Device and method for driver assistance

Abstract

Driver assistance device, which includes: a top-view image output unit (100) which outputs a top-view image at a viewing angle in which the environment of the vehicle is viewed downwards from above; a lane detection unit (200) which detects a left lane of the vehicle and a right lane of the vehicle from the top-down image using a rectangular filter (top hat filter); a speed measuring unit (300) which outputs a speed of the vehicle; and an integrated control unit (400) which selects and executes a parking assistance function and a lane departure warning function in accordance with the speed, wherein the driver assistance device is configured to perform a step of detecting a parking area guide line in a first top-down image photographed at a first time point (S120), a step of predicting a first gradient of left and right parking lines in a second top-down image photographed at a second time point after the first time point (S160), and estimating a second gradient of the left and right parking lines based on position information of the vehicle (S170), a step of determining whether a difference value between the first and second gradients is less than a preset threshold difference value (S180), and a step of generating a top-down composite image (S190),including left and right parking guide lines, corresponding to the left and right parking lines, if the difference value is less than the threshold difference value to detect the parking space.

Description

CROSS-REFERENCE TO RELATED REGISTRATIONS

[0001] The present application claims priority over Korean patent application No. 10-2014-0141126, filed on October 17, 2014, and Korean patent application No. 10-2014-0149632, filed on October 30, 2014, the entire contents of which are incorporated herein by reference for all purposes. TECHNICAL AREA

[0002] The present invention relates to a driver assistance device and a driver assistance method, and more precisely to a driver assistance device and a driver assistance method which detect a lane in the vicinity of a vehicle and, in accordance with a speed of the vehicle, perform a parking assistance and lane departure warning function. BACKGROUND

[0003] Recently, many vehicles have been equipped with an intelligent parking assist system (SPAS) and a lane departure warning system (LDWS) for the safety of the driver and pedestrian.

[0004] Both SPAS and LDWS use a lane detection technique. However, both systems use separate lane detection algorithms, making it inefficient for the system as a whole.

[0005] Recently, a technique has been studied that integrates a lane detection technique to operate both parking assistance and lane departure warning systems in accordance with the different situation.

[0006] DE 10 2013 103 952 A1 describes a system and a method for providing lane detection in a vehicle by detecting lane markings, wherein the system uses a surround-view camera system that provides a top-down image around the vehicle. The method includes detecting left and right lane markings in the top-down image and subsequently determining whether the lane markings are aligned from one frame to the next and whether they are aligned from frame to frame in the top-down image. If the lane markings are not aligned, one or more of the cameras are then calibrated, and if the lines are aligned, a model fitting method is then used to specifically identify the position of the lane markings on the road.

[0007] DE 10 2007 039 374 A1 describes a motor vehicle with a lane change or lane keeping assistance system, comprising monitoring means for the area in front and / or behind the vehicle, and a parking assistance system, comprising several ultrasonic sensors provided at the front and / or rear for environmental detection, wherein the parking assistance system can be activated during driving by simultaneously switching to operation of at least one ultrasonic sensor with an enlarged detection range and the information supplied by its at least one ultrasonic sensor can be processed by the lane change or lane keeping assistance system. SUMMARY

[0008] The present invention was made in an effort to unify a lane detection algorithm used in a parking assistance system and a lane departure warning system in order to eliminate the inefficiency of the system.

[0009] The present invention was made in an effort to improve lane detection accuracy and to automatically activate the parking assistance system or the lane departure warning system in accordance with the speed of the vehicle, in order to increase convenience for a user.

[0010] The objectives of the present invention are not limited to the aforementioned objectives, and other objectives not mentioned above will be understood by the person skilled in the art with reference to the following descriptions.

[0011] The above objectives are achieved by a driver assistance device according to claim 1 and a driver assistance method according to claim 7. Claims 2 to 6 relate to particularly advantageous implementations of the driver assistance device according to claim 1. Claims 8 to 15 relate to particularly advantageous implementations of the driver assistance method according to claim 7.

[0012] An exemplary embodiment of the present invention provides a driver assistance device, including: a top-view image output unit which outputs a top-view image from a top-down perspective, a lane detection unit which detects a left lane of the vehicle and a right lane of the vehicle from the top-view image using a rectangular filter (top hat filter), a speed measurement unit which outputs a speed of the vehicle, and an integrated control unit which selectively executes one of a parking assistance function and a lane departure warning function in accordance with the speed.

[0013] Another exemplary embodiment of the present invention provides a driver assistance method, including: an image output step of assembling images in the vehicle's environment to output a top-view image; an image conversion step of converting the top-view image into a grayscale image; a lane detection step of detecting the left and right lanes of the vehicle from the grayscale image using a rectangular filter; an effectiveness determination step of detecting an interval between the left and right lanes to determine the effectiveness of the detected lane; a speed measurement step of measuring the vehicle's speed; and a function activation step of selecting and activating a parking assistance function and a lane departure warning function in accordance with the speed.

[0014] According to a driver assistance device and a driver assistance method of the present invention, one or more effects are as follows: First, a track is detected in a top-down image of a left side and a right side of the vehicle using a rectangular filter, and the effectiveness of a parallel track is determined by a preset interval value, thereby improving the track detection accuracy.

[0015] Secondly, a parking assistance system or a lane departure warning system is automatically activated in accordance with the vehicle's speed, thereby increasing user comfort.

[0016] The effects of the present invention are not limited to the aforementioned effects, and other effects not mentioned above will be clearly understood by those skilled in the field from the recitation of the claims. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a diagram illustrating a configuration of a driver assistance device according to an exemplary embodiment of the present invention. Fig. Figure 2 is a block diagram illustrating a control configuration of a top view image output unit and a lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention. Fig. Figure 3 is a flowchart illustrating a method of detecting a parking space by a driver assistance device according to an exemplary embodiment of the present invention. Fig. Figure 4 is a flowchart illustrating a driver assistance method according to an exemplary embodiment of the present invention. Fig. Figure 5 is an exemplary view illustrating that a lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention calculates a feature point using a rectangular filter. Fig. Figure 6 is an exemplary view illustrating that a lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention extracts a line component from the calculated feature point. Fig. Figure 7 is an exemplary view illustrating that a lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention detects a parking area guide line. Fig. Figure 8 is an exemplary view illustrating that a lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention detects a parking line. Fig. Figure 9 is an exemplary view illustrating an image in which left and right parking guide lines overlap in a driver assistance device according to an exemplary embodiment of the present invention. Fig. Figure 10 is an exemplary view for determining the effectiveness of a lane taking into account a lane interval detected by the lane detection unit of a driver assistance device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0017] The advantages and features of the present invention and methods for carrying it out will become clear with reference to exemplary embodiments, which are described in detail below together with the accompanying drawings. However, the present invention is not limited to the exemplary embodiments set forth below and can be implemented in many other forms. The present embodiments serve to provide a complete description of the present invention and are presented to give a person skilled in the art in the field to which the present invention belongs a comprehensive understanding of the scope of the invention, and the present invention is defined only by the scope of the claims. Identical reference numerals within the description denote identical elements.

[0018] A driver assistance device and a driver assistance method according to an exemplary embodiment of the present invention are described below with reference to the drawings.

[0019] Fig. Figure 1 is a diagram illustrating a configuration of a driver assistance device according to an exemplary embodiment of the present invention. A driver assistance device according to an exemplary embodiment of the present invention includes a top-view image output unit 100, a lane detection unit 200, a speed measurement unit 300, and an integrated control unit 400.

[0020] In particular, the driver assistance device according to an exemplary embodiment of the present invention can include the top-view image output unit 100, which outputs a top-view image from a viewing angle in which the vehicle's surroundings are viewed from top to bottom, the lane detection unit 200, which detects a left lane of the vehicle and a right lane of the vehicle from the top-view image using a rectangular filter, a speed measurement unit 300, which outputs a speed of the vehicle, and the integrated control unit 400, which selectively activates a parking assistance function and a lane departure warning function in accordance with the speed.

[0021] The top-down image is an image taken from a perspective where the vehicle's surroundings are viewed from the top of the vehicle looking downwards. This image is generated by photographing the vehicle's surroundings using a camera located inside the vehicle and combining these images. The top-down image output unit 100 can be part of an all-around monitoring system (AVM). The vehicle can include at least one camera positioned at the front, rear, left, and right sides. The top-down image output unit 100 outputs a top-down image obtained by photographing the vehicle's surroundings using cameras located at the front, rear, and sides of the vehicle and combining these images into a single image.

[0022] When a rectangular filter is used, it is called a rectangular transformation, which is used to capture a portion of a grayscale image where the brightness changes. For example, if there is a bright area within a dark area of ​​the grayscale image, applying the rectangular filter will capture a point where the dark and light areas meet. This is because the brightness changes most significantly at the point where the dark and light areas meet. When the track is captured in the grayscale image, the track is a dark area, and the rest of the ground is dark. Applying the rectangular filter to the grayscale image will capture a boundary portion of the track. A method for calculating a feature point using the rectangular filter by the Track Detection Unit 200 is described with reference to Fig. 2 described.

[0023] The track detection unit 200 applies the rectangular filter to the grayscale image of the top view in a horizontal direction, and detects a maximum value of a brightness change and considers the section where the maximum value is detected as a left feature point and a right feature point of the track.

[0024] The lane detection unit 200 takes into account a central section of the left feature point and the right feature point of the lane as a vehicle center point.

[0025] The track detection unit 200 scans all directions based on the left and right feature points to detect a section that will be recognized as a track. A method for detecting the section that will be recognized as a track from the feature point calculated by the track detection unit 200 is described below with reference to Fig. 3 described.

[0026] The track detection unit 200 compares an angular error formed by the section which is detected as a track, and grants effectiveness to the line component if the angular error is equal to or less than a predetermined value, and represents the line component as a track candidate group based on the center point of the track.

[0027] The lane detection unit 200 can determine whether an interval between a detected left lane of the vehicle and a detected right lane of the vehicle has a preset interval value in order to determine lane effectiveness. A method for determining by the lane detection unit 200 whether an interval between the detected left lane and the detected right lane has a preset interval value Vd is described below with reference to Fig. 4 described.

[0028] As illustrated in Fig. 2, the lane detection unit 200 can include a detection unit 150 and a control unit 160, and the top view output unit 100 inputs a top view image to the lane detection unit 200.

[0029] This means that the track detection unit 200 applies the rectangular filter to the grayscale image of the top view in a horizontal direction, detects a maximum value of the brightness change, considers the section where the maximum value is detected as a left feature point and a right feature point of the track, considers a middle section of the left feature point and the right feature point of the track as a midpoint of the track, searches all directions based on the left feature point and the right feature point to detect a section which is detected as a track, compares an angular error formed by the section which is detected as a track, and allows effectiveness of the line components if the angular error is equal to or less than a predetermined value, and represents the line component as a track candidate group based on the midpoint of the track.

[0030] The top-view image output unit 100 can include multiple cameras that photograph the surroundings, including front and rear images and left and right images of the vehicle.

[0031] The lane detection unit 200 can include the recognition unit 150, which detects a parking area guide line and left and right parking lines from the top view image, and the control unit 160, which predicts a gradient direction of a parking line that runs perpendicular to the parking area guide line and a first gradient of the left and right parking lines, estimates a second gradient of the left and right parking lines based on the first gradient and a position coordinate of the vehicle in order to generate left and right parking guide lines, corresponding to the left and right parking lines, if a difference value between the first and second gradients is less than a predetermined threshold difference value, in order to generate a composite image in which the left and right parking guide lines overlap the top view image.

[0032] The recognition unit 150 can include an image conversion unit 152, which converts the top-view image into a grayscale image, a rectangular filter unit 154, which extracts a plurality of guide line feature points and a plurality of parking line feature points using an illumination change of the grayscale image, and a lane detection unit 156, which extracts line components of the plurality of guide line feature points and the plurality of parking line feature points in order to detect the parking area guide lines and the left and right parking lines.

[0033] As illustrated in Fig. 7, the rectangular filter unit 154 shows a value obtained by measuring the change in illumination with respect to the parking area guide line located horizontally in a target parking area.

[0034] In this case, the rectangular filter unit 154 measures a value in the shape of a triangle along a width of the parking area guide line and extracts a mean value of the triangular shape as the majority of guide line feature points.

[0035] As illustrated in Fig. Figure 8 shows the rectangular filter unit 154, similar to the above description, a value obtained by measuring the change in illumination with respect to the left and right parking lines, which are vertically located in the target parking area.

[0036] The control unit 160 can include a forecasting unit 162, an estimation unit 164 and a generation unit 166.

[0037] In particular, the control unit 160 can include the prediction unit 162, which predicts a gradient direction of the parking line, which runs perpendicular to the parking area guide line, and a first gradient of the left and right parking lines; the estimation unit 164, which estimates a second gradient of the left and right parking lines based on the first gradient and the position coordinates of the vehicle; and the generation unit 166, which generates the composite image when the difference value between the first and second gradients is less than a predetermined threshold difference value.

[0038] The prediction unit 162 can predict the first gradient based on a coordinate of the image of the left and right parking lines and the position coordinates of the vehicle.

[0039] The threshold difference value can range from 1° to 11°.

[0040] The driver assistance device according to the exemplary embodiment of the present invention can further include a parking assistance unit 500, which detects a parking space based on the lane detected by the lane detection unit, and parks the vehicle in the parking space when the parking space is detected, and in this case the integrated control unit 400 actuates the parking assistance unit 500 when the speed is equal to or lower than a first threshold value.

[0041] For example, if the vehicle's speed is equal to or less than 20 km / h (kilometers per hour), the Parking Assist Unit 500 can perform the parking assistance function. The Parking Assist Unit 500 can be an intelligent parking assistance system (SPAS). When the Parking Assist Unit 500 performs the parking assistance function, it takes into account the distance between the left and right lanes of the vehicle, as output by the Lane Detection Unit 200. If the distance between the left and right lanes of the vehicle is less than the width of the vehicle, the Parking Assist Unit 500 can search for a different parking space to park the vehicle.

[0042] Fig. Figure 3 is a flowchart illustrating a method of detecting a parking space by a driver assistance device according to an exemplary embodiment of the present invention.

[0043] Referring to Fig. 3. The driver assistance device extracts a plurality of guide line feature points from an initial grayscale image, which is converted from an initial top-down image photographed at an initial time point using a change in illumination in a horizontal direction in step S110, extracts a line component for the plurality of guide line feature points to detect the parking area guide lines in step S120, and predicts a gradient direction of the parking line that is perpendicular to the parking area guide line in step S130.

[0044] The driver assistance device then extracts a plurality of parking line feature points using a vertical illumination change from a second grayscale image, which is taken from the second top-down image at a second time point which is later than the first time point in step S140, extracts a line component for the plurality of parking line feature points to detect the left and right parking lines in step S150, and predicts the first gradient based on an image coordinate of the left and right parking lines and a position coordinate of the vehicle in step S160.

[0045] The driver assistance device can estimate the second gradient based on the first gradient and the vehicle's position coordinates in step S170.

[0046] The driver assistance device determines whether a difference value between the first and second gradients is smaller than a predetermined threshold difference value in step S 180, and generates the top-view composite image in which the left and right parking guide lines overlap when the difference value is smaller than the threshold difference value in step S 190.

[0047] Next, in step S200, the driver assistance device can display the top-view composite image.

[0048] The driver assistance device according to the exemplary embodiment of the present invention can further include a driver warning unit 600 which issues a warning to the driver if the vehicle leaves the lane while driving, and the integrated control unit 400 can actuate the driver warning unit 600 if the speed is equal to or higher than a second threshold value.

[0049] The Driver Warning Unit 600 detects movements of the vehicle's steering wheel to determine whether a lane-changing movement is taking place and issues a warning to the driver if the vehicle leaves its lane without having been moved to change lanes.

[0050] The Driver Warning Unit 600, for example, issues a lane departure warning when the vehicle's speed exceeds 50 km / h. This lane departure warning, issued by the Driver Warning Unit 600, can be part of a Lane Departure Warning System (LDWS). The Driver Warning Unit 600 detects the lane identified by the Lane Detection Unit 200 and warns the driver if the vehicle drifts out of the detected lane without driver input. The Driver Warning Unit 600 receives a steering input signal via a Control Area Network (CAN) to determine whether the driver intended the vehicle to leave the lane.

[0051] Fig. Figure 5 is an example view illustrating how the Lane Detection Unit 200 calculates a feature point using a rectangular filter. The Lane Detection Unit 200 converts the top-down image to a grayscale image. The Lane Detection Unit 200 applies the rectangular filter to the grayscale image in a horizontal direction. The Lane Detection Unit 200 captures a maximum value in a local area to capture a midpoint CP of the lane. The Lane Detection Unit 200 captures the maximum value in the local area with respect to a boundary between a left side of the lane and a right side of the lane, based on the midpoint, to calculate a left lane feature point LP and a right lane feature point RP. The left lane feature point LP and the right lane feature point RP calculated by the Lane Detection Unit 200 are located between a dark part D and a light part B.

[0052] Fig. Figure 6 is an exemplary view illustrating that a line component is extracted from the feature point detected by the lane detection unit 200. Based on the left lane feature point LP and the right lane feature point RP, the lane detection unit 200 scans all radiating directions to extract a line component. The lane detection unit 200 can scan a 360-degree direction with respect to the left lane feature point and the right lane feature point to extract a line component. According to an exemplary embodiment, the lane detection unit 200 can scan a clockwise direction with respect to the right lane feature point RP. Fig. The lane detection unit 200 can search either the left lane marker point LP and a direction counterclockwise relative to the left lane marker point LP. It can also search a direction counterclockwise relative to the right lane marker point RP or a direction clockwise relative to the left lane marker point LP.

[0053] The track detection unit 200 checks the effectiveness of the line components by comparing the angular error of the extracted line components. The angular error taken into account by the track detection unit 200, according to an exemplary embodiment, is a preset value and can be adjusted within a range of 5 degrees. If the effectiveness of the line components is satisfied, the track detection unit 200 stores a track candidate group with respect to the center point CP of the track.

[0054] Fig. Figure 10 is an example view illustrating how the Lane Detection Unit 200 determines lane effectiveness by considering an interval of detected lanes. The Lane Detection Unit 200 determines lane effectiveness in at least one of a parking situation and a road driving situation. The Lane Detection Unit 200 determines whether a detected interval of the vehicle's left and right lanes has a preset value Vd to determine lane effectiveness. The interval value Vd, which is preset in the Lane Detection Unit 200, can be stored within the unit. The interval value Vd of the parking line can be set in real time in the Lane Detection Unit 200 by receiving geographic information via wireless communication. The Lane Detection Unit 200 can determine lane effectiveness in continuous top-down images.

[0055] Fig.Figure 4 is a flowchart illustrating a driver assistance method according to an exemplary embodiment of the present invention.

[0056] In particular, a driver assistance method according to an exemplary embodiment of the present invention may include an image output step of compiling images in the vehicle's environment to output a top-view image, an image conversion step of converting the top-view image into a grayscale image, a lane detection step of detecting a left lane and a right lane of the vehicle from the grayscale image using a rectangular filter, an effectiveness determination step of detecting an interval between the left lane and the right lane to determine the effectiveness of the detected lane, a speed measurement step of measuring a speed of the vehicle, and a function actuation step of selecting and actuating, in accordance with the speed, a parking assistance function and a lane departure warning function.

[0057] In the lane detection step, a maximum value of a brightness change is detected in an area where the rectangular filter is applied to the grayscale image, and the area where the maximum value is detected is considered as a left feature point and a right feature point of the lane in order to detect a left lane of the vehicle and a right lane of the vehicle.

[0058] In the lane detection step, a central area of ​​the left feature point and the right feature point of the lane is considered as a center point of the lane in order to detect the left lane and the right lane of the vehicle.

[0059] In the lane detection step, all directions are searched based on the left feature point and the right feature point, and a portion that is detected as a lane is captured to determine the left lane and the right lane of the vehicle.

[0060] In the lane detection step, an angular error formed by the section being detected as a lane is compared, and if the angular error is equal to or less than a predetermined value, effectiveness of the line components is granted, and based on the center point of the lane, the line component is represented as a lane candidate group to detect the left lane and the right lane of the vehicle.

[0061] In the lane detection step, a parking space is detected based on the detected lane, and in the function activation step, if the speed is equal to or less than a first threshold value, the parking assistance function can be activated.

[0062] In the track acquisition step, a maximum value of brightness change is captured in an area where the rectangular filter is applied to the grayscale image. The section where the maximum value is captured is considered a left feature point and a right feature point of the track. A central section of the left feature point and the right feature point of the track are considered the midpoint of the track. Based on the left feature point and the right feature point, all directions are searched to capture a section that is captured as a track. An angular error formed by the section captured as a track is compared, and the effectiveness of the line component is granted if the angular error is equal to or less than a predetermined value. The line component is then represented as a track candidate group based on the midpoint of the track.This detects the left lane and the right lane of the vehicle.

[0063] The lane detection step can include a step of detecting a parking area guide line in a first top-down image photographed at a first time point, a step of predicting a first gradient of left and right parking lines in a second top-down image photographed at a second time point after the first time point, and estimating a second gradient of the left and right parking lines based on vehicle position information, a step of determining whether a difference value between the first and second gradients is less than a preset threshold difference value, and a step of generating a top-down composite image, including left and right parking lines, according to the left and right parking lines if the difference value is less than the threshold difference value, in order to detect the parking space.

[0064] The step of predicting and estimating the gradient can include a step of predicting the first gradient and a step of estimating the second gradient.

[0065] The first gradient prediction step includes a step of extracting a plurality of parking line feature points using a vertical illumination change from a second grayscale image, which is converted from the second top-down image; a step of extracting a line component for the plurality of parking line feature points to detect the left and right parking lines; and a step of predicting the first gradient based on an image coordinate of the left and right parking lines and a position coordinate of the vehicle.

[0066] In the step of estimating a second gradient, the second gradient can be estimated based on the first gradient, which is predicted in the step of predicting the first gradient and the position coordinate of the vehicle.

[0067] The threshold difference value can range from 1° to 11°.

[0068] In the step of generating a top-view composite image, the top-view composite image is created in which the left and right parking guide lines generated on the basis of each of the first and second gradients overlap the left and right parking lines.

[0069] The driver assistance method according to the exemplary embodiment of the present invention may further include a step of displaying the top-view composite image.

[0070] The parking assistance function can be a function of automatically parking the vehicle in the detected parking space.

[0071] During the function activation step, if the speed is equal to or higher than a second threshold, the lane departure warning function can be activated.

[0072] According to the lane departure warning function, movement of the vehicle's steering wheel is detected to determine whether there is movement to change lanes, and a warning is issued to the driver if the vehicle leaves the lane without having been moved to change lanes.

[0073] The driver assistance method according to the exemplary embodiment of the present invention may further include a step of detecting a lane after changing lanes, when the vehicle leaves a lane in accordance with movement to change lanes.

[0074] The driver assistance method according to the exemplary embodiment of the present invention can further include a step of displaying the lane detected in the lane detection step by means of a line with a preset color for outputting the lane. Therefore, the driver can check whether the lane is detected in real time.

[0075] Preferred embodiments of the present invention have been described and illustrated above, but the present invention is not limited to the specific embodiments above; it is obvious that various modifications can be made by those skilled in the art without departing from the core of the present invention as claimed in the claims, and such modifications should not be understood individually from the technical spirit or expectation of the present invention.

Claims

[1] Driver assistance device comprising: a top-view image output unit (100) which outputs a top-view image at a viewing angle in which the environment of the vehicle is viewed downwards from above; a lane detection unit (200) which detects a left lane of the vehicle and a right lane of the vehicle from the top-down image using a rectangular filter (top hat filter); a speed measuring unit (300) which outputs a speed of the vehicle; and an integrated control unit (400) which selects and executes a parking assistance function and a lane departure warning function in accordance with the speed, wherein the driver assistance device is configured to perform a step of detecting a parking area guide line in a first top-down image photographed at a first time point (S120), a step of predicting a first gradient of left and right parking lines in a second top-down image photographed at a second time point after the first time point (S160), and estimating a second gradient of the left and right parking lines based on position information of the vehicle (S170), a step of determining whether a difference value between the first and second gradients is less than a preset threshold difference value (S180), and a step of generating a top-down composite image (S190),including left and right parking guide lines, corresponding to the left and right parking lines, if the difference value is less than the threshold difference value to detect the parking space. [2] Device according to claim 1, wherein the top view image output unit (100) includes a plurality of cameras which photograph the surrounding image including front and rear images and left and right images of the vehicle. [3] Device according to claim 1 or 2, wherein the lane detection unit (200) determines whether an interval between a detected left lane of the vehicle and a detected right lane of the vehicle has a preset interval value in order to determine lane effectiveness. [4] Device according to any one of the preceding claims, further comprising: a parking assistance unit (500) which detects a parking space based on the lane detected by the lane detection unit (200) and parks the vehicle in the parking space when the parking space is detected, wherein the integrated control unit (400) activates the parking assistance unit (500) when the speed is equal to or less than a first threshold value. [5] Device according to any one of the preceding claims, further comprising: a driver warning unit (600) which issues a warning to the driver if the vehicle leaves its lane while driving, wherein the integrated control unit (400) activates the driver warning unit (600) when the speed is equal to or less than a second threshold. [6] Device according to one of the preceding claims, wherein the driver warning unit (600) detects movement of a steering wheel of the vehicle to determine whether a movement to change lanes is taking place and issues a warning to the driver when the vehicle leaves the lane without having been moved to change lanes. [7] Driver assistance systems, which include: an image output step of assembling images from the vehicle's environment to output a top-down image; an image conversion step of converting the top-view image into a grayscale image; a track detection step of capturing a left track and a right track of the vehicle from the grayscale image using a rectangular filter (top hat filter); an effectiveness determination step of capturing an interval between the left lane and the right lane in order to determine the effectiveness of the captured lane; a speed measurement step of measuring the speed of the vehicle; and a function activation step of selecting and activating a parking assistance function and a lane departure warning function in accordance with the speed, wherein the lane detection step includes a step of detecting a parking area guide line in a first top-down image photographed at a first time point (S120), a step of predicting a first gradient of left and right parking lines in a second top-down image photographed at a second time point after the first time point (S160), and estimating a second gradient of the left and right parking lines based on position information of the vehicle (S170), a step of determining whether a difference value between the first and second gradients is less than a preset threshold difference value (S180), and a step of generating a top-down composite image (S190),including left and right parking guide lines, corresponding to the left and right parking lines, if the difference value is less than the threshold difference value to detect the parking space. [8] Method according to claim 7, wherein in the lane detection step a parking space is detected on the basis of the detected lane and in the function activation step, when the speed is equal to or less than a first threshold value, the parking assistance function is activated. [9] A method according to claim 7 or 8, wherein the step of predicting and estimating the gradient includes a step of predicting the first gradient and a step of estimating the second gradient, the step of predicting the first gradient includes a step of extracting a plurality of parking line feature points using a vertical illumination change from a second grayscale image, which is converted from the second top-down image, a step of extracting a line component for the plurality of parking line feature points to detect left and right parking lines, and a step of predicting the first gradient based on an image coordinate of the left and right parking lines and a position coordinate of the vehicle, and in the step of estimating the second gradient, the second gradient is estimated based on the first gradient.which is predicted in the step of predicting the first gradient and the position coordinates of the vehicle. [10] Method according to one of claims 7 to 9, wherein the threshold difference value is from 1° to 11°, wherein in the step of generating a top view composite image the top view composite image in which the left and right parking guide lines generated on the basis of one of the first and second gradients overlap the left and right parking lines is generated, and furthermore a step of displaying the top view composite image is included. [11] Method according to any one of claims 7 to 10, wherein the parking assistance function automatically parks the vehicle in the detected parking space. [12] Method according to any one of claims 7 to 11, wherein in the function activation step, when the speed is equal to or higher than a second threshold value, the lane departure warning function is activated. [13] Method according to any one of claims 7 to 12, wherein, according to the lane departure warning function, a movement of a steering wheel of the vehicle is detected to determine whether a movement to change lanes is taking place and a warning is issued to the driver when the vehicle leaves the lane without being moved to change lanes. [14] Method according to any one of claims 7 to 13, further comprising: a step in the process of capturing a track after changing tracks, when the vehicle leaves a lane in accordance with a movement to change lanes. [15] Method according to any one of claims 7 to 14, wherein the track detection step further includes a step of displaying the detected track by means of a track with a preset color for outputting the track.

Images