in a congested state or in a dwelling area, it is inconvenient to check the departure time of the preceding vehicle from time to time.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and it is an object of the present invention to provide an alarm device for notifying departure facts of a forward vehicle, Method.

Another object of the present invention is to provide an alarm device and method for notifying a driver of a front vehicle capable of detecting a pattern such as a camera image-based brake so as to inform the driver of whether the brake lamp or the like is turned on or off using a single color camera .

Yet another object of the present invention is to provide an alarm device and method for indicating the departure fact of a preceding vehicle capable of calculating a camera image based distance value so as to determine a distance between a front vehicle and a rear vehicle using one color camera .

It is another object of the present invention to provide an alarm device and method for storing shape and size information of brakes and the like for each vehicle type in a database to detect a departure fact of a front vehicle in which pattern detection and distance value calculation are easy.

Yet another object of the present invention is to provide an alarm device and method for informing the departure fact of a preceding vehicle that pattern detection and calculation of a distance value are easy using a knowledge-based algorithm so that brake pattern detection and distance value calculation can be performed even when a database is not provided .

It is still another object of the present invention to provide an alarm device and method for notifying a departure fact of a vehicle ahead of the event when an event occurs, such as a display, a speaker, or a lamp.

According to an aspect of the present invention, there is provided an alarm device for notifying a departure of a preceding vehicle, comprising: an image collection module installed in a rear vehicle for collecting image data of a preceding vehicle; An image processing module for detecting a taillight pattern by processing the image data and determining whether a distance value between the rear vehicle and the front vehicle is within a predetermined range, And an event occurrence detection module for determining whether or not the event has occurred.

According to another aspect of the present invention, an alarm method for notifying a departure fact of a forward vehicle of the present invention includes a step of collecting image data of a preceding vehicle from a point of time when a rear vehicle is stopped, extracting only a red color pattern A step of extracting a specific shape pattern by processing the boundary of the red color pattern, a step of detecting a tail light pattern by comparing the specific shape pattern with previously stored reference shape information, And determining the departure fact of the preceding vehicle.

As described above, according to the configuration of the present invention, the following effects can be expected.

First, by informing the driver of the departure of the front vehicle through a speaker or a lamp, it is expected to provide convenience to the elderly who are weak in concentration and have a low power.

Second, since the pattern detection and the distance value calculation can be performed using one color camera, the structure is not complicated, and it is economical to install it independently of a vehicle black box or navigation.

Third, since it can be installed integrally with an existing vehicle black box or navigation, it is expected to have a reflective effect that can further enhance the utility value of a black box or navigation.

Fourth, since the camera image-based pattern can be detected by using the library in which the shape information of the vehicle type is stored, the detection reliability of the brake and the like is very high, and even if it is not stored, Is expected to be effective.

Fifth, since the camera image-based distance value can be calculated using the library in which the size information of the vehicle type and the like is stored, the detection reliability of the brake and the like is very high, and since the sensor based distance value can be calculated even if it is not stored, Is expected to be effective.

1 is an example of the installation of an alarm device for notifying the departure fact of the preceding vehicle according to the present invention.
2 is a block diagram showing a configuration of an alarm device for notifying the departure fact of a preceding vehicle according to the present invention;
3 is a flowchart showing an alarm method for notifying the departure fact of a preceding vehicle according to the present invention;

Brief Description of the Drawings The advantages and features of the present invention, and how to achieve them, will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. The dimensions and relative sizes of layers and regions in the figures may be exaggerated for clarity of illustration. Like reference numerals refer to like elements throughout the specification.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of an alarm device for notifying a departure of a front vehicle according to the present invention having the above-described configuration will be described in detail with reference to the accompanying drawings.

Referring to FIG. 1, the alarm device 10 for notifying the departure of the front vehicle of the present invention includes an image collection module 100 installed in the rear vehicle 20 and collecting image data of the front vehicle 30, The image processing module processes the collected image data to detect the taillight pattern 32 and determines whether the distance value between the rear vehicle 20 and the front vehicle 30 is within a predetermined range An event occurrence detection module 300 for detecting the departure of the front vehicle 30 by detecting the absence of the detected taillight and an event occurrence detection module 300 for detecting the departure of the front vehicle 30, 20 to the driver of the vehicle.

The alarm device 10 for notifying the start of the full-length car according to the present invention can be integrated with the video recording device of the vehicle black box or the navigation device for the vehicle, or can be installed independently.

In addition, the rear vehicle 20 acquires the image data of the front vehicle 30 in the stopped state, and the obtained image data mainly includes the brake lights or the tail lights 32 that are turned on in the front vehicle 30 in the stop state It is targeted.

Further, the light-off alarm device 10 of the taillight according to the present invention will be described as an example of a daylight. In the nighttime, since the taillight state is both in the running state and in the stopped state, the lighting state is turned on differently. In other words, since the two brakes operate at night and one brake operates at the start, A case in which the tail light 32 of the vehicle being stopped at the start is extinguished at the time of starting from the lighting state will be described. Therefore, it is interpreted that, in the specification, off-state includes a case of changing from a high-luminance on state to a low-luminance on state as well as a case of changing from a lighted state to an unlit state.

The image collection module 100 is configured to receive an image of the tail light 32 of the front vehicle 30 at a predetermined position of the rear vehicle 20, (110), a color filter array (120) composed of RGB filters, and an image sensor (130) for generating image data for converting the received optical signals into electrical image signals required for image processing have.

The image sensor 130 may include a CCD (Charge Coupled Device) image sensor, a CMOS (Complementary Metal Oxide Semiconductor) image sensor, and the like. In addition, the color camera module may include an image, a web, or various digital cameras in addition to a CCD and a CMOS.

Meanwhile, the lens 110 of the present invention may be coated in a green color. By performing hardware green coating on the upper color camera lens 110, it is possible to more accurately detect the complementary red color. For example, there is a certain limit in detecting the red color through image processing by software, and coating of the lens 110 itself with the green color is more effective in judging the taillight especially illuminated in the daytime . The complementary colors red and green are yellow when mixed and can be distinguished most clearly during the daytime.

The image processing module 200 includes a color pattern detecting unit 210 for extracting only specific color image data from the image data of the multicolor to detect a red color pattern, And a distance value calculating unit 230 for estimating a distance between the front vehicle 30 and the rear vehicle 20. [

The color pattern detecting unit 210 extracts only red color from the raw image data and detects a red color pattern. For example, in a vehicle, color distribution map information of a sample in which a taillight area and a bumper other vehicle body area are separated is calculated, and color values (for example, RGB pixel values) of each area are calculated to obtain sample color distribution map information It is possible to easily detect only the red color pattern.

After the color pattern is detected, noise is removed using a local average filter in a specific color pattern so as to emphasize the pattern boundary of the specific shape, and the brightness of the specific color pattern is obtained by using spectrum equalization I can correct it. For example, a correction may be performed to remove the shadow effect, or a certain color pattern may be changed to a gray pattern, and then the gray pattern contrast value may be smoothed to have a constant contrast value distribution.

The shape pattern detector 220 can detect only a specific pattern, which can be regarded as the tail light 32, in the red color pattern. For example, using a knowledge based algorithm, the taillight 32 must be line-symmetric and depart from the fact that the shape and size are the same except for the directionality, Only the pattern of the shape can be detected. For example, a pair of resembling patterns spaced apart by a predetermined distance can be seen as a tail light 32.

In addition, in recent years, the tail lights 32 tend not only to be installed on both sides of the vehicle rear side but also to the rear center upper end. Therefore, when there are more similar patterns in the center of a specific pattern to be symmetrical, It is based on the fact that the similar-shaped pattern can be seen by the taillight 32.

Since the tail light 32 can be regarded as being progressed at the same time without selectively being turned on or off, the tail light 32 can be seen when the pair of specific patterns are turned on or off at the same time.

According to another embodiment of the present invention, the library 240 may further include taillight shape information and size information for each vehicle type.

The image processing module 200 can finally determine the tail light 32 by comparing it with the tail light 32 pattern provided by the color pattern detector 210 based on the previously stored library 240. [

The distance value calculating unit 230 can calculate the distance to the forward vehicle 30 by storing the size information as well as the shape of the various taillights 32 in the library 240 for each type of vehicle. That is, the distance value between the rear vehicle 20 and the front vehicle 30 can be estimated by comparing the size of the taillight 32 of the collected image data with the size of the preliminarily stored taillight 32 as a reference.

Alternatively, the focal length of the camera lens 110 may be used to calculate the distance from the image sensor 130 to the focal point of the vehicle, thereby measuring the distance to the vehicle ahead.

In other cases, when the color pattern is recognized on the image sensor 130 according to the distance, a phenomenon in which the red color appears gradually as the upper distance value becomes smaller within a predetermined distance range, The distance to the front vehicle 30 can be measured by calculating how much the vehicle changes in white color.

If the rear vehicle 20 and the front vehicle 30 are not located on the same straight line, the angle of inclination of the vehicle can be calculated and the direction value of the vehicle can be calculated. can do. In particular, the direction value can be used to recognize a specific pattern by reflecting a direction value when a pair of specific patterns are not mutually symmetric.

The present invention may further include a distance sensor 250 and an illuminance sensor 260. [

The distance value calculation unit 230 can calculate the image-based distance value. However, if the image-based distance value can not be calculated, it is necessary to calculate the sensor-based distance value either side by side or in parallel.

The distance sensor 250 may include a laser or an ultrasonic sensor that emits ultrasonic waves to the front vehicle 30. The ultrasonic sensor can calculate the distance by measuring the signal that the divergent laser or the ultrasonic wave is reflected to the forward vehicle 30 and returns. Although the sensor-based distance value may be more accurate than the image-based distance value, a separate sensor must be provided and the calculation may be complicated.

The illuminance sensor 260 can determine the day and the night. In addition, the headlamp of the rear vehicle 20 may include an illuminance that is reflected back to the front vehicle 30. For example, if the rear vehicle 20 does not turn on the headlamp at night, it is easy to identify the tail light 32 of the front vehicle 30. However, if the vehicle is turning on the headlamp, It is difficult to identify the taillight 32, or extraction of red color may become difficult.

The event generation alarm module 400 may include any method that can inform the driver of the departure of the front vehicle 30. For example, it may show the occurrence of an event on the display panel 410 or inform the user through the speaker 420 or the lamp 430. In addition, the event can be notified through vibration.

According to another embodiment of the present invention, an alarm is provided as an example, but it is not excluded to directly control the machine of the rear vehicle 20 directly.

Hereinafter, an alarm method for notifying the departure of the preceding vehicle according to the present invention will be described in detail with reference to the drawings.

3, the image data of the front vehicle 30 is collected from the point of time when the rear vehicle 20 is stopped. (S10) While the vehicle is stopped, the rear vehicle 20 is stopped using the image collection module 100 The image data of the tail light 32 of the preceding vehicle 30 is obtained.

Only the red color pattern is extracted from the collected image data. (S20) The color pattern detecting unit 210 extracts only a specific color pattern from the image data and detects a red color pattern.

(S30) The shape pattern detection unit 220 extracts only a specific shape pattern from the red color pattern.

The taillight pattern is detected by comparing the specific shape pattern and the previously stored reference shape information. (S40, S50) It is determined whether or not a specific shape pattern is stored in the library 240. If there is already stored information, Thereby detecting the taillight 32 pattern.

The distance value calculating unit 230 calculates the distance value between the rear vehicle 20 and the front vehicle 30 by comparing the size of the specific shape pattern with the size information of the reference shape stored previously. It is determined whether or not the distance value can be calculated based on the size information of the taillight 32 previously stored.

If the specific shape pattern is not stored in the library 240 so that it can not be compared with the reference shape information or is stored but is hard to be calculated, a tail light pattern is detected using a knowledge based algorithm. (S70) It is possible to determine the pattern as a taillight 32 pattern in the case of a pair of resembling patterns whose specific pattern is spaced apart by a predetermined distance.

(S80). If the distance value can not be calculated by the distance value calculation unit 230, the distance value is calculated using the distance sensor 250 can do.

(S90 and S100). [0051] The event is generated through the display panel 410, the speaker 420, or the lamp 430 by detecting the absence of the detected taillight 32 and determining the departure of the front vehicle 30. You can inform the driver of the fact or control the vehicle's machine directly.

As described above, the present invention detects the occurrence of an event such as a brake of a forward vehicle through image-based pattern detection and distance value calculation, and informs the driver of the rear vehicle visually or audibly . Many other modifications will be possible to those skilled in the art, within the scope of the basic technical idea of the present invention.

10: Alarm device informing the departure of the preceding vehicle
20: rear vehicle 30: forward vehicle
32: Tail light
100: image acquisition module 110: lens
120: color array 130: image sensor
200: image processing module 210: color pattern detector
220: shape pattern detection unit 230: distance value calculation unit
240: Library 250: Distance sensor
260: illuminance sensor 300: event occurrence detection module
400: Event occurrence alarm module 410: Display panel
420: speaker 430: lamp

Claims

An image collection module installed in the rear vehicle for collecting image data of the forward vehicle;
An image processing module processing the collected image data to detect a taillight pattern, and determining whether a distance value between the rear vehicle and the front vehicle is within a predetermined range; And
And an event occurrence detection module for detecting the occurrence of an event by detecting the absence of the detected taillight.

The method according to claim 1,
And an event occurrence alarm module for visually or audibly informing the driver of the occurrence of the event.

The method according to claim 1,
Wherein the image acquisition module comprises:
A lens for receiving an optical signal for an image of the front vehicle;
A color filter array composed of RGB filters; And
And an image sensor for generating the image data by changing the received optical signal into an electrical image signal necessary for image processing,
Characterized in that the lens is coated in a green color.

The method according to claim 1,
The image processing module includes:
A color pattern detector for extracting only a specific color pattern from the image data and detecting a red color pattern; And
And a shape pattern detecting section for extracting only a specific shape pattern from the red color pattern and detecting the tail light pattern.

5. The method of claim 4,
Further comprising a distance value calculating unit for estimating a distance value between the front vehicle and the rear vehicle.

6. The method of claim 5,
Further comprising a pre-stored library of shape information of the taillights for each vehicle type,
Wherein the image processing module compares the taillight pattern information with a taillight pattern provided by the shape pattern detection section based on taillight pattern information of the library.

The method according to claim 6,
Further comprising a pre-stored library of taillight size information for each vehicle type,
Wherein the distance value calculation unit calculates a distance value between the rear vehicle and the front vehicle based on taillight size information of the library.

The method according to claim 6,
Wherein the image processing module detects the tail light shape pattern using a knowledge based algorithm in the red color pattern,
Wherein the knowledge-based algorithm is characterized in that the pair of resembling patterns separated by a predetermined distance are viewed as tail lights.

8. The method of claim 7,
Further comprising an ultrasonic distance sensor for measuring a distance between the front vehicle and the rear vehicle,
Based distance value is calculated by the ultrasonic distance sensor when the image-based distance value is not calculated in the distance value calculating unit.

5. The method of claim 4,
Further comprising an illuminance sensor,
Wherein the light emitted from the headlamp of the rear vehicle is reflected by the front vehicle and is reflected to detect the red color pattern.

Collecting image data of the preceding vehicle from a point of time when the rear vehicle is stopped;
Extracting only a red color pattern from the collected image data;
Processing a boundary of the red color pattern to extract a specific shape pattern;
Comparing the specific shape pattern with previously stored reference shape information to detect a taillight pattern; And
And detecting an off state of the detected taillight to determine a departure of the preceding vehicle.

12. The method of claim 11,
If the specific shape pattern is not stored in the reference shape information, detecting a taillight pattern using a knowledge based algorithm,
Wherein the step of detecting the taillight pattern using the knowledge based algorithm comprises determining whether the specific shape pattern is a pair of resembling patterns that are spaced apart by a predetermined distance .

12. The method of claim 11,
Further comprising the step of calculating a distance value between the rear vehicle and the front vehicle by comparing the size of the specific shape pattern with the size information of a previously stored reference shape,
Further comprising the step of calculating a distance value by a distance sensor when the distance value can not be calculated.

12. The method of claim 11,
The step of extracting only the red color pattern includes:
The color distribution information of the sample in which the tail light area and the vehicle body area are separated is calculated, the color value of each area is calculated, and only the red color pattern is detected using the sample color distribution map information for each area,
The step of extracting the specific shape pattern may include:
A color filter is used to remove noise by using a local average filter in the red color pattern, a red color pattern is changed to a gray pattern, and then a specific color pattern is generated using a spectrum equalization And the brightness of the preceding vehicle is corrected.