KR101665340B1 - Apparatus for sensing forward obstacle - Google Patents

Abstract

One feature of the front obstacle sensing apparatus includes an image acquiring unit that generates original image data of an image in a direction in which the vehicle is traveling, a display unit that is connected to the image acquiring unit, The obstacle candidate image data group is generated by using the original image data, and the obstacle candidate image data group is generated by using the row position of the pixel located at the lowermost and the uppermost pixel and the row position of the pixel located at the leftmost and rightmost pixel, An obstacle candidate region including the candidate video data group is extracted and a distance between the obstacle candidate region located closest to the vehicle and the obstacle candidate region is determined. If the determined distance is less than the set distance, An operation control unit connected to the operation control unit, And a warning section operating in accordance with the warning signal.

Description

[0001] APPARATUS FOR SENSING FORWARD OBSTACLE [0002]

The present invention relates to a front obstacle sensing device.

Today's vehicles are equipped with a variety of additional facilities to provide convenience to the driver. For example, with the development of global positioning system (GPS), a car navigation system has been introduced to provide convenience and information to drivers.

Such a vehicle navigation system provides drivers with convenience by providing current position information, destination location information, and route information to the destination.

In addition, recently, various systems such as a black box or a lane departure warning system (LDW) in a vehicle have been installed.

It is necessary to develop a device that reduces the risk of accidents by inputting information generated from outside the vehicle.

However, it is not easy to detect a vehicle positioned ahead in a state where the external environment continuously changes in the running state. Accordingly, there is a need for a technique for reliably detecting a vehicle located ahead.

Korean Patent Laid-Open Publication No. 10-2011-0117990 (Publication Date: October 28, 2011, entitled "Front Vehicle Detection Apparatus and Method)

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and an object of the present invention is to prevent collision of a vehicle by predicting a collision state between a vehicle and an obstacle.

A front obstacle sensing apparatus according to one aspect of the present invention includes an image acquiring unit that generates original image data of an image in a direction in which a vehicle is traveling, an image acquisition unit connected to the image acquiring unit, And generating an image data group based on the row positions of the pixels positioned at the leftmost and rightmost positions of the pixels located at the lowest and uppermost positions in the obstacle candidate image data group, An operation control section for extracting a candidate region and determining a distance between the obstacle candidate region located closest to the vehicle and the warning signal if the determined distance is less than a set distance, And a warning unit connected to the control unit and operating in accordance with the warning signal .

The operation control unit extracts a part of the original image data to generate processing image data, generates binary image data by binarizing the processing image data, labels the binarized image data to generate the obstacle candidate image data group Can be obtained.

Wherein the operation control unit converts the binarization value of the corresponding pixel among '0' and '1' to a value for displaying white when the tone value of each pixel of the processing video data is equal to or greater than the set tone value when the running time is the week, Converts the binarization value of the corresponding pixel among '0' and '1' to a value for displaying black when the tone value of each pixel of the binarized image data is less than the set tone value, and converts the processing image data into binarized image data And converts the binarization value of the corresponding pixel among '0' and '1' to a value for displaying black when the tone value of each pixel of the binarized image data is equal to or greater than the set tone value when the running time is night, When the tone value of each pixel of the data is less than the set tone value, the binarization value of the corresponding pixel among '0' and '1' is converted into a value for displaying white, It is recommended to convert the image data into binary image data.

Wherein the operation control unit extracts video data for day-and-night determination from the video data for processing, and when the average tone value of the video data for day / night determination is less than the setting value for day / night determination, When the average gradation value of the image data for day is equal to or larger than the setting value for day / night determination, the running time of the vehicle can be judged as the week.

It is preferable that the setting value for day / night determination is changed by adding or subtracting the set addition value according to the ratio of the total number of pixels showing black color and the total number of pixels showing white color in the binarized image data.

The operation control unit may set the obstacle candidate region located in the range of the pixel row determined by the rightmost pixel position to the leftmost pixel position of the distance system of the distance meter as the final obstacle region as the obstacle candidate left- And determine a distance between the vehicle and the final obstacle area, which is the closest to the vehicle, of the final obstacle area as the separation distance.

According to this feature, since the vehicle driver is informed of the collision with the obstacle located in front of the vehicle during driving, the risk of collision of the vehicle is reduced, thereby greatly reducing the risk of loss of life as well as economic loss.

1 is a schematic block diagram of a front obstacle sensing apparatus according to an embodiment of the present invention.
2A and 2B are operational flowcharts of a front obstacle sensing apparatus according to an embodiment of the present invention.
3 is a diagram illustrating a method of extracting image data for processing from original image data according to an embodiment of the present invention.
4 is a diagram illustrating a method of extracting image data for day / night determination from original image data according to an embodiment of the present invention.
5 is a diagram illustrating a method of extracting an obstacle candidate image data group and an obstacle candidate region from binarized image data according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating an image in which a distance meter is displayed on an image obtained by the image obtaining unit according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but it should be understood that there may be other elements in between do. On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

A front obstacle sensing apparatus according to an embodiment of the present invention will now be described with reference to the accompanying drawings.

First, the structure of a front obstacle sensing apparatus according to an embodiment of the present invention will be described in detail with reference to FIG.

The front obstacle sensing device according to the present embodiment is for mounting on a vehicle such as a passenger car or a truck and determining whether or not the vehicle collides with another vehicle, and can constitute a part of a black box or a car navigation system.

 The front obstacle sensing apparatus according to an embodiment of the present invention shown in FIG. 1 includes an image acquisition unit 10, an operation control unit 20 connected to the image acquisition unit 10, and an operation control unit 20 connected to the operation control unit 20 A storage unit 30, a display unit 40 connected to the operation control unit 20, and a warning unit 50.

The image acquiring unit 10 acquires original image data for an image in a direction in which the vehicle travels (that is, a forward image) mounted on the vehicle. The image acquiring unit 10 acquires original image data It is possible to continuously acquire or acquire a forward image at a predetermined time interval.

The image acquiring unit 10 can acquire an image using a digital camera having an image input sensor such as a CCD (charge coupled device).

Accordingly, when the image signal related to the forward image is acquired by the image input sensor, signal processing of the acquired image signal is performed, and then the image signal in the analog state is converted into digital image data (hereinafter referred to as 'original image data' Quot;) and input to the operation control unit 20.

At this time, the size (i.e., resolution) of the original image data may be 1024 pixels or 768 pixels, respectively.

The operation control unit 20 receives the original image data corresponding to the image through the image acquisition unit 10 and determines the distance to the obstacle located in front of the obstacle. If the distance to the obstacle is less than the set distance, The warning unit 50 is operated to warn the driver of the risk of collision.

2, the operation control unit 20 includes an operation of extracting only a part of the original image data inputted from the image obtaining unit 10, an operation of extracting the image data for processing for extracting only the part of the original image data inputted from the image obtaining unit 10, An operation of binarizing the grayscale value of each pixel of the processing video data to have a value of '1' or '0' using a value (that is, a pixel value) and a set value, an operation of binarizing processed video data (Hereinafter referred to as " obstacle candidate image data group ") corresponding to the obstacle candidates by using the image data of the obstacle candidate image data group, And the row position of the pixel except the leftmost and rightmost positions are used to calculate a plurality of blocks including the corresponding obstacle candidate image data group Extracting a water candidate region, removing an obstacle candidate region that is not located in a lane on which the vehicle is currently traveling among a plurality of obstacle candidate regions to determine a final obstacle region, And determines the position of the obstacle located at the closest distance to the vehicle.

When the position of the obstacle closest to the vehicle is determined, the operation control unit outputs a warning signal to the warning unit 50 to warn the driver of the risk of the vehicle collision if the determined distance is less than the set distance.

The storage unit 30 stores data or signals generated during the operation of the front obstacle sensing apparatus of this example, such as original image data and processing image data, input through the image acquiring unit 10 as a storage medium such as a memory, Predetermined data necessary for the operation such as the set value and the set distance, and the like are stored.

The display unit 40 outputs a corresponding image by a control operation of the operation control unit 20 such as a liquid crystal display (LCD) or an organic light emitting display (OLED).

The warning unit 50 operates according to the warning signal outputted from the operation control unit 20 to warn the driver of the vehicle of the collision situation with the obstacle in the traveling direction of the vehicle.

The alarm unit 50 may include a buzzer for outputting a warning sound or a warning light such as a light emitting diode (LED).

Next, the operation of the front obstacle sensing apparatus according to this embodiment will be described with reference to Figs. 2 to 6. Fig.

First, when power required for operation of the front obstacle sensing apparatus of the present embodiment is supplied and the front obstacle sensing apparatus is operated, the operation of the operation control unit 20 is also started (S10).

Accordingly, the operation control unit 20 reads the original image data from the image obtaining unit 10 and stores the original image data in the storage unit 30 (S11).

Next, the operation control unit 20 extracts a part of the original image data D11 as shown in FIG. 2 and extracts processing image data (S12).

The processing video data is composed of a plurality of pixels PX arranged in a matrix and arranged in a row direction and a column direction.

Accordingly, in order to extract the processing image data from the original image data D11, the operation control unit 20 sets a plurality of pixels sequentially arranged in the row direction for each row to a predetermined number of pixel row units (for example, Unit number), and a plurality of pixels arranged side by side in the row direction for each row are divided into a plurality of unit-row pixel groups (e.g., a plurality of first unit-row pixel groups R1).

Then, the operation control unit 20 extracts a set number of pixels PXpr located in the middle of each first unit row pixel group R1 excluding the pixels located at the edge portions thereof, and stores them as a plurality of spare row direction image data (30). At this time, the position of the spare row direction image data, that is, the position and the tone value of the corresponding pixel PXpr are all stored in the storage unit 30.

In this example, only the outermost pixels, which are located at the left and right edges of each first unit row pixel group R1, are excluded, but the number of pixels that are excluded from the left and right edges may be changed. For example, Since the number of pixels per line is 5 and the number of pixels excluded from the left and right edges is one, only three pixels PXpr located in the middle portion are preliminary row direction image data of each unit row pixel group . Next, similarly to the above operation, the operation control unit 20 divides a plurality of pixel columns arranged in the column direction into a set number of pixel column units (for example, the first pixel column unit number) (for example, 2) The plurality of pixel columns are divided into a plurality of (e.g., two) unit column pixel groups C11 and C12.

Then, the operation control unit 20 sets the unit row pixel group C12 located below the two unit column pixel groups C11 and C12 among the plurality of spare row direction image data PXpr stored in the storage unit 30 And stores the preliminary column direction image data PXpc belonging to the preliminary column direction image data PXpc as processing image data in the storage unit 30. In this case as well, both the pixel position and the tone value of the processing video data are stored in the storage unit 30.

In this case, instead of detecting the collision with the obstacle by using all of the original image data inputted through the image obtaining unit 10, the collision with the obstacle is detected by using a part of the original image data, The amount of the image data to be processed is reduced, the image processing time is reduced, the capacity of the storage unit 30 is reduced, and the manufacturing cost of the front obstacle sensing apparatus is reduced.

Then, the operation control unit 20 outputs the original image data (hereinafter, referred to as 'day / night image data') belonging to the unit column pixel group C12 located above the two unit column pixel groups C11 and C12 Image data for day-and-night decision is extracted (S13).

(For example, the second pixel row unit number) and the pixel column unit number (e.g., the second pixel column unit number) by using the image data for day and night, And extracting the image data for processing from the original image data D11 is the same as the operation for extracting the image data for processing.

Therefore, the operation control section 20 divides a plurality of pixels arranged side by side in the row direction into a plurality of unit-row pixel groups (for example, a second plurality of unit-row pixel groups R2) And divided into a plurality (e.g., two) of unit column pixel groups (e.g., second unit column pixel groups) C11 and C12.

Next, the operation control unit 20 sets the row direction video data belonging to the unit column pixel group C11 of the remaining one of the divided unit column pixel groups C11 and C12 (i.e., the unit column pixel group located at the upper side) And stores it in the storage unit 30 after extracting it as data.

In the case of this example, the number of the second pixel row line number is 7, and the five pixels located consecutively in the center except for the left and right outmost pixels in each second unitary row group are extracted and divided into the second pixel row pixel group , The second pixel column unit number is 2 and is equal to the first pixel column unit number.

Generally, a front obstacle such as a vehicle is located on the lower side of the captured image obtained through the image acquiring unit 10, and a sky is located on the upper side.

Therefore, although the pixel group C11 located at the bottom of the plurality of first unit column pixel groups C11 and C12 divided for extracting the image data for processing is selected, a plurality of pixel groups C11 and C12 And selecting the pixel group C12 located at the top of the second unit column pixel groups C11 and C12 are different from each other.

4, the operation control unit 20 calculates the average gray-level value for all the pixels of the image data for day / night determination (S14).

Then, the operation control unit 40 reads the setting value for day / night determination stored in the storage unit 30 and compares the average gray-level value calculated in the step S14 with the setting value for day / night determination (S15, S16) .

If the calculated average gradation value is greater than or equal to the setting value for day / night determination, the operation control unit 20 determines the current driving time of the vehicle equipped with the front obstacle sensing apparatus as the week and stores the current driving time as the daytime (S17).

However, if the calculated average gradation value is less than the setting value for day / night determination, the operation control unit 20 determines the current driving time of the vehicle equipped with the front obstacle sensing apparatus as night and stores the present driving time at night (S18) .

At this time, the setting value for day / night determination is a value that varies from the initial value according to the external environment (i.e., brightness) of the vehicle. That is, the intensity of light in the case of daytime and nighttime is different from each other, and the tone values of the pixels located at the same position in the original image data of the image photographed in the daytime and in the original image data of the image taken in the nighttime are different. Therefore, the setting value for day / night determination is changed by the operation of the operation control section 20 in accordance with the current running state (that is, the bright and cloudy state of the weather).

If it is determined that the image data for processing is extracted and the current running time is the daytime or the nighttime, the operation control unit 20 determines whether the current time is the daytime or the nighttime, And performs binarization processing on the image data.

Accordingly, the operation control unit 20 determines whether the current driving state is the daytime using the value stored in the storage unit 30 (S19) for binarizing the processing image data.

When the driving condition is daytime, a shadow is formed near the part where the object is located, and the shadow part is darker than the other parts. As a result, the shadows in the original image data acquired during the day have darker tone values than the other portions.

However, when the driving state is nighttime, the traveling vehicle is driven with the headlights turned on, and the light of the headlights irradiated from the vehicle is irradiated in the traveling direction of the vehicle.

Accordingly, the light of the headlamp is irradiated to an object (e.g., another vehicle) located in front of the vehicle traveling in the corresponding direction, so that the part of the object that receives the light of the headlamp is reflected by another part or the other object In bright light.

As a result, the object portion irradiated with the light of the headlight from the original image data captured at night has a brightness value that is brighter than other portions.

For this reason, when the determined running state is the week, the operation control unit 20 compares the gray level value with the set gray level value in each pixel of the processing video data to determine a dark portion which is a shadow portion (S110).

If the tone value of the pixel is equal to or greater than the set tone value (S110), the operation control unit 20 determines that the pixel is a dark pixel and sets the binarization value of the pixel to '0' (S111). Otherwise, a pixel having a value smaller than the set tone value is determined as a bright pixel, and the binarization value of the pixel is set to '1' (the tone value at this time is '255') (S112).

Through these operations, binarization processing is performed on all the pixels of the processing image data, and binary image data for the processing image data is generated and stored in the storage unit 30 (S110 - S113).

However, when the determined running state is nighttime, the operation control unit 20 processes the binarization operation of the processing image data in a manner opposite to that during the daytime.

That is, the operation control unit 20 compares the gradation value and the set gradation value for all the pixels of the image data for processing to determine the bright part to which the light is irradiated at night (S114) The binarization value of the corresponding pixel is converted into '1' for displaying white, and the binarization value of the pixel not having the set tone value is converted into '0' for displaying black (S114-S117).

As described above, in the generated binarized image data obtained by binarizing the binarization value for all the pixels of the processing image data to '0' or '1' by comparing the tone value and the set tone value of all the pixels forming the processing image data, The part to be searched (that is, the part of the shadow in the case of daytime and the part of light in the case of the nighttime) is represented by '1' representing white and the part not represented by black is represented by '0'.

Next, the operation control unit 20 adjusts the setting value for day / night determination using the binarized image data (S118).

For this, the operation control unit 20 determines the total number of pixels (hereinafter, referred to as 'total black pixels') B and the total number of pixels representing white (Hereinafter referred to as a "ratio calculation coefficient") BWV is calculated by dividing the total number W of white pixels by the total number B of black pixels, .

BWV = W / B

Therefore, when the calculated ratio calculation coefficient is equal to or smaller than the reference value (e.g., 0.02), the setting value for day / night determination is increased by adding the set addition value (for example, 1) The setting value for day / night determination is subtracted from the set value for day / night determination and the setting value for day / night determination is decreased. As described above, the setting value for day / night determination newly added or subtracted by the set addition value is stored in the storage unit 30.

On the contrary, when the ratio calculation coefficient exceeds the reference value (for example, 0.02), it subtracts the set value (for example, 1) for the current day / night judgment determination value in the daytime, And adjusts the current day / night determination setting value and stores it in the storage unit 30 as a new day / night setting determination value.

In this way, since the setting value for day / night determination is adjusted according to the ratio calculation coefficient that varies depending on the clear flow of the weather, the binarized image data to which the driving time and weather conditions of the vehicle are applied is generated.

Then, the operation control unit 20 labels the binarized image data to extract obstacle candidate image data groups, and then generates an obstacle candidate image data group (S119). Using these obstacle candidate image data groups, (S1110). In this example, the labeling process uses blob labeling.

Therefore, when the binarized image data D12 as shown in FIG. 5 exists, the operation control unit 20 determines whether the pixel value of each pixel changes from '0' to '1' The change determines the position of the pixel and calculates an area in which a plurality of pixels having a pixel value of '1' are located adjacent to each other in at least one of a row direction and a horizontal direction, To the storage unit 30 (S119).

In the case of FIG. 4, all four obstacle candidate image data groups OCG11-OCG14 are formed.

At this time, the operation control unit 20 knows the positions (i.e., row positions and column positions) of the pixels belonging to each of the obstacle candidate image data groups OCG11-OCG14.

Accordingly, the operation control unit 20 determines the column positions (Y1, Y2) of the pixels located at the lowest and uppermost positions in each of the obstacle candidate image data groups OCG11-OCG14 and the row positions (X1, Y2), (X1, Y2) obtained by combining the two column positions (Y1, Y2) and the two row positions (X1, X2) The area of the rectangle formed with vertexes of the pixels located at the vertexes of the obstacle candidate areas OCA11 to OCA14 is stored in the storage unit 30 at step S120. At this time, the operation control unit 20 also stores the positions of the pixels belonging to the obstacle candidates OCA11-OCA14 in the storage unit 30. [

(PX _{5 , 10} ), (PX _{5 , 12} ), and (PX ₄ ), which form four vertexes of a quadrangle in the obstacle candidate image area OCA 11, among the plurality of obstacle candidate areas OCA 11 to OCA 14 shown in FIG. (PX _9,10 ) and (PX _9,12 )].

Next, the operation control unit 20 selects only the obstacle candidate regions located within the predetermined distance from the same lane as the vehicle on which the front obstacle sensing apparatus of this example is mounted, among the determined obstacle candidate regions (OCA), and determines the final obstacle region.

To this end, the operation control unit 20 determines whether there is an obstacle candidate area (OCA) calculated by the operation in step S1110 (S121).

The obstacle candidate region may not be calculated if the vehicle is traveling on a road such as a desert or the like without building or trees on both sides of the driving road and no other vehicle or obstacle is present in the front.

Therefore, in this case, the operation control unit 20 determines that there is no obstacle ahead and proceeds to the first step (S11).

When there is at least one calculated obstacle candidate area (OCA), the operation control part 20 judges whether the calculated at least one obstacle candidate area (OCA) is a valid area for the obstacle judgment operation ahead.

Accordingly, the operation control unit 20 reads the distance measurement data stored in the storage unit 30 and outputs the distance measurement data to the display unit 40 to display the distance meter DI (S122) (see FIG. 6).

The distance meter DI displayed on the display unit 40 is displayed on the display unit 40 in consideration of the shape of the actual traveling road when the actual driving vehicle is viewed from the traveling road, ) Is proportional to the measured distance.

6, the distance meter DI displayed on the display unit 40 includes a plurality of horizontal lines HL partitioned in the horizontal direction along the running direction of the road and one horizontal line HL extending in the middle of the plurality of horizontal lines HL There is a vertical line LL.

Therefore, the distance meter DI is divided into a plurality of areas by the plurality of horizontal lines HL, and the vertical length of each area (i.e., the distance between two adjacent horizontal lines HL in the traveling direction) The vertical lengths LL1 of the respective regions are the same distance on the actual roads although they have different sizes on the display of the display unit 40. For example, 2m.

When the distance meter DI is displayed on the display unit 40 as described above, the operation control unit 20 uses the positional information of each pixel of the obstacle candidate area OCA stored in the storage unit 30, (Hereinafter, referred to as an " obstacle leftmost pixel position ") and a position of a pixel positioned at the rightmost position (hereinafter, referred to as a " rightmost pixel position of an obstacle " (S123).

Next, the operation control unit 20 uses the position information of the pixel for the distance meter DI stored in the storage unit 30 when the distance meter DI is displayed on the display unit 40, (Hereinafter, referred to as 'distance system leftmost pixel position') PX _{X11 and Y11} and the position of the rightmost pixel (hereinafter referred to as the 'rightmost pixel of the distance system'_Quot; position ") PX _{X12, Y11} (S124).

Then, the operation control unit 20 determines whether the leftmost pixel position of the obstacle is the same as the leftmost pixel position of the distance system or on the right side (S125).

If it is determined that the leftmost pixel position of the obstacle is equal to or further to the right of the leftmost pixel position of the distance system, the operation control unit 20 determines whether the rightmost pixel position of the obstacle is the same as or further to the left of the rightmost pixel position of the distance system (S127).

However, if it is determined that the leftmost pixel position of the obstacle is located to the left of the leftmost pixel position of the distance system, the operation control unit 20 does not select the obstacle candidate region OCA as the final obstacle region (S126).

If it is determined in step S127 that the rightmost pixel position of the obstacle is equal to or further to the left of the rightmost pixel position of the distance system, the operation control unit 20 sets the obstacle candidate area OCA as the final obstacle area (S129 ).

If it is determined in step S127 that the rightmost pixel position of the obstacle is located to the right of the rightmost pixel position of the distance system, the operation control unit 20 excludes the obstacle candidate area OCA from the final obstacle area (S128).

In this manner, the operation control unit 20 compares the leftmost pixel position and the rightmost pixel position of the obstacle with the positions of the leftmost pixel and the rightmost pixel of the distance system with respect to all the obstacle rear regions (OCA) Only the obstacle candidate region OCA located within the range (X11 - X12) of the pixel row defined by the position and the distance to the left side pixel position is selected (S130).

At this time, the operation control unit 20 can compare only the row positions at the row positions and column positions of the most left and rightmost pixel positions of the obstacle and the rightmost and rightmost pixel positions of the distance meter with each other.

When the final obstacle area is determined, the operation control unit 20 determines a final obstacle area located at the lowermost end of the omnidirectional area in the final obstacle area (S131), and determines a final obstacle area The final obstacle area is referred to as the " nearest obstacle area ") as an obstacle located closest to the vehicle.

Next, the operation control unit 20 uses the position of the closest obstacle area located within the range system to determine the distance between the vehicle (i.e., the vehicle on which the front obstacle sensing apparatus of this example is mounted) and the closest obstacle area (S132 ). At this time, the operation control unit 20 can calculate the separation distance from the vehicle based on the position of the center pixel of the closest obstacle area.

When the distance between the nearest obstacle area and the vehicle is calculated, the operation control unit 20 determines whether the calculated distance is less than the set distance (S133).

If the calculated distance is less than the set distance, the operation control unit 20 determines that there is a risk of collision between the vehicle and an obstacle displayed in the nearest obstacle area, and outputs a warning signal to the warning unit 50 (S134).

Therefore, the warning unit 50 operates according to the warning signal applied from the operation control unit 20 to warn the driver of the risk of collision with the obstacle ahead.

However, if the calculated distance exceeds the set distance, the operation control unit 20 determines that there is no risk of collision between the vehicle and the obstacle and returns to the initial step (S100).

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

10: Image acquisition unit 20: Operation control unit
30: storage unit 40: display unit
50: Warning section

Claims (6)

An image acquiring unit for generating original image data for an image in a direction in which the vehicle is traveling;
An obstacle candidate image data group, which is connected to the image acquisition unit, generates an obstacle candidate image data group using the original image data, extracts an obstacle candidate region from the obstacle candidate image data group, An operation control unit for determining a distance between the obstacle candidate areas and outputting a warning signal when the determined distance is less than a set distance; And
And an alarm unit connected to the operation control unit and operating according to the warning signal,
The obstacle candidate region extraction is performed by dividing a plurality of pixels sequentially arranged in the row direction for each row by the number of first pixel line units that are set and arranging a plurality of pixels arranged in the row direction for each row in a plurality of first unit rows The set number of pixels PXpr located in the middle portion excluding the leftmost and rightmost pixels in each first unit row pixel group R1 in the obstacle candidate candidate image data group divided by the pixel group R1 is extracted, And dividing the plurality of pixel arrays into a plurality of unit column pixel groups (C11, C12) and dividing the plurality of preliminary row direction image data (PXpr) by a predetermined number Column direction image data PXpc belonging to the unit column pixel group C12 located below the unit column pixel groups C11 and C12. The method of claim 1,
The operation control unit extracts a part of the original image data to generate processing image data, generates binary image data by binarizing the processing image data, labels the binarized image data to generate the obstacle candidate image data group Obtained front obstacle detection device. 3. The method of claim 2,
The operation control unit,
When the gray scale value of each pixel of the processing video data is equal to or greater than the set gray level value when the running time is the week, the binarization value of the corresponding pixel among '0' and '1' is converted into a value for displaying white, Converts the binarization value of the corresponding pixel among '0' and '1' to a value for displaying black when the tone value of each pixel is less than the set tone value, converts the processing image data into binary image data,
When the driving time is night, when the tone value of each pixel of the binarized image data is equal to or greater than the set tone value, the binarization value of the corresponding pixel among '0' and '1' is converted into a value indicating black color, And converting the binarization value of the corresponding pixel among '0' and '1' to a value for displaying white when the tone value of each pixel is less than the set tone value, and converting the processing image data into binarized image data . 4. The method of claim 3,
Wherein the operation control unit extracts video data for day-and-night determination from the video data for processing, and when the average tone value of the video data for day / night determination is less than the setting value for day / night determination, When the average gradation value of the image data for the vehicle is equal to or larger than the set value for day / night determination, the running time of the vehicle is determined as the week. 5. The method of claim 4,
Wherein the setting value for day / night determination is added or subtracted according to a ratio of the total number of pixels showing black color and the total number of pixels showing white color in the binarized image data. 3. The method of claim 2,
The operation control unit may set the obstacle candidate region located in the range of the pixel row determined by the rightmost pixel position to the leftmost pixel position of the distance system of the distance meter as the final obstacle region as the obstacle candidate left- And determines a distance between the vehicle and a final obstacle area located nearest to the vehicle, out of the final obstacle area as the distance.

Images