JP2001014595A - Device for detecting traffic lane for vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the erroneous recognition of a traffic lane when driving a vehicle by detecting the type of a white line with high accuracy. SOLUTION: This traffic lane detecting device for a vehicle picks up the image of a travel road situation with an imaging means installed on the front part of a vehicle and detects the type of the white line of the travel road on the basis of an image obtained by the imaging means. In such a case, the device is provided with a white line candidate point detecting part 23 which processes the image picked up by the imaging means and detects white line candidate points, a white line candidate point selecting part 24 which selects only a white line candidate point recognized as a white line among the white line candidate points and calculates the detection rate of white line candidate points from the selected white line candidate points, a white line candidate data storing part 25 which stores the detection rate of the white line candidate points for a fixed quantity in the past and a white line discriminating part 26 which discriminates the type of a white line on the basis of the stored detection rate of the white line candidate points.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traveling lane detecting device for detecting the type of white line in a traveling lane on which a vehicle such as an automobile travels.

[0002]

2. Description of the Related Art It is very important to recognize a traveling lane on which a vehicle is traveling on a traveling path of the vehicle from the viewpoint of accident prevention. JP-A-8-320 discloses a technique for recognizing such a driving lane by detecting the type of white line.
No. 997 is provided.

In this invention, the maximum value of the edge point count value of each window is changed periodically or constantly based on the traveling image data in front of the own vehicle imaged by the imaging means. Is determined to be a constant value, and if it changes periodically, the white line is recognized as a dashed lane marker, and if the value is constantly constant, it is recognized as a solid lane marker. Based on the lane marker recognition result, the classification of the traveling lane of the own vehicle and the adjacent part is recognized, and whether the own lane is the traveling lane at the left end, the overtaking lane at the center, the overtaking lane at the right end, and the like. recognize.

According to this invention, since the window is set at the lower end of the image corresponding to the part immediately before the host vehicle, the lane marker can be detected as a substantially straight line, and there is no possibility that the lane marker is blocked by a preceding vehicle or the like.

[0005]

Problems to be Solved by the Invention
In the invention of No. 7, the white line is determined to be a dashed line when the maximum value of the edge point count value appearing in the traveling road image changes periodically, and the white line is determined to be a solid line when constantly constant, Since the white line is uniquely determined to be a broken line and a solid line depending on the appearance of the edge point count value, the recognition accuracy of the type of the white line is low. In particular, in the case of the broken line, the length and pitch of the white line cannot be detected. For this reason, it cannot be determined whether the dashed line is a dashed line on an expressway or a dashed line on a general road, and erroneous recognition of the traveling lane due to erroneous recognition of the type of white line during traveling is likely to occur.

The present invention has been made in view of the problems of the prior art,
An object of the present invention is to make it possible to detect the type of a white line with high accuracy in spite of a short detection range of the white line, and to prevent erroneous recognition of a traveling lane during traveling.

[0007]

According to the present invention, in order to solve the above-mentioned problems, according to the first aspect of the present invention, a state of a traveling road is imaged by an image pickup means provided at a front portion of a vehicle, and the state of the traveling road is picked up by the image pickup means. A vehicle lane detecting device for detecting the type of a white line on the traveling road based on the obtained image, wherein the white line candidate point detecting unit detects the white line candidate point by processing the captured image. From the white line candidate points, only the white line candidate points recognized as white lines are selected, the white line candidate point detection rate data holding unit from the selected white line candidate points, and the stored white line candidate point detection rates are stored. The system includes a white line candidate data holding unit that stores a predetermined amount of past white lines, and a white line type determination unit that determines the type of white line based on the stored white line candidate point detection rate.

According to a second aspect of the present invention, the white line type determination unit determines the white line as a solid line when the average value of the detection rates stored in the white line candidate data holding unit is equal to or larger than a predetermined reference value. When the average value is less than the reference value and the detection rates of the respective data are present in an area larger than the upper determination value of the reference value and an area smaller than a predetermined lower determination value, the white line is determined as a broken line. It is constituted as follows.

According to this invention, when the average value of the detection rates stored in the white line candidate data holding unit is equal to or more than a predetermined reference value, that is, approximately 70% or more, the white line is determined to be a solid line. The value is less than the reference value (70%), the peak of each detection rate protrudes from the upper fixed line (upper judgment value), and the bottom is lower than the lower fixed line (lower judgment value). In the case of the detection rate form protruding downward, it is determined that the white line is a broken line.

Therefore, according to the present invention, the detection rates of the white line candidate points are stored in a fixed amount in the past, and the average, maximum, and minimum values of the detection rates are obtained and compared with the reference value, whereby the traveling lane is detected. Is determined as a solid line or a broken line, the solid line and the broken line can be distinguished with high accuracy only by storing a small amount of white line detection rate data.

According to a third aspect of the present invention, in addition to the second aspect, the white line type discriminating unit judges the white line as a broken line, and then determines the peak of the detection rate stored in the white line candidate data holding unit. The distance is calculated, and the type of the traveling lane can be detected from the distance.

According to this invention, even after it is determined that the white line is a broken line, the peak-to-peak distance of the curve connecting the individual detection rates is calculated, and this distance is made to correspond to the white line of an expressway or a general road. Thus, the driver can correctly recognize whether the current driving lane is a highway or a general road.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to an embodiment shown in the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not merely intended to limit the scope of the present invention, but are merely illustrative examples unless otherwise specified. Absent.

FIG. 1 is a control block diagram of an automobile driving lane detecting apparatus according to an embodiment of the present invention, FIG. 2 is a control flowchart, FIG. 3 is an overall configuration diagram, FIG. 4 is a white line type discrimination diagram, and FIG. FIG. 6 is an explanatory diagram of the type of broken line.

In FIG. 3, which shows the entire configuration of the vehicle traveling lane detecting device, 10 is a vehicle (hereinafter referred to as a vehicle), 1 is a video camera (imaging means) mounted in front of the vehicle 10, 2 is a controller, and 3 is a controller. The controller 2 and the vehicle speed detector 3 detect a vehicle speed of the vehicle 10.
Is mounted on the car 10.

In FIG. 1 and FIG.
The image in front of the car 10 taken at is input to the amplifier 21 of the controller 2 and is amplified to a predetermined size.
After being digitized by the D conversion unit 22, it is input to the white line candidate point detection unit 23 (step (1)).

The white line candidate point detecting section 23 processes the input image and detects data as white line candidate points by removing noise. (Step (2)). The white line candidate point detection data is input to the white line candidate point selection unit 24.

The white line candidate point selection unit 24 calculates a white line detection rate based on the white line candidate point detection data. The detection rate: f is given by the following equation (1). f = (number of detected points / number of lines in detection range) × 100 (%) (1) Then, the detection rate data is stored in the white line candidate data holding unit 25.
(Step (3)).

In the white line candidate data holding section 25, the processing time (for example, 0.1 second) and the vehicle speed detection data (vehicle speed for each processing time) of the vehicle 10 inputted from the vehicle speed detector 3 are used for each cycle. Distance: L is calculated by the following equation (2), and this L is stored in a fixed amount in the past (for example, for the past 20). L = (vehicle speed) × (processing time = 2 seconds) (2)

Then, the white line candidate data holding unit 25 simultaneously stores the detection rates of the left and right white line candidate points by a fixed amount in the past (step (4)).

The white line type determination unit 26 calculates the average value f _m from detection rate f of the stocked in the white line candidate data holding unit 25 white line. Then the white line type determination unit 26 is the average value: f _m is less than the reference value f _0, and smaller than the larger area and a lower determination value f ₀₂ than the detection rate f the upper determination value f ₀₁ of each data area , It is determined that the white line is a broken line (step (5)).

[0022] That is, as shown in FIG. 4 (A), the average value _{f m} of stocked detection rate f is the reference value _f 0 = 70% or more (not necessarily limited to 70%, past detection rate (Determined by data), the white line type determination unit 26 determines that the white line is a solid line. Further, as shown in FIG. 4 (B), an average value f _m is the reference value f _{0 =} less than 70% of the stocked detection rate f, f detection rate f of the data is the upper determination value _When there is an area in which ₀₁ = 80% or more and an area in which the lower determination value f ₀₂ = 20% or less exists, it is determined that the white line is a broken line.

Accordingly, the white line type discriminating section 26 obtains the average, maximum and minimum values of the past fixed amount of the white line detection rate, and compares the average, maximum and minimum values of the detection rate with the reference value. It is determined whether the white line of the traveling lane is a solid line or a broken line, so that the solid line and the broken line can be correctly determined even if the number of stored white line detection data is small.

That is, by determining whether the white line is a solid line as described above, when the left side is a solid line and the right side is a broken line, the vehicle is traveling in the left lane. When the middle and left sides are broken lines and the right side is solid lines, the traveling lane of the own vehicle can be easily recognized from the image captured by the video camera 1 such that the vehicle is traveling in the right lane.

Furthermore in the white line type determining unit 26, when the white line is determined to be a broken line, the distance L ₀ between the peaks of the connected curve detection rate f of the white line stock number (20), wherein (2) calculated calculated based on the distance L, which is stocked in the white line candidate data holding unit 25 by the formula, broken lines of the highway, as shown in FIG. 6 (a) if the distance L ₀ is 20m determined that, L ₀ is judged to dashed ordinary road if 15 m.

That is, as shown in FIG.
The distance L ₀ between 0 peaks individual tethered detection rate f curve can be calculated from the distance L corresponding to the processing time calculated by the image and the (2) equation. And since pitch this peak appears (i.e. the distance between the peaks) is equal to the pitch L ₀ dashed white lines as shown in FIG. 6, the distance between peaks L ₀ of said calculated dashed image, which is 20
m, it can be determined that the road is a general road if it is 15 m.

Therefore, according to this embodiment, the white line type determination section 26 determines the white line as a solid line or a broken line as described above, and further calculates the peak-to-peak distance of the detection rate for the broken line. Since it is determined whether the white line is an expressway or a general road based on the distance, it is possible to easily recognize whether the current traveling lane is an expressway or a general road.

[0028]

As described above, according to the present invention, a white line candidate point detecting section for processing a captured image to detect white line candidate points, and a white line candidate point recognized as a white line among the white line candidate points A white line candidate point selection unit that selects only the white line candidate points from the selected white line candidate points, and a white line candidate data holding unit that stores the white line candidate point detection ratio in a past fixed amount. Since the type of the white line can be determined based on the stored detection rate of the white line candidate points, the type of the white line can be detected with high accuracy despite the short detection range of the white line. This is to prevent erroneous recognition.

According to the second aspect of the present invention, the detection rates of the white line candidate points are stored in a fixed amount in the past, and the average value, the maximum value, and the minimum value of the detection rates are obtained and compared with the reference value. Since it is determined whether the traveling lane is a solid line, the solid line and the broken line can be distinguished with high accuracy only by storing a small amount of white line detection rate data.

According to the third aspect of the present invention, even after it is determined that the white line is a broken line, the peak-to-peak distance of the curve connecting the individual detection rates is calculated, and this distance is calculated on an expressway or a general road. By corresponding to the white line of the road,
The driver can correctly recognize whether the current lane is a highway or a general road.

[Brief description of the drawings]

FIG. 1 is a control block diagram of an automobile driving lane detecting device according to an embodiment of the present invention.

FIG. 2 is a control flowchart in the embodiment.

FIG. 3 is an overall configuration diagram in the embodiment.

FIG. 4 is a diagram for discriminating white line types in the embodiment.

FIG. 5 is a determination diagram of a broken line type in the embodiment.

FIG. 6 is an explanatory diagram of broken line types in a traveling lane.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Video camera 2 Controller 23 White line candidate point detection part 24 White line candidate point selection part 25 White line candidate data holding part 26 White line type discrimination part 3 Vehicle speed detector 10 vehicles

Claims (3)

[Claims] 1. A vehicle traveling system in which a state of a traveling road is imaged by an imaging unit installed at a front part of a vehicle, and a type of a white line on the traveling road is detected based on an image obtained from the imaging unit. In the lane detection device, a white line candidate point detection unit that processes an image captured from the imaging unit to detect a white line candidate point, and selects only white line candidate points recognized as white lines among the white line candidate points, A white line candidate point selecting unit for calculating a detection rate of white line candidate points from the selected white line candidate points; a white line candidate data holding unit for storing the white line candidate point detection ratio in a past fixed amount; and the stored white line candidate points And a white line type determination unit that determines the type of white line based on the detection rate of the vehicle. 2. The white line type determination unit determines that the white line is a solid line when the average value of the detection rates stored in the white line candidate point data holding unit is equal to or greater than a predetermined reference value, and the average value is equal to the reference value. The white line is determined to be a dashed line when the detection rate of each data is less than the upper determination value of the reference value and in the area smaller than the predetermined lower determination value. The vehicle traveling lane detecting device according to claim 1. 3. The white line type determination unit calculates the distance between peaks of the detection rate stored in the white line candidate point data holding unit after determining the white line as a broken line, and determines the type of the traveling lane from this distance. The traveling lane detecting device for a vehicle according to claim 2, wherein the device is configured to be capable of detecting.