JP6293570B2 - Lane mark recognition apparatus and method, and vehicle - Google Patents

Description

  The present invention relates to a lane mark recognition device for recognizing a lane mark provided on a road based on a captured image of a camera mounted on a vehicle.

  2. Description of the Related Art Conventionally, a lane mark recognition device that recognizes a lane mark provided on a road from an image of a road ahead of the vehicle captured by an in-vehicle camera is known.

  In addition to the lane mark, a travel guidance obstacle detection device that recognizes a travel guidance obstacle disposed on or around the travel lane of the host vehicle such as a lane marking and a pylon has been proposed (for example, Patent Document 1). reference).

  The travel-guided obstacle detection device described in Patent Document 1 is such that the size of a three-dimensional object detected from an image is within a preset threshold and the three-dimensional object detected in the image is When the luminance change in the direction has a certain pattern, the three-dimensional object is detected as a traveling guidance obstacle.

  According to the traveling guidance obstacle detection device described in Patent Document 1, sorting is performed by size based on the width and height of a three-dimensional object detected in an image, and further sorting is performed by a repeating pattern of luminance. Thus, it is possible to accurately detect a traveling guidance obstacle such as a lane separator or a pylon from among three-dimensional objects existing in front of the host vehicle.

Japanese Patent No. 4901275

  In the conventional lane mark recognition device, the driving guidance obstacle detected by the technique of Patent Document 1 is excluded from the processing target as an image part that is not a candidate for the lane mark, and by recognizing the lane mark from the remaining image part, It is conceivable to improve the processing speed and the detection accuracy of the lane mark.

  However, in the detection of the travel-guided obstacle in the technique of Patent Document 1, it is necessary to collate the detected object with information such as a size and a luminance change pattern regarding a plurality of types of obstacles. For this reason, if a traveling guidance object is detected by the technique of Patent Document 1 and an attempt is made to exclude it as an image part that is not a candidate for a lane mark, the calculation cost may become enormous.

  The present invention has been made in view of such a background, and an object of the present invention is to provide a lane mark recognition apparatus and the like that can improve the detection accuracy of lane marks while reducing the calculation cost in lane mark recognition. .

  The present invention has been made to achieve the above object, and a lane mark recognition device for recognizing a lane mark provided on a road based on an image around the vehicle imaged by a camera mounted on the vehicle, The present invention relates to a vehicle and a lane mark recognition method.

And the lane mark recognition device of the present invention,
An edge image generating unit that generates an edge image obtained by extracting an edge point having a luminance change amount equal to or greater than a predetermined value from a captured image of the camera;
Among the edge points, a candidate image part extraction unit that extracts candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part;
Of the candidate image portions extracted by the candidate image portion extraction unit, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or greater than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion unit that excludes the candidate image portion that is more than a predetermined distance away from the camera ;
And a lane mark recognizing unit that recognizes a lane mark based on a candidate image portion that has not been excluded by the specific candidate excluding unit.

  According to studies by the present inventors, lane marks are often imaged as diagonal lines on a captured image. Further, since the lane mark is laid on the road, the length in the vertical direction in the real space is almost equal to zero if the road is flat. Therefore, it is possible to determine with high accuracy whether or not the candidate image portion represents a lane mark based on the length in a specific direction corresponding to the vertical direction in the real space corresponding to the candidate image portion.

According to the lane mark recognizing device of the present embodiment configured focusing on this point, the lane that satisfies the condition that the length in the specific direction corresponding to the vertical direction in the real space corresponding to the candidate image portion is equal to or greater than the predetermined length. Candidate image portions that are not likely to be image portions of marks are excluded. For this reason, compared with the case where collation with information, such as a size and a brightness change pattern regarding a detected object and a plurality of types of obstacles, is required, calculation cost is reduced.
Furthermore, according to the lane mark recognizing device having the above configuration, the exclusion process is performed on a long-distance candidate image portion that does not directly affect the control of the vehicle. As a result, the calculation cost of the entire apparatus is reduced while maintaining the lane mark detection accuracy.

In the lane mark recognition apparatus of the present invention,
The predetermined length deviates from the length in the specific direction of the image portion of the lane mark in the captured image when it is assumed that there is a lane mark at a position in the real space corresponding to the vehicle and the candidate image portion. The length is preferred.

  According to the lane mark recognizing apparatus having the above configuration, candidate image portions having lengths in a specific direction that deviate from the lengths in the specific direction of the image portion of the assumed lane mark on the captured image are excluded. As a result, according to the lane mark recognizing device having the above configuration, the detection accuracy of the lane mark can be improved while the calculation cost is reduced.

In the lane mark recognition apparatus of the present invention,
The specific candidate excluding unit is preferably configured to exclude the candidate image portion whose position in the specific direction is higher than the vanishing point of the captured image.

  As described above, since the lane mark is laid on the road, in the captured image while running on a flat road, the positions of all the specific directions of the image portion of the lane mark are lower than the vanishing point. There are often. That is, when the position of a part of the candidate image portion in the specific direction on the captured image is higher than the vanishing point of the captured image, the candidate image portion is highly likely not to be the lane mark image portion.

  According to the lane mark recognizing device having the configuration configured by paying attention to this point, it is possible to achieve both reduction in calculation cost and improvement in detection accuracy of lane marks.

In the lane mark recognition apparatus of the present invention,
The specific candidate exclusion unit is preferably configured not to exclude a candidate image portion existing in a central region of the captured image.

  In a situation where the vehicle straddles the lane mark, the candidate image portion of the lane mark can extend vertically in the center of the captured image in the captured image.

  According to the lane mark recognizing device having the configuration configured by paying attention to this point, even if it is a candidate image portion that satisfies a predetermined condition, if it exists in the center of the captured image, The candidate image portion is not excluded.

  As a result, according to the lane mark recognizing apparatus having the configuration, the calculation cost is reduced and the probability of malfunction of the lane mark exclusion process is reduced, so that the detection accuracy of the lane mark can be improved.

In the lane mark recognition device of the configuration,
It is preferable that the specific candidate exclusion unit is configured not to exclude a candidate image portion that exists in the center of the captured image only when the vehicle is changing lanes.

  When the vehicle straddles the lane mark, the vehicle is likely to change lanes. According to the lane mark recognizing device having the configuration configured by paying attention to this point, the calculation cost is reduced and the vehicle exists at the center of the captured image only when the probability that the vehicle straddles the lane mark is high. Therefore, the detection accuracy of the lane mark can be improved.

In the lane mark recognition apparatus of the present invention,
An obstacle recognition unit that recognizes the candidate image portion excluded by the specific candidate exclusion unit as an obstacle;
The vehicle is steered so that the vehicle travels in the lane identified by the lane mark recognized by the lane mark recognition unit and avoids contact between the vehicle and the obstacle recognized by the obstacle recognition unit. It is preferable to include a steering control unit that performs support.

  According to the lane mark recognition apparatus having the above configuration, the calculation result in the lane mark recognition process can be used for obstacle recognition. Then, steering of the vehicle is controlled so as to avoid contact with the obstacle recognized as the vehicle. As a result, the calculation cost of the entire apparatus is reduced.

The vehicle of the present invention
An edge image generating unit that generates an edge image obtained by extracting an edge point having a luminance change amount equal to or greater than a predetermined value from a captured image of the camera;
Among the edge points, a candidate image part extraction unit that extracts candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part;
Of the candidate image portions extracted by the candidate image portion extraction unit, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or greater than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion unit that excludes the candidate image portion that is more than a predetermined distance away from the camera ;
And a lane mark recognizing unit that recognizes a lane mark based on a candidate image portion that has not been excluded by the specific candidate excluding unit.

  According to the vehicle of the present invention, the same operational effects as those of the lane mark recognition apparatus of the present invention described above can be obtained.

The lane mark recognition method of the present invention is
An edge image generation step of generating an edge image obtained by extracting an edge point where the amount of change in luminance with respect to the surrounding portion is a predetermined value or more from the captured image of the camera;
A candidate image part extraction step of extracting candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part of the edge points;
Of the candidate image portions extracted by the candidate image portion extraction step, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or longer than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion step of excluding the candidate image portion that is more than a predetermined distance away from the camera ;
A lane mark recognition step for recognizing a lane mark based on a candidate image portion that has not been excluded by the specific candidate exclusion step.

  By applying the lane mark recognition method of the present invention to a vehicle and implementing it, the same operational effects as those of the lane mark recognition device of the present invention described above can be obtained.

The block diagram of a lane mark recognition apparatus. The flowchart of a lane mark recognition process. The flowchart of a lane mark recognition process. It is a figure explaining a captured image and an edge image, (a) is a figure explaining a captured image, (b) is a figure explaining an edge image. The figure explaining the edge image when not excluding the candidate image of a perpendicular direction. The figure explaining the relationship between recognition of a lane mark and an obstruction, and steering control.

  An embodiment of a lane mark recognition apparatus of the present invention will be described with reference to FIGS.

(Configuration of lane mark recognition device)
Referring to FIG. 1, a lane mark recognition device 10 is mounted on a vehicle 1 (corresponding to a vehicle of the present invention) including a camera 2 (color camera), a speaker 5, a display 6, and a steering mechanism 7. Yes.

  The lane mark recognition device 10 is an electronic unit that includes a CPU, a memory, various interface circuits, and the like (not shown), and a lane mark recognition and steering control program held in the memory is executed by the CPU, thereby taking a captured image. It functions as an acquisition unit 11, an edge image generation unit 12, a candidate image part extraction unit 13, a specific candidate exclusion unit 14, a lane mark recognition unit 15, an obstacle recognition unit 16, and a steering control unit 17. Further, the lane mark recognition apparatus 10 implements the lane mark recognition method of the present invention.

(Lane mark recognition process)
Hereinafter, processing for recognizing a lane mark provided on a road by the lane mark recognition apparatus 10 will be described with reference to the flowcharts shown in FIGS. 2 (a) and 2 (b). The lane mark recognizing device 10 executes processing according to the flowcharts shown in FIGS. 2A and 2B for each predetermined control cycle, and the lane mark (continuous line lane) of the road on which the vehicle 1 is traveling. Mark and broken line lane mark).

  FIG. 2A / STEP 1 is processing by the captured image acquisition unit 11. The captured image acquisition unit 11 inputs a video signal output from the camera 2 in front of the vehicle 1 (corresponding to the periphery of the vehicle of the present invention), and color components (R value, G value, B value) of the video signal. ) To obtain a color captured image 21 having R value, G value, and B value as data of each pixel. The acquired data of the captured image 21 is held in the image memory 20.

  Next, FIG. 2A / STEP2 is processing by the edge image generation unit 12. Note that the processing step by the edge image generation unit 12 corresponds to the edge image generation step in the lane mark recognition method of the present invention.

  The edge image generation unit 12 performs a process of converting the color component of each pixel of the captured image 21 into luminance, and generates a grayscale image (multi-valued image). Then, the edge image generation unit 12 is a pixel in which the luminance difference (luminance change amount) from the edge point (the surrounding pixel (image portion)) is a predetermined value or more from the grayscale image. The luminance changes from dark to bright. Edge image 22 (see FIG. 1) is generated by extracting positive edge points and negative edge points whose luminance changes from light to dark.

  Note that when the camera 2 is a monochrome camera, a grayscale captured image is obtained from the luminance of each pixel, and thus the above-described processing for generating a grayscale image from a color captured image is not necessary.

  Here, Im1 shown in FIG. 3A is an example of a grayscale image generated from the captured image 21, and a broken line lane mark 51a on the left side that divides the lane in which the vehicle 1 (own vehicle) is traveling. 55a and right solid line lane marks 61a, 62a and 63a. Im1 includes light images 71a to 78a extending around the right solid line lane marks 61a and 62a. Further, for the sake of explanation, a vanishing point 81a of the captured image 21 and a horizontal line 82a indicating the same height in the camera coordinate system with a one-dot chain line are added.

  Im2 illustrated in FIG. 3B is an edge image generated by performing edge extraction processing on the grayscale image Im1. In the edge image Im2, the left broken line lane mark image portions 51c to 55c, the right solid line lane mark image portions 61c, 62c, and 63c, and the light image portion 71c to 78c are linearly continuous. The image portion includes many edge points. For the sake of explanation, a point 81c corresponding to the vanishing point 81a of the captured image 21 (hereinafter referred to as vanishing point 81c) and a line 82c corresponding to the horizontal line 82a indicated by a one-dot chain line (hereinafter referred to as horizontal line 82c). Is added.

  In the following, in the coordinate system (camera coordinate system) of the captured image 21, the gray scale image Im1, and the edge image Im2 (hereinafter collectively referred to as the captured image 21), the horizontal direction is referred to as the x-axis direction, and the vertical direction is the y-axis. It is called direction. The horizontal direction of the real space coordinate system is referred to as the X-axis direction, the vertical direction is referred to as the Y-axis direction, and the depth is referred to as the Z-axis direction.

  Next, FIG. 2A / STEP 3 to FIG. 2A / STEP 4 are processes by the candidate image portion extraction unit 13. The candidate image portion extraction unit 13 labels the edge image Im2 with a label for a group of edge points (hereinafter referred to as edge portions) and the adjacent edge portions as image portions of the same object. Perform clustering to associate as there is.

  Further, the candidate image portion extraction unit 13 in FIG. 2 (a) / STEP4, among the image portions extracted by clustering, the image portion having line characteristics (for example, the ratio of linearly continuous edge points is predetermined). Image portions that are equal to or higher than the level) are extracted as candidate image portions that are candidates for the image portion of the lane mark.

  Note that the processing step by the candidate image portion extraction unit 13 corresponds to the candidate image portion extraction step in the lane mark recognition method of the present invention.

  The subsequent FIG. 2A / STEP 5 and FIG. 2A / STEP 6 (FIG. 2A / STEP 61 to FIG. 2A / STEP 66) are processes by the specific candidate excluding unit 14. In addition, the process process by the specific candidate exclusion part 14 is corresponded to the specific candidate exclusion process in the lane mark recognition method of this invention.

  The specific candidate excluding unit 14 performs reverse projection conversion from the camera coordinates to the real space coordinates for each candidate image portion extracted in FIG. 2A / STEP4, and the real space corresponding to each candidate image portion. The position of the object at is calculated (FIG. 2 (a) / STEP5). Note that an image portion whose size is equal to or smaller than a predetermined size (a size that is not assumed to be an image portion of a lane mark) may be excluded from the candidate image portions.

  The specific candidate excluding unit 14 selects each candidate image portion (FIG. 3) whose distance from the vehicle 1 in the corresponding real space is equal to or greater than a predetermined distance from the candidate image portions extracted in FIG. 2B, candidate image portions 53c to 55a, 63c, and 71c to 77c at positions far from the vehicle 1 (camera 2) from the two-dot chain line 83c in FIG. 2A / STEP 61 to FIG. / STEP 66 is executed. Hereinafter, the candidate image part to be processed according to FIG. 2A / STEP 61 to FIG. 2A / STEP 66 is referred to as “current candidate image part”.

  The specific candidate excluding unit 14 determines whether or not the current candidate image portion is vertical on the captured image 21 based on the ratio of the vertical edge to the edges included in the current candidate image portion (FIG. 2A / (STEP 61). For example, when the ratio of the vertical edges is a predetermined ratio (for example, 60%) or more, it is determined that the current candidate image portion is vertical on the captured image 21. Note that the “vertical edge” refers to an edge extending in the y direction formed by a collection of edge points where a change in shading (change in luminance) with respect to a change in position in the horizontal direction is steep. In the present embodiment, the “vertical edge” includes not only an edge extending in the vertical direction (y direction) in the photographic image 21 in a strict sense but also an edge that can be regarded as extending substantially in the vertical direction.

  For example, the left broken line lane mark image portions 53c to 55a and the right solid line lane mark image portion 63c shown in FIG. 3B hardly include vertical edges. For this reason, the determination result for the image portions 53c to 55c of the left solid line lane mark and the image portion 63c of the right broken lane mark is negative.

  On the other hand, the image portions 71c to 77c of the light image shown in FIG. 3B include many vertical edges, and the ratio is equal to or higher than a predetermined ratio. For this reason, the said determination result about the image parts 71c-77c of the image of light becomes affirmative.

  When the determination result is affirmative (FIG. 2 (a) / STEP 61... YES), the specific candidate exclusion unit 14 determines whether or not the current candidate image portion satisfies the first condition (FIG. 2 ( a) / STEP 62).

  Here, the first condition is a condition for the length in a specific direction corresponding to the vertical direction (Y direction) of the real space coordinate system corresponding to the current candidate image portion.

  The length in the specific direction corresponding to the vertical direction (Y direction) of the real space coordinate system corresponding to the current candidate image portion is equal to the length on the captured image 21 of the current candidate image portion and FIG. It is calculated from the position of the object in the real space corresponding to the current candidate image portion calculated in STEP 5.

  As a first condition, estimated from a distance between the vehicle 1 and an object in the real space corresponding to the current candidate image portion (a distance calculated from the position of the object in the real space and the position of the vehicle 1 in the real space) The condition that the difference between the length in the y direction of the image portion of the lane mark to be performed and the length in the y direction of the current candidate image portion is equal to or greater than a predetermined threshold (in other words, the y direction of the current candidate image portion The condition that the length deviates from the length in the y direction of the image portion of the lane mark estimated from the distance between the vehicle 1 and the object in the real space corresponding to the current candidate image portion can be employed. . Here, the length in the y direction of the image portion of the lane mark on the captured image 21 estimated from the distance is created in advance by using a predetermined function that inputs the distance and the imaging magnification, for example. It is obtained by searching the map by distance.

  For example, in FIG. 3B, for example, the image portions 73c to 77c of the light image are images whose length in the y direction is larger than the image portion of the lane mark assumed from the actual distance. For this reason, it is determined that the image portions 73c to 77c of the light image satisfy the first condition. On the other hand, the image portions 71c to 72c of the light image are images that cannot be determined to be lane marks because their lengths in the y direction are small considering errors. For this reason, it is not determined that the first condition is satisfied for the image portions 71c to 72c of the light image.

  Further, as the first condition, a condition that the length in the vertical direction (Y direction) of the object in the real space corresponding to the current candidate image portion is equal to or longer than a predetermined length may be employed.

  When the determination result of FIG. 2A / STEP 62 is negative (FIG. 2A / STEP 62... NO), the specific candidate excluding unit 14 determines whether or not the current candidate image portion satisfies the second condition. Is determined (FIG. 2A / STEP 63). Here, the second condition is a condition that at least a part of the current candidate image portion is at a position higher than the vanishing point of the captured image in the y direction.

  For example, in FIG. 3B, each of the image portions 71c to 75c of the light image is at a position higher than the horizontal line 82c. For this reason, the said determination result about the image parts 71c-75c of the image of light becomes affirmative.

  When either the determination result of FIG. 2 (a) / STEP 62 or the determination result of FIG. 2 (a) / STEP 63 is affirmative (FIG. 2 (a) / STEP 62... YES or FIG. 2 (a) / STEP 63... YES The specific candidate excluding unit 14 determines whether or not the vehicle 1 is changing lanes (FIG. 2A / STEP 64). Whether or not the lane is being changed includes, for example, characteristics of the traveling road (whether the road recognized by the lane recognition process so far is branched, characteristics such as the number of lanes), the operating state of the winker, the vehicle 1 And the steering operation speed. For example, there is no branch on the travel path, there are two or more lanes, a blinker in one direction is lit, the speed of the vehicle 1 is equal to or higher than a predetermined speed, and the steering operation speed in the direction is predetermined. When it is within the range, it can be determined that the lane is being changed.

  When the determination result of FIG. 2A / STEP 64 is affirmative (FIG. 2A / STEP 64... YES), the specific candidate excluding unit 14 has the current candidate image portion in the central region of the captured image 21. (FIG. 2A / STEP 65). Here, the central region of the captured image 21 refers to a predetermined range on the captured image 21 corresponding to the front portion of the vehicle 1 (the front portion having the same width w as the vehicle width, see FIG. 4).

  For example, as shown in FIG. 4, during the lane change, lane mark candidate image portions 91c to 95c whose image portions are substantially vertical and have a predetermined length can be extracted. The determination result of FIG. 2A / STEP65 for such a candidate image portion is a positive determination result.

  When the determination result of FIG. 2 (a) / STEP64 is negative (FIG. 2 (a) / STEP64... NO), or when the determination result of FIG. 2 (a) / STEP65 is negative (FIG. 2 (a) ) / STEP 65... NO), the specific candidate excluding unit 14 excludes the current candidate image part (FIG. 2A / STEP 66), and executes the next candidate image part process.

  When the determination result of FIG. 2 (a) / STEP 61 or FIG. 2 (a) / STEP 63 is negative FIG. 2 (a) / STEP 61... NO, FIG. 2 (a) / STEP 63. a) When the determination result of / STEP65 is affirmative (FIG. 2 (a) / STEP64... YES), the specific candidate exclusion unit 14 does not exclude the current candidate image portion.

  Next, FIG. 2B / STEP 7 is processing by the lane mark recognition unit 15. The processing step by the lane mark recognition unit 15 corresponds to the lane mark recognition step in the lane mark recognition method of the present invention.

  In the lane mark recognition unit 15, the length in the corresponding real space obtained in FIG. 2 (b) / STEP5 is set based on a predetermined range (the length of the broken lane mark (specified length) defined by the law). In this case, a candidate for a broken line lane mark is detected based on candidate image portions that are continuous in a certain direction.

  The lane mark recognizing unit 15 uses a known method described in, for example, Japanese Patent Laid-Open No. 11-219435 when the lane is divided by continuous line lane marks (white line, yellow line, etc.). Detect candidates for.

  The lane mark recognition unit 15 recognizes the lane mark based on these lane mark candidates.

  For example, as shown in FIG. 5, left lane marks 51b to 54b are recognized on the left side of the lane in which the vehicle 1 is traveling, and right lane marks 61b, 62b, and 63b are recognized on the right side.

  In subsequent STEP 8, the lane mark recognition unit 15 recognizes the center position of the lane based on the positions of the left and right lane marks. In the example of FIG. 5, the center position Xc between the inner position XL1 of the left lane marks 51b to 54b on the left side and the inner position XR1 of the right lane mark 61b on the right side is recognized.

  Next, FIG. 2B / STEP9 is processing by the obstacle recognition unit 16.

  The obstacle recognition unit 16 recognizes the candidate image part excluded in FIG. 2A / STEP 66 as an obstacle. For example, as shown in FIG. 5, the illumination 77b is recognized as an obstacle. Further, the illumination 78b whose distance in the real space is within a predetermined distance when viewed from the vehicle 1 can be recognized as an obstacle by performing more detailed obstacle determination such as pattern matching.

  Next, FIG. 2B / STEP 10 is processing by the steering control unit 17. The steering controller 17 assists the driving operation of the driver of the vehicle 1 by operating the steering mechanism 7 so that the vehicle 1 travels near the center position Xc in the lane.

  Here, the steering control unit 17 increases the deviation between the vehicle 1 and the center position Xc, and the interval WR between the vehicle 1 and the right lane mark 61b or the interval WL between the vehicle 1 and the left lane marks 51b to 54b is increased. When the value is equal to or less than a predetermined determination value, the steering mechanism 7 is operated to control the vehicle 1 closer to the center position Xc.

  When it is detected that the vehicle 1 may come into contact with the lights 77b and 78b, the steering control unit 17 performs control to avoid contact or warns the driver via the speaker 5 or the display 6. Do.

(Operation and effect of this embodiment)
According to studies by the inventors of the present application, the lane mark is often imaged as a diagonal line on the captured image 21. Since the lane mark is laid on the road, the length in the vertical direction (Y direction) in the real space is substantially equal to zero if the road is flat. For this reason, it is highly accurate whether the candidate image portion represents a lane mark by the length in a specific direction corresponding to the vertical direction in the real space corresponding to the candidate image portions 51c to 55c, 61c to 63c, and 71c to 78c. Can be determined.

  According to the lane mark recognition apparatus 10 of the present embodiment configured with attention paid to this point, the specific direction corresponding to the vertical direction in the real space corresponding to the candidate image portions 51c to 55c, 61c to 63c, and 71c to 78c. Candidate image portions 71c to 78c that are not likely to be lane mark image portions that satisfy the condition that the length of the lane mark is equal to or greater than a predetermined length are excluded. For this reason, compared with the case where collation with information, such as a magnitude | size and a brightness | luminance change pattern regarding a detected object and multiple types of obstacle, is required, calculation cost is reduced.

  Further, according to the lane mark recognition device 10 having the above configuration, the length in the vertical direction (y direction) deviated from the length in the vertical direction (y direction) of the image portion of the assumed lane mark on the vehicle 1 and the captured image. Candidate image portions 73c to 78c having are excluded. As a result, according to the lane mark recognition apparatus 10 having the configuration, it is possible to improve the lane mark detection accuracy while reducing the calculation cost.

  Further, as described above, since the lane mark is laid on the road, in the captured image 21 while traveling on a flat road, the entire vertical position of the image portion of the lane mark is from the vanishing point 81a. Are often in lower positions. That is, when the position of the candidate image portion in the vertical direction on the captured image is higher than the vanishing point of the captured image, there is a probability that the candidate image portions 71c to 75c are not lane mark image portions. high.

  According to the lane mark recognizing device 10 having the configuration configured focusing on this point, it is possible to achieve both reduction in calculation cost and improvement in detection accuracy of lane marks.

  In a situation where the vehicle 1 straddles the lane mark, in the captured image 21, the candidate image portions 91 a to 91 c of the lane mark can extend in the vertical direction to the center of the captured image 21.

  According to the lane mark recognizing device 10 having the configuration configured focusing on this point, even if the candidate image portion satisfies the first condition or the second condition, it exists in the center of the captured image 21. If it is, the candidate image portions 91a to 91c are not excluded.

  As a result, according to the lane mark recognizing device 10 having the configuration, the calculation cost is reduced and the probability of malfunction in the lane mark exclusion process is reduced, so that the accuracy of detecting the lane mark can be improved.

  Moreover, when the vehicle 1 straddles the lane mark, the vehicle 1 is highly likely to change lanes. According to the lane mark recognizing device having the configuration configured by paying attention to this point, the calculation cost is reduced and the vehicle exists at the center of the captured image only when the probability that the vehicle straddles the lane mark is high. Since the candidate image portions 91a to 91c are not excluded, lane mark detection accuracy can be improved.

  Further, according to the lane mark recognition apparatus 10 having the configuration, the calculation result in the lane mark recognition process can be used for the recognition of the obstacles 77b and 78b. As a result, the calculation cost of the entire apparatus is reduced.

  Moreover, according to the lane mark recognition apparatus having the configuration, the exclusion process is performed on the long-distance candidate image portions 53c to 55c and 71c to 77c that do not directly affect the control of the vehicle 1. As a result, the calculation cost of the entire apparatus is reduced while maintaining the lane mark detection accuracy.

(Modification)
The determination as to whether the candidate image portion is vertical (FIG. 2A / STEP 61) may be omitted.

  Further, FIG. 2A / STEP 63 may be omitted.

  The process of determining whether or not the vehicle is changing lanes (FIG. 2 (a) / STEP 65) may be omitted. Further, in addition to the determination of whether or not the vehicle is changing lanes (FIG. 2 (a) / STEP65), the determination of whether or not the candidate image portion exists in the center (FIG. 2 (a) / STEP66) is omitted. May be.

  The process of recognizing the excluded candidate image portion as an obstacle may be omitted (FIG. 2B / STEP 11).

  DESCRIPTION OF SYMBOLS 1 ... Vehicle (own vehicle), 2 ... Camera, 7 ... Steering mechanism, 10 ... Lane mark recognition apparatus, 11 ... Captured image acquisition part, 12 ... Edge image generation part, 13 ... Candidate image part extraction part, 14 ... Specific image Exclusion part, 15 ... Lane mark recognition part, 16 ... Obstacle recognition part, 17 ... Steering control part, 20 ... Image memory, 21 ... Captured image, 22 ... Edge image.

Claims (8)

A lane mark recognition device for recognizing a lane mark provided on a road based on an image around the vehicle imaged by a camera mounted on the vehicle,
An edge image generating unit that generates an edge image obtained by extracting an edge point having a luminance change amount equal to or greater than a predetermined value from a captured image of the camera;
Among the edge points, a candidate image part extraction unit that extracts candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part;
Of the candidate image portions extracted by the candidate image portion extraction unit, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or greater than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion unit that excludes the candidate image portion that is more than a predetermined distance away from the camera ;
A lane mark recognition apparatus, comprising: a lane mark recognition unit that recognizes a lane mark based on a candidate image portion that has not been excluded by the specific candidate exclusion unit. The lane mark recognition apparatus according to claim 1,
The predetermined length deviates from the length in the specific direction of the image portion of the lane mark on the captured image when it is assumed that there is a lane mark at a position in the real space corresponding to the vehicle and the candidate image portion. A lane mark recognizing device characterized in that the length is the same. In the lane mark recognition apparatus according to claim 1 or 2,
The lane mark recognition apparatus, wherein the specific candidate exclusion unit is configured to exclude the candidate image portion whose position in the specific direction is higher than a vanishing point of a captured image. In the lane mark recognition device according to any one of claims 1 to 3,
The lane mark recognition apparatus, wherein the specific candidate exclusion unit is configured not to exclude a candidate image portion existing in a central region of the captured image. In the lane mark recognition apparatus according to claim 4,
The lane mark recognition apparatus, wherein the specific candidate exclusion unit is configured not to exclude a candidate image portion existing in the center of the captured image only when the vehicle is changing lanes. In the lane mark recognition apparatus according to any one of claims 1 to 5,
An obstacle recognition unit that recognizes the candidate image portion excluded by the specific candidate exclusion unit as an obstacle;
The vehicle is steered so that the vehicle travels in the lane identified by the lane mark recognized by the lane mark recognition unit and avoids contact between the vehicle and the obstacle recognized by the obstacle recognition unit. A lane mark recognition apparatus comprising: a steering control unit that performs support. A vehicle for recognizing a lane mark provided on a road based on an image around the vehicle imaged by a camera mounted on the vehicle,
An edge image generating unit that generates an edge image obtained by extracting an edge point having a luminance change amount equal to or greater than a predetermined value from a captured image of the camera;
Among the edge points, a candidate image part extraction unit that extracts candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part;
Of the candidate image portions extracted by the candidate image portion extraction unit, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or greater than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion unit that excludes the candidate image portion that is more than a predetermined distance away from the camera ;
A vehicle comprising: a lane mark recognition unit that recognizes a lane mark based on a candidate image portion that is not excluded by the specific candidate exclusion unit. A lane mark recognition method for recognizing a lane mark provided on a road based on an image around the vehicle imaged by a camera mounted on a vehicle,
An edge image generation step of generating an edge image obtained by extracting an edge point where the amount of change in luminance with respect to the surrounding portion is a predetermined value or more from the captured image of the camera;
A candidate image part extraction step of extracting candidate image parts of lane marks by combining edge points estimated to be extracted from the same image part of the edge points;
Of the candidate image portions extracted by the candidate image portion extraction step, the length in the specific direction corresponding to the vertical direction in the corresponding real space is equal to or longer than a predetermined length , and the corresponding position in the real space is A specific candidate exclusion step of excluding the candidate image portion that is more than a predetermined distance away from the camera ;
A lane mark recognition method, comprising: a lane mark recognition step for recognizing a lane mark based on a candidate image portion that has not been excluded by the specific candidate exclusion step.