JP2008225953A - White line recognition device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a white line recognition device used in a lane keeping assist system, or the like, to accurately recognize a white line, even when the width of a white line is different from the normal width. <P>SOLUTION: The while line recognition device comprises a photographing means (110) for photographing an image of a traveling road where a vehicle travels, an image processing means (120) for recognizing a white line by the comparison with a threshold set to correspond to the width of a white line on the traveling road based on the photographed image, an information acquiring means (130) for acquiring information of a predetermined kind regarding the traveling road, and a threshold value changing means (150) for changing the set threshold according to the content of the acquired information. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technical field of a white line recognition apparatus and method for recognizing a white line on a traveling road on which a vehicle travels based on an image captured by, for example, a camera.

  As this type of white line recognition device, there is a device that captures a travel road with a camera or the like mounted on a vehicle, and then analyzes the captured image to recognize a white line on the travel road, thereby supporting the travel of the vehicle. In such an apparatus, various techniques have been disclosed for accurately recognizing a white line with high accuracy.

  For example, there is a technique of estimating a parameter relating to a traveling road based on the position of a white line in a captured image (see Patent Document 1). There is also a technique of using a plurality of thresholds for recognizing white lines (see Patent Documents 2 and 3). Furthermore, there is a technique of determining whether or not the image is a white line after binarizing the image (see Patent Document 4).

  On the other hand, a technique of obtaining information about a lane using a means other than the imaging means is also disclosed. For example, there is a technique of obtaining information on the lane width of a travel path from a navigation system (see Patent Documents 5 and 6).

JP 2005-329765 A JP 2003-168123 A JP 2001-92970 A JP 2000-235699 A JP 2003-15490 A JP-A-4-152406

  However, in the white line recognition device as described above, for example, in a place where a special white line is locally used such as a branch point or a merge point, the width of the white line is significantly different from the normal, There is a technical problem that white lines may not be recognized accurately.

  The present invention has been made in view of the above-described problems, for example, and an object of the present invention is to provide a white line recognition apparatus and method that can accurately recognize a white line even in a place where the width of the white line is different from a normal width.

  In order to solve the above problems, the white line recognition device of the present invention corresponds to the width of the white line on the travel road based on the image pickup means for picking up the image of the travel road on which the vehicle travels and the captured image. The image processing means for recognizing the white line by comparison with a threshold value set as: an information acquisition means for acquiring a predetermined type of information about the travel route; and the setting according to the content of the acquired information Threshold value changing means for changing the threshold value.

  According to the white line recognition device of the present invention, during the operation, first, the image of the traveling road is taken by the imaging means. The imaging means is typically mounted on the vehicle as part of a white line recognition device mounted on the vehicle, and images the front scenery of the vehicle.

  Then, the white line is recognized by the image processing means by comparison with a threshold value set as corresponding to the width of the white line on the travel path based on the captured image. Recognition of the white line is performed by comparing a point where the color density in the direction intersecting the white line in the image changes abruptly or a point where the differential value is steep, with a threshold value. For the white line recognition here, various existing recognition techniques can be employed as long as the set threshold value is used.

  The “white line” according to the present invention refers to a line or pattern that divides a lane and is drawn on the road, and is not limited to a white line. That is, a line with a different color, such as a yellow line, may be used. For example, it may be a pattern that does not form a simple line, such as that seen at a branch point or a junction. The white line may have any color, pattern, or shape as long as the width can be recognized.

  In parallel with or in parallel with the recognition of such a white line, the information acquisition means typically acquires a predetermined type of information related to the travel path from other than the imaging means. Here, the “predetermined type of information relating to the travel path” may be the width of the white line associated with the road position itself, or the white line width may be locally special, such as a road branching point or a junction point. The information may indirectly indicate that the width is changed. The information acquisition means is typically mounted on the vehicle as part of the white line recognition device, but may be at least partially disposed outside the vehicle via the communication means, or via the communication means. Information may be acquired from outside the vehicle.

  Further, the threshold for recognizing the white line used in the image processing means is changed according to the content of the information acquired in this way. For example, the white line can be recognized more quickly and accurately according to the content of the information that indicates the width of the line directly, such as 10 cm white, or shows the width of the white line in stages, such as thick, standard, thin or special. The threshold value is changed. As a result, the white line can be recognized quickly and accurately even in a place where the width of the white line changes.

  However, the “threshold change” according to the present invention may be a change unrelated to the information acquired by the information acquisition means. Even in this case, it is considered that it is difficult or impossible to recognize the white line at least using the normal threshold value, and it is correspondingly useful to reduce the probability of causing recognition failure by increasing or decreasing the threshold value. .

  If the information acquisition means as described above is not provided, the threshold value for recognizing the white line cannot be changed according to the condition of the road. For this reason, when the width of the white line is different from normal, there is a possibility that a white line that does not exist is erroneously detected, or conversely, an existing white line cannot be detected.

  However, in the present invention, in particular, since the information acquisition unit acquires information on the travel path from other than the imaging unit, it is possible to acquire, for example, information on whether or not a white line having a special width exists around the travel path. That is, information that cannot be obtained from the imaging range of the imaging unit, information that cannot be obtained from the image, and the like can be acquired. The information acquisition unit may be configured to acquire information using a plurality of methods as long as it is a method other than the imaging unit.

  As described above, in the white line recognition apparatus according to the present invention, the information acquisition means acquires information that cannot be obtained from other than the imaging means, and therefore the threshold value for recognizing the white line is set to a value corresponding to the condition of the traveling road. It is possible to change. Therefore, the white line can be recognized with higher accuracy even in a place where the width of the white line is different from the usual.

  In one aspect of the white line recognition apparatus of the present invention, the white line recognition device further includes a determination unit that determines whether the white line has a predetermined width based on the acquired information, and the threshold value changing unit includes: When the white line width is determined to be different from the predetermined width, the set threshold value is changed.

  According to this aspect, the determination unit determines whether or not the width of the white line on the travel path is a predetermined width. The “predetermined width” does not indicate a single value, but a value representing a range having a certain width. Therefore, in the determination, it is typically determined whether or not the width of the white line is within a predetermined width range. When it is determined that the width of the white line is different from the predetermined width, the threshold for recognizing the white line used in the image processing unit is changed.

  Thus, by determining the width of the white line based on the acquired information, it is possible to more easily determine whether or not to change the threshold value. Accordingly, it is possible to prevent the apparatus configuration from becoming complicated, and as a result, it is possible to realize downsizing of the apparatus.

  In another aspect of the white line recognition apparatus of the present invention, the determination unit determines whether a branch point or a merge point exists on the travel path based on the acquired information, and the threshold value changing unit includes: The threshold value for recognizing the white line on the travel path is changed so that the white line is easily recognized when it is determined by the determination means that a branch point or a merge point exists on the travel path.

  According to this aspect, it is determined by the determination means whether or not a branch point or a merging point exists on the travel path. At this time, if it is determined that there is a branch point or a merging point on the travel path, the threshold value is increased by the threshold value changing means.

  Unusual white lines are often used at junctions and junctions on the roadway. For example, especially on an expressway, the width of the white line may become thicker when approaching a junction or junction. If an attempt is made to recognize a white line using a normal threshold at such a point, the white line may have a different width from the normal width, and the white line may not be recognized accurately.

  However, in the present invention, in particular, based on the information acquired by the information acquisition unit, the determination unit determines whether a branch point or a merge point exists, and the threshold value is changed so that a white line is easily recognized based on the result. To do. For example, when the white line is thicker than usual, the threshold value is increased. Therefore, even if the width of the white line is different from the normal width at the branch point or the merge point, the white line can be recognized with high accuracy by the changed threshold value.

  Moreover, since the determination means only has to determine whether or not there is a branch point or a merging point on the travel path, the determination process is relatively simple compared to the case of handling various information on the travel path. The configuration can also be simplified. Further, the information acquired by the information acquisition means may be relatively small. For example, if there is map data, the determination means can make the determination. At this time, if information directly indicating the presence of a branch point or a merge point can be acquired, the determination by the determination means is simple. Or if the information which shows the connection state of a road is acquirable, the determination in a determination means will only have to determine a branch and merge from a connection state.

  As described above, in the white line recognition apparatus according to the present invention, by using the characteristic that the width of the white line is different from the normal at the branch point or the junction, the configuration of the determination unit is simplified and the information acquisition unit is used. It is possible to recognize the white line with higher accuracy when a white line having a width different from normal is used while reducing the amount of information to be acquired.

  In another aspect of the white line recognition apparatus of the present invention, the information acquisition means includes a navigation system.

  According to this aspect, acquisition of information by the information acquisition unit is performed from the navigation system mounted on the vehicle. For this reason, for example, it is possible to acquire map data around the vehicle together with the current position of the vehicle. Therefore, it is possible to predict the width of the white line around the traveling path of the vehicle, and change the threshold value for recognizing the white line accordingly. Therefore, even when the width of the white line on the travel path is different from the normal width, the white line can be recognized with high accuracy.

  In a typical navigation system, map data and various additional information are recorded on a DVD, for example. If the information about the width of the white line is recorded in detail here, the white line can be recognized with higher accuracy, but the width of the white line is predicted to some extent even if the information is not directly related to the width of the white line. If it is information that can be made possible, the accuracy of recognizing a white line can be improved.

  In another aspect of the white line recognition apparatus of the present invention, the information acquisition means includes a communication system.

  According to this aspect, acquisition of information by the information acquisition unit is performed by communicating with the outside of the vehicle by the communication system. For this reason, it is possible to acquire the information regarding the width of a white line, or the information which can predict the width of a white line, for example by communicating with the database in places other than a vehicle. Therefore, the threshold value for recognizing the white line can be changed based on the information. Therefore, even when the width of the white line on the travel path is different from the normal width, the white line can be recognized with high accuracy.

  By using the communication system, the information acquisition means can acquire information from outside the vehicle. For this reason, information such as map data may not be held in advance by the vehicle itself. Further, when an external database or the like is used, more information can be used as compared with the case where the vehicle itself holds information. Therefore, the width of the white line can be predicted more accurately, and the threshold value can be changed to a more appropriate value. Therefore, it is possible to recognize the white line with higher accuracy.

  Such a communication system may be a communication system included in in-vehicle communication navigation that acquires navigation map data or the like via communication means.

  In order to solve the above-described problem, the white line recognition method of the present invention corresponds to the width of the white line on the travel road based on the imaging step of capturing an image of the travel road on which the vehicle travels and the captured image. The image processing step for recognizing the white line by comparison with a threshold value set as: an information acquisition step for acquiring a predetermined type of information related to the travel route; and the setting according to the content of the acquired information And a threshold value changing step for changing the threshold value.

  According to the white line recognition method of the present invention, information that is typically not obtained in the imaging means step is acquired by the information acquisition step, so that the determination by the determination step is performed more accurately. Therefore, the threshold value for recognizing the white line can be changed to a value corresponding to the situation of the traveling road. Therefore, the white line can be recognized with higher accuracy even in a place where the width of the white line is different from the usual.

  In the white line recognition method according to the present invention, various aspects similar to the various aspects of the above-described white line recognition apparatus of the present invention can be employed.

  The operation and other advantages of the present invention will become apparent from the best mode for carrying out the invention described below.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Here, as an example of the white line recognition device of the present invention, a white line recognition device used in a lane keeping assistance (LKA) system will be described.

<First Embodiment>
First, the configuration of the white line recognition device according to the present embodiment will be described with reference to FIGS. 1 and 2. FIG. 1 is a block diagram showing the configuration of the white line recognition device, and FIG. 2 is a side view showing an installation example of the white line recognition device on the vehicle. In FIG. 2, the image processing unit 120, the determination unit 140, and the threshold value changing unit 150 in FIG. 1 are collectively illustrated as a white line recognition unit 400.

  As shown in FIG. 1, the white line recognition apparatus 100 according to the present embodiment includes an imaging unit 110 as an example of the “imaging unit” of the present invention and an image processing unit 120 as an example of the “image processing unit” of the present invention. The information acquisition unit 130 as an example of the “information acquisition unit” of the present invention, the determination unit 140 as an example of the “determination unit” of the present invention, and the threshold change as an example of the “threshold change unit” of the present invention Part 150. In addition, the white line recognition apparatus 100 according to the present embodiment is connected to the LKA system 400 via the image processing unit 120.

  As illustrated in FIG. 2, the imaging unit 110 is, for example, a camera, and is disposed at a position where the traveling path on which the vehicle travels can be imaged.

  The information acquisition unit 130 is a navigation system, for example, and can acquire information related to the travel route. Moreover, the communication system which acquires information by communicating with the exterior may be sufficient.

  The white line recognition unit 300 (that is, the image processing unit 120, the determination unit 140, and the threshold value changing unit 150 in FIG. 1) includes, for example, an arithmetic circuit and a memory, and performs arithmetic processing on the input data. The result can be output.

  Next, the operation of the white line recognition apparatus according to the present embodiment will be described with reference to FIGS. 3 and 4 in addition to FIGS. FIG. 3 is a flowchart showing the operation steps of the white line recognition apparatus according to the first embodiment. FIG. 4 is a conceptual diagram showing a white line candidate detection method.

  As shown in FIG. 3, when the white line recognition apparatus 100 according to the present embodiment operates, first, an image of a traveling road on which the vehicle travels is captured by the imaging unit 110 (step S1). At this time, the image of the traveling road to be imaged is typically an image in front of the vehicle 500, but may be an image of the rear of the vehicle 500 or the like.

  The captured image is analyzed by the image processing unit 120, and white line candidates (objects that are likely to be white lines) in the image are detected (step S2). For example, white line candidates can be detected by the luminance in the image. An example of the detection method will be described in detail below.

  Assume that white lines 201 and 202 appear in the captured image as shown in FIG. First, the luminance on a straight line extending in the horizontal direction (a chain line in FIG. 4A) is measured at a predetermined height in the image. Since roads are typically close to black, the brightness is low. On the other hand, the white line portion is estimated to have high luminance. Therefore, a portion having higher brightness than the surroundings can be detected as a white line candidate.

  The luminance measured on the chain line in FIG. 4A can be represented by a graph as shown in FIG. Further, the measured luminance can be differentiated to obtain a luminance change amount, which can be represented by a graph as shown in FIG. When using FIG.4 (b), the part with high brightness | luminance can be detected as a white line candidate. On the other hand, when using FIG.4 (c), between the rising and falling edges of a graph can be detected as a white line candidate. As described above, it is possible to detect white line candidates using the captured image of the traveling road. The detected white line candidates are measured for their respective widths W1 and W2.

  On the other hand, in parallel with the above-described imaging of the travel road image and the processing of the captured image, the information acquisition unit 130 acquires information about the travel road (step S3). For example, the navigation system can be used to acquire information related to the white line on the surrounding traveling road along with the current position of the vehicle 500. Here, the information on the white line may be the width of the white line associated with the road position itself, or may be information indirectly indicating that the width of the white line locally changes to a special width. Good.

  Next, the determination unit 140 determines whether or not the white line on the traveling road has a predetermined width based on the acquired information on the traveling road (step S4). Note that the value of the predetermined width is typically set based on the width of a white line that is relatively used on a normal road. When it is determined that the white line on the travel path is not a predetermined width (step S4: NO), the process proceeds to step S5. When it is determined that the white line on the travel path has a predetermined width (step S4: YES), the process of step S5 is omitted, and the process proceeds to step S6.

  If the white line on the road is not a predetermined width, the threshold value is changed by the threshold value changing unit 150 because the white line may not be properly recognized with a preset threshold value (step S5). For example, upper limit values WU and WL are set as initial values for the threshold. This initial value corresponds to the case where the white line has a predetermined width. That is, when the white line has a predetermined width, it does not need to be changed. When changing the threshold value, for example, ΔW1 is added to the upper limit value WU to obtain WU ′ (ie, WU ′ = WU + ΔW1). On the other hand, ΔW2 is subtracted from the lower limit value WL to become WL ′ (that is, WL ′ = WL−ΔW2). Here, ΔW1 and ΔW2 may be, for example, experimentally or theoretically derived values, or may be derived by a difference between a predetermined width and the width of the white line obtained from the travel route information.

  Next, the widths W1 and W2 of the white line candidates detected in step S2 and the threshold values WU 'and WL' (however, if the threshold value has not been changed in step S5, WU and WL. Hereinafter, only when it has been changed) Are respectively compared with each other and a determination is made as to whether or not to recognize them as white lines (step S6). Here, if WL ′ ≦ W1 ≦ WU ′ (step S6: YES), the white line candidate having the width W1 is recognized as a white line (step S7). Similarly, if WL ′ ≦ W2 ≦ WU ′, a white line candidate having a width W2 is recognized as a white line. When W1 <WL ′ or WU ′ <W1 (step S7: NO), the white line candidate having the width W1 is not recognized as a white line. Similarly, when W2 <WL ′ or WU ′ <W2, a white line candidate having a width W2 is not recognized as a white line.

  As described above, the white line is recognized based on the captured image and the information on the travel path. Such a series of processes may be continuously repeated or may be performed at predetermined intervals.

  When a white line is recognized in this way, a lane (that is, a portion sandwiched between white lines) can be recognized based on the white line. For this reason, by outputting the white line information recognized by the white line recognition device 100 to the LKA system 400 (see FIGS. 1 and 2), for example, the power steering of the vehicle 500 is controlled, and the vehicle 500 is out of the lane. It is possible to support the driver's driving by driving without any trouble.

Second Embodiment
Next, a white line recognition apparatus according to the second embodiment will be described with reference to FIGS. 5 and 6. FIG. 5 is a flowchart showing the operation steps of the white line recognition apparatus according to the second embodiment, and FIG. 6 is a plan view showing a special white line at the branch point. The second embodiment differs from the first embodiment described above in the determination process in the determination unit 140, and the other processes and the configuration of the apparatus are the same. Therefore, in the second embodiment, the operation of the determination unit 140 will be mainly described, and description of other configurations will be omitted as appropriate.

  As shown in FIG. 5, in the operation process of the white line recognition apparatus according to the second embodiment, steps S11 to S13 and S15 to S17 correspond to S1 to S3 and S5 to S7 in the first embodiment, respectively. The same operation is performed.

  However, in the determination by the determination unit 140 based on the information on the travel route acquired from the information acquisition unit 130, it is determined whether or not a branch point or a junction exists around the travel route (step S14). Then, when it is determined that there is a branch point or a merging point in the vicinity of the travel path (step S14: YES), the process proceeds to step S15 and the threshold value is changed. When it is determined that there is no branch point or junction point around the travel path (step S14: NO), the process of step S15 is omitted, and the process proceeds to step S16. That is, the threshold value is not changed.

  As shown in FIG. 6, there is a high possibility that a special white line 300 different from a normal white line is provided at a branch point and a merge point. Such a special white line 300 is wider than a normal white line, and there is a possibility that accurate recognition may not be performed with a threshold for recognizing a normal white line. For this reason, it is possible to recognize the special white line 300 as a white line by changing the threshold when there is a branch point or a merging point around the road.

  The special white line 300 has various shapes. For example, the shape differs depending on the type of the branch point and the merge point (that is, the shape of the branch point and the merge point itself), or whether the road is a general road or a highway. Therefore, the white line can be recognized more accurately by making the threshold value changing method in step S15 correspond to these conditions.

  In the second embodiment, the determination unit 140 only needs to determine whether there is a branch point or a merging point in the vicinity of the traveling road. Therefore, whether the width of the white line in the first embodiment is a predetermined width or not. Compared with the case of determining whether or not, less information on the travel route is required. That is, even when it is difficult to acquire information other than the information recorded on the DVD being used as in a typical navigation system, information on whether or not a branch point and a junction point exist. Can be obtained reliably. Therefore, it is possible to prevent a situation in which there is a shortage of information regarding the travel route used by the determination unit 140 and an accurate determination cannot be made.

  As described above, in the white line recognition apparatus according to the second embodiment, the determination in the determination unit 140 is determined as a determination as to whether or not there is a branching point or a merging point around the traveling road. A beneficial effect is obtained.

  The present invention is not limited to the above-described embodiment, and can be appropriately changed without departing from the gist or concept of the invention that can be read from the claims and the entire specification. The method is also included in the technical scope of the present invention.

The block diagram which shows the structure of a white line recognition apparatus. It is a side view which shows the example of installation to the vehicle of a white line recognition apparatus. It is a flowchart which shows the process of the operation | movement of the white line recognition apparatus based on 1st Embodiment. It is a conceptual diagram which shows the detection method of a white line candidate. It is a flowchart which shows the process of the operation | movement of the white line recognition apparatus based on 2nd Embodiment. It is a top view which shows the special white line in a branch point.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... White line recognition apparatus, 110 ... Imaging part, 120 ... Image processing part, 130 ... Information acquisition part, 140 ... Determination part, 150 ... Threshold change part, 201, 202 ... White line, 250 ... Special white line, 300 ... White line recognition part , 400 ... LKA system, 500 ... vehicle

Claims (6)

Imaging means for capturing an image of a travel path on which the vehicle travels;
Image processing means for recognizing the white line based on a comparison with a threshold value set as corresponding to the width of the white line on the travel path based on the captured image;
Information acquisition means for acquiring a predetermined type of information relating to the travel path;
A white line recognizing device comprising: threshold changing means for changing the set threshold according to the content of the acquired information. A determination means for determining whether the white line has a predetermined width based on the acquired information;
2. The white line according to claim 1, wherein the threshold value changing unit changes the set threshold value when the determination unit determines that the width of the white line is different from the predetermined width. Recognition device. The determination means determines whether or not a branch point or a merging point exists on the travel path based on the acquired information,
The threshold value changing means changes the set threshold value so that the white line is easily recognized when it is determined by the determining means that a branch point or a merging point exists on the travel path. The white line recognition device according to claim 2.   The white line recognition apparatus according to any one of claims 1 to 3, wherein the information acquisition unit includes a navigation system or acquires the information via the navigation system.   5. The white line recognition apparatus according to claim 1, wherein the information acquisition unit includes a communication system or acquires the information through the communication system. 6. An imaging process for capturing an image of a travel path on which the vehicle travels;
An image processing step for recognizing the white line based on a comparison with a threshold value set as corresponding to the width of the white line on the travel path based on the captured image;
An information acquisition step of acquiring a predetermined type of information about the travel path;
A white line recognition method comprising: a threshold value changing step of changing the set threshold value according to the content of the acquired information.

Images