JP2012093854A - Road-surface image generation vehicle, road-surface image generation device, and road-surface image generation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide technique capable of generating images of a road surface along a curve when a vehicle travels along the curve.SOLUTION: A road-surface image generation vehicle 10 includes: road-surface photographing means of photographing a road surface repeatedly; a first line camera 14 which photographs a road surface within a first visual field range 62; a second line camera 16 which photographs a road surface within a second visual field range 60 which intersects the first visual field range; means of photographing the road surfaces by the first line camera and second line camera in first timing, photographing the road surfaces by the first line camera and second line camera in second timing, specifying a matching point between a photographed image of the first line camera in the first timing and a photographed image of the second line camera in the second timing; specifying a matching point of a photographed image of the second line camera in the first timing and a photographed image of the first line camera in the second timing, and arraying images photographed by the road-surface photographing means based upon the first matching point and second matching point.

Description

  The present invention relates to a technique for arranging a plurality of images taken from a traveling vehicle and generating a continuous image of a road surface.

  Patent Document 1 discloses an apparatus that captures a road surface while traveling and generates a continuous image of a road surface by arranging a plurality of captured images. This road surface image generation device has a line camera that captures a high-definition field of view that is long in the width direction of the vehicle. Each time the vehicle advances by the width of the line camera's visual field range (distance along the traveling direction of the vehicle), the road surface is photographed by the line camera. By arranging a plurality of images obtained by repeatedly photographing with a line camera, it is possible to generate an image of the entire road surface in a range photographed by the line camera. In addition, this image generation apparatus includes an area camera that captures a visual field range wider than the visual field range of the line camera. The area camera captures the road surface less frequently than the line camera. Each time the vehicle travels a predetermined distance set shorter than the width of the field camera viewing range (distance along the traveling direction of the vehicle), the area camera captures the road surface. Therefore, a part of the image captured by the area camera at a certain capturing timing overlaps with a part of the image captured by the area camera at the immediately preceding capturing timing. This image generation apparatus calculates the amount of change in the traveling direction of the vehicle between these image capturing timings by arranging two images captured continuously by the area camera so that their overlapping portions coincide. Then, by arranging a plurality of images photographed by the line camera based on the calculated amount of change in the traveling direction of the vehicle, an image of the entire road surface in the range photographed by the line camera is generated. In this way, by arranging a plurality of images taken by the line camera in consideration of changes in the traveling direction of the vehicle, when the road surface is curved, an image of the entire curved road surface can be generated. it can. Without this technique, a linearly extending road surface image is generated even though a curved road surface is photographed.

JP 2006-214854 A

  According to the technique using the area camera described above, a change in the traveling direction of the vehicle can be accurately identified. However, since the area camera captures a wide range with a high resolution, it cannot continuously capture in a short cycle. For this reason, when the vehicle is driven at a high speed, there is no overlap in the images continuously captured by the area camera, and the traveling direction cannot be specified. That is, in the conventional technique, it is necessary to drive the vehicle at a low speed so that there is an overlapping portion between two images continuously captured by the area camera. For this reason, it took time to photograph the road surface, and there was a risk of disturbing the surrounding traffic during the photographing. Therefore, the present specification provides a technique capable of generating a road surface image that reflects a change in the traveling direction of the vehicle even when the vehicle is traveling at high speed.

  The road surface image generation vehicle disclosed in this specification captures a road surface while traveling and generates a continuous image of the road surface. The road surface image generation vehicle includes a road surface photographing unit, a first line camera, a second line camera, a first matching point specifying unit, a second matching point specifying unit, and an image arranging unit. . The road surface photographing means repeatedly photographs the road surface. The first line camera photographs a road surface within a first visual field range that is long in one direction. The second line camera photographs the road surface within the second visual field range. The second visual field range is long in one direction and intersects the first visual field range. The first matching point specifying means matches the first image captured by the first line camera at the first timing with the fourth image captured by the second line camera at a second timing after a predetermined time has elapsed from the first timing. A first matching point that is a point is specified. The second coincidence point specifying means obtains a second coincidence point that is a coincidence point between the second image photographed by the second line camera at the first timing and the third image photographed by the first line camera at the second timing. Identify. The image arrangement means arranges the images photographed by the road surface photographing means between the first timing and the second timing based on the positional relationship between the first coincidence point and the second coincidence point.

  FIG. 13 illustrates the traveling direction 150 of the road surface image generation vehicle, the visual field range 110 of the first line camera, and the visual field range 120 of the second line camera. In FIG. 13, the traveling direction 150 and the visual field ranges 110 and 120 at the first timing are shown as the traveling direction 150a and the visual field ranges 110a and 120a, and these at the second timing are represented by the traveling direction 150b, the visual field range 110b, 120b. As illustrated, in this road surface image generation vehicle, the visual field range 110 of the first line camera and the visual field range of the second line camera 120 intersect each other. At the first timing, a road surface image (first image) within the visual field range 110a is captured by the first line camera, and a road surface image (second image) within the visual field range 120a is captured by the second line camera. At the subsequent second timing, since the road surface image generation vehicle is moving, a road surface image (third image) in the visual field range 110b is captured by the first line camera, and the second line camera is in the visual field range 120b. A road surface image (fourth image) is taken. When the first to fourth images are captured, a first matching point that is a matching point between the first image and the fourth image is specified. The first coincidence point means a coincidence point (point 160 in FIG. 10) between the visual field range 110a and the visual field range 120b. In addition, a second matching point that is a matching point between the second image and the third image is characterized. The second coincidence point means a coincidence point (point 170 in FIG. 10) between the visual field range 110b and the visual field range 120a. Thus, when two coincidence points (the first coincidence point 160 and the second coincidence point 170) are specified, the visual field ranges 110a and 120a at the first timing, and the visual field ranges 110b and 120b at the second timing, The relative positional relationship of becomes clear. By identifying the coincidence points 160 and 170, the angle between the vehicle traveling direction 150a at the first timing and the vehicle traveling direction 150b at the second timing (that is, the amount of change in the traveling direction) is determined. The road surface image generating vehicle arranges images photographed by the road surface image photographing means based on the positional relationship between the first coincidence point and the second coincidence point. Therefore, even when the traveling direction of the vehicle changes, the images can be arranged according to the change in the traveling direction. Therefore, it is possible to generate an accurate image of the entire road surface in the range photographed by the road surface image photographing means. In addition, the road surface image generation vehicle has a first timing and a second timing based on an image photographed by the first line camera and an image photographed by the second line camera (that is, an image within two linear visual field ranges). The amount of change in the direction of travel between and is specified. Since the first line camera and the second line camera do not need to shoot a wide field of view, unlike the conventional area camera, it is possible to shoot continuously in a short cycle. Therefore, this road surface image generation vehicle can generate a continuous image of the road surface while traveling at a high speed.

  In the road surface image generation vehicle described above, the second matching point specifying unit specifies the first matching point after the first matching point specifying unit specifies the first matching point, and the second image and the second image centered on the first matching point. It is preferable that the degree of coincidence in a range where the second image and the third image overlap is calculated while relatively rotating the three images, and the second coincidence point is specified based on the calculated degree of coincidence.

  If the first matching point is specified, the second matching point overlaps the second image and the third image when the second image and the third image are relatively rotated around the first matching point. Exists in range. The range in which the second image and the third image overlap varies depending on the angle at which the two images are relatively rotated. The overlapping range at the best matching angle can be the second matching point. Since the range where the second image and the third image overlap is a narrow range, the second matching point can be easily specified. By specifying the second matching point based on the matching degree in the range where the second image and the third image overlap, it is possible to reduce the load required for the arithmetic processing for specifying the second matching point.

  In the road surface image generation vehicle described above, it is preferable that the first line camera and the road surface photographing means are configured by one line camera.

  According to such a configuration, the road surface image generation vehicle can be further simplified.

  A road surface image generation method that performs the above-described processing is one of the inventions disclosed in this specification. A road surface photographing means; a first line camera; a second line camera; a first control means for photographing a road surface with the first line camera at a first timing; A road surface imaging vehicle having second control means for capturing a road surface with the first line camera at a second timing after a predetermined time has elapsed from the timing and also capturing a road surface with the second line camera is also disclosed in the present specification. One. An accurate image of the entire road surface can be generated by performing the above-described processing using each image captured by the road surface image capturing vehicle. Also, a road surface image generation device and a road surface image generation method for performing identification of the first coincidence point, identification of the second coincidence point, and image arrangement for each image photographed by the road surface photographing vehicle, This is one of the inventions disclosed in this specification.

1 is a schematic side view of a road surface image generation vehicle 10. FIG. 1 is a schematic top view of a road surface image generation vehicle 10. FIG. The flowchart which shows whole image generation processing. The figure which shows the image C1 which arranged the images A1-A1000 straight. The figure which shows the image D1 which arranged the images B1-B1000 straightly. The figure which overlap | superposed the image C1 and the image D1 on the subject 70 as a reference | standard. Explanatory drawing of the positional relationship of image A1, A1000, B1, and B1000. The figure which shows the positional relationship of image A1, A1000, B1, and B1000 when the matching point 82 is specified. Explanatory drawing about the movement path | route 90. FIG. The figure which shows the image which arranged the images A1-A1000 along the movement path | route 90. FIG. The figure which shows the image which superimposed the images B1-B1000 arranged along the movement path | route 90 on the images A1-A1000 arranged along the movement path | route 90. FIG. Explanatory drawing of an angle calculation process. The figure which shows the positional relationship of the visual field range 110a, 110b, 120a, 120b.

The features of the embodiment described below will be described.
(Characteristic 1) Each time the road surface image generation vehicle travels by the width of the visual field range (distance along the traveling direction of the road surface image generation vehicle), the first line camera captures the road surface within the visual field range.
(Feature 2) Each time the road surface image generation vehicle travels by the width of the visual field range (distance along the traveling direction of the road surface image generation vehicle), the second line camera captures the road surface within the visual field range.
(Feature 3) The interval between the first timing and the second timing is longer than the time interval at which the first line camera and the second line camera repeatedly take images.
(Feature 4) The interval between the first timing and the second timing is set to an interval where a part of the first image and the fourth image overlap and a part of the second image and the third image overlap.
(Characteristic 5) The first coincidence point specifying means arranges the images continuously photographed by the first line camera in a line and the images continuously photographed by the second line camera in a line. By comparing the images, the first matching point is specified. That is, the first coincidence point specifying means starts the calculation from the state where the vehicle has traveled straight between the first timing and the second timing, and specifies the first coincidence point.

  FIG. 1 is a schematic side view of a road surface image generation vehicle 10 according to the embodiment, and FIG. 2 is a schematic top view of the road surface image generation vehicle 10. As shown in FIGS. 1 and 2, the road surface image generation vehicle 10 includes a vehicle 12, a line camera 14, a line camera 16, an encoder 18, a storage device 20, and a control device 22.

  The vehicle 12 is a microbus modified for measurement. A support 24 that protrudes forward from the vehicle 12 is installed on the top of the vehicle 12. The line camera 14 and the line camera 16 are fixed to the front upper part of the vehicle 12 by a support 24. The line camera 14 and the line camera 16 are fixed directly below. The line cameras 14 and 16 include an image sensor unit in which a large number (about 4000) of image sensors (CCD, CMOS, etc.) are arranged in the longitudinal direction and several (about 10) are arranged in the short direction. It is built-in and takes a picture of a long and narrow field of view that extends linearly. The line cameras 14 and 16 photograph the road surface directly below. A range 60 in FIG. 2 indicates the visual field range of the line camera 14. The line camera 14 photographs a road surface within the visual field range 60. As shown in FIG. 2, the visual field range 60 extends long in a direction inclined by an angle θ <b> 1 with respect to the traveling direction 50 of the vehicle 12. The length of the visual field range 60 in the longitudinal direction is about 4000 mm, and the length of the visual field range 60 in the short direction is about 1 mm. A range 62 in FIG. 2 indicates the visual field range of the line camera 16. The line camera 16 captures a road surface within the visual field range 62. As shown in FIG. 2, the visual field range 62 extends long in a direction inclined by an angle θ2 on the opposite side to the visual field range 60 with respect to the traveling direction of the vehicle 12. In this embodiment, the angle θ1 and the angle θ2 are equal. The length of the visual field range 62 in the longitudinal direction is about 4000 mm, and the length of the visual field range 60 in the short direction is about 1 mm. The line cameras 14 and 16 are arranged so that the visual field range 60 and the visual field range 62 intersect to form an X shape. The position where the visual field range 60 and the visual field range 62 intersect with each other is shifted from the center of each visual field range by the size of the lens of each camera. However, since the deviation is so small that it can be ignored, the following description will be made assuming that the visual field range 60 and the visual field range 62 intersect at the center point. The intersecting visual field range composed of the visual field ranges 60 and 62 is symmetrical on the left and right of the vehicle 12.

  The encoder 18 has a rotating part 18a. The rotating part 18a is in contact with the road surface and rotates as the vehicle 12 travels. The encoder 18 detects the travel distance of the vehicle 12 by detecting the rotation of the rotating portion 18a. The rotating portion 18 a is disposed in the vicinity of the rear wheel of the vehicle 12 and at the intermediate point in the width direction of the vehicle 12. The encoder 18 detects the average value of the travel distance of the left and right rear wheels. The encoder 18 outputs a pulse each time the vehicle 12 travels by the width of the visual field ranges 60 and 62 along the traveling direction 50. The length of the visual field ranges 60 and 62 in the short direction is 1 mm, and the visual field ranges 60 and 62 are inclined at an angle θ with respect to the traveling direction, so the width of the visual field ranges 60 and 62 along the traveling direction 50 is Is 1 / sin θ (mm).

  The storage device 20 is connected to the control device 22. The storage device 20 stores images taken by the line cameras 14 and 16.

  Line cameras 14 and 16, an encoder 18, and a storage device 20 are connected to the control device 22. The control device 22 controls the operations of the line cameras 14 and 16 and the storage device 20. The above-described pulse is input from the encoder 18 to the control device 22. The control device 22 performs imaging with the line cameras 14 and 16 every time one pulse is input (that is, every time the vehicle 12 travels 1 / sin θ (mm)). Since the line cameras 14 and 16 capture the road surface every time the vehicle 12 travels by the width along the traveling direction 50 of the visual field ranges 60 and 62, the road surface on which the vehicle 12 travels by each of the line cameras 14 and 16 is approximately. Photographed without gaps. The control device 22 stores the captured images in the storage device 20 together with the imaging order. Further, every time 1000 pulses are input from the encoder 18 (that is, every time the vehicle advances 1000 / sin θ (mm)), the control device 22 executes the entire image generation process described below.

  Next, the entire image generation process performed by the control device 22 will be described. FIG. 3 is a flowchart showing the entire image generation process. As described above, the line cameras 14 and 16 photograph the road surface every pulse, and the control device 22 executes the entire image generation process every 1000 pulses. Therefore, when the control device 22 starts the entire image generation process, 1000 images are obtained by the line camera 14 and 1000 images are obtained by the line camera 16. Hereinafter, the images captured by the line camera 14 are referred to as images A1, A2, A3,... A1000 in the order of capturing, and the images captured by the line camera 14 are referred to as images B1, B2, B3,. In the entire image generation process, an image of the entire road surface within the imaging range of the images A1 to A1000 is generated by arranging the images A1 to A1000. Even during the execution of the entire image generation process, the line cameras 14 and 16 capture an image for each pulse. An image captured during the execution of the whole image generation process is used for the next whole image generation process.

  In step S2, as shown in FIG. 4, the control device 22 arranges the images A1 to A1000 in a straight line along the direction inclined by the angle θ1 with respect to the longitudinal direction of the image. That is, the images A1 to A1000 are arranged in a straight line along the traveling direction 50 of the vehicle 12 when each image is captured. As a result, an image C1 obtained by joining the images A1 to A1000 is generated. Similarly, the controller 22 linearly displays the images B1 to B1000 along the direction inclined by the angle θ2 with respect to the longitudinal direction of each image (that is, the traveling direction 50 of the vehicle 12 when each image is captured). Line up. Thereby, as shown in FIG. 5, an image D <b> 1 in which the images B <b> 1 to B <b> 1000 are connected is generated. 4 and 5 and the respective drawings described below, the widths of the images A1 to A1000 and the images B1 to B1000 are shown larger than the actual width, and the images A1 to A1000 and the images B1 to B1000 are shown. Some illustrations are omitted. The line camera 14 and the line camera 16 photograph the same road surface. Therefore, as shown in FIGS. 4 and 5, the same subjects 70 and 72 are shown in a portion where the imaging ranges of the image C1 and the image D1 overlap.

  In step S4, the control device 22 identifies the same subject among the subject that appears below the center of the image C1 and the subject that appears below the center of the image D1. The subject is identified by collating a plurality of points with the same shape of the subject. 4 and 5, the subject 70 is identified as the same. Then, the control device 22 calculates a relative position and a relative angle between the image C1 and the image D1 in which the identified subject 70 matches (hereinafter, the relative position and the relative angle are referred to as a relative positional relationship). That is, as shown in FIG. 6, when the image D1 is superimposed on the image C1, the relative positional relationship between the image D1 and the image C1 in which the subject 70 in both images matches is calculated.

  In step S6, the control device 22 specifies a coincidence point between the image A1 and the image B1000 based on the relative positional relationship calculated in step S4. For example, when the relative positional relationship shown in FIG. 6 is obtained, a point 80 where the image A1 and the image B1000 intersect is specified as a coincidence point. The position of the coincidence point 80 is specified by the distance from the center point 76 of the image A1 and the distance from the center point 78 of the image B1000.

In step S8, the control device 22 arranges images A1, B1, A1000, and B1000 as shown in FIG. The images A1, B1, A1000, and B1000 are arranged in consideration of the visual field range 60 of the line camera 14 and the visual field range 62 of the line camera 16. The relative positional relationship between the image A1 and the image B1 captured at the same timing is equal to the relative positional relationship between the visual field range 60 and the visual field range 62. Therefore, the images A1 and B1 are arranged such that the center point of the image B1 overlaps the center point 76 of the image A1, and the relative angle of the image B1 with respect to the image A1 is an angle θ1 + θ2. Hereinafter, the set of the image A1 and the image B1 whose relative positional relationship is fixed is referred to as an image set (A1, B1). Similarly, the image A1000 and the image B1000 are arranged such that the center point 78 of the image B1000 overlaps the center point of the image A1000, and the relative angle of the image B1000 with respect to the image A1000 is an angle θ1 + θ2. Hereinafter, a set of the image A1000 and the image B1000 whose relative positional relationship is fixed is referred to as an image set (A1000, B1000). Further, the image A1 and the image B1000 are arranged so as to overlap at the coincidence point 80. Further, the angle φ between the image A1 and the image B1000 is temporarily set to an angle θ1 + θ2. Next, the control device 22 rotates the image group (A1000, B1000) with respect to the image group (A1, B1) around the coincidence point 80 (that is, the image group (A1, B1) with respect to the image group (A1, B1)). While changing the relative angle φ of A1000 and B1000), the degree of coincidence of these images is calculated in the overlapping range of the images A1000 and B1. As described above, since the line cameras 14 and 16 include several image sensors in the short direction, the overlapping range of the image A1000 and the image B1 is composed of a plurality of pixels. The control device 22 calculates the degree of coincidence based on the pixel values in the overlapping range. Then, the angle φ _{max at} which the highest matching degree is obtained is specified, and the overlapping range between the image A1000 and the image B1 at the angle φ _max is specified as the matching point 82 as shown in FIG. The position of the coincidence point 82 is specified by the distance from the center point 76 and the distance from the center point 78. Thus, when specifying the second matching point 82, only the angle φ of the image set (A1000, B1000) with respect to the image set (A1, B1) is undetermined, so the angle φ is changed. By calculating the degree of coincidence of overlapping portions between the image A1000 and the image B1, the coincidence point 82 can be easily identified. In this way, by specifying the coincidence point 82 in consideration of only the overlapping portion between the image A1000 and the image B1 when the angle φ is changed, it is possible to reduce the burden of calculation processing for identifying the coincidence point 82. . When the matching points 80 and 82 are specified, as shown in FIG. 8, the relative positional relationship between the images A1, B1, A1000, and B1000 is determined. Each of the images A1, B1, A1000, and B1000 shows the relative positional relationship between the visual field ranges 60 and 62 when these images are taken. Accordingly, when the relative positional relationship between the images A1, B1, A1000, and B1000 is determined, the field of view range 60 when the image A1 is photographed, the field of view range 62 when the image B1 is photographed, and the field of view when the image A1000 is photographed. The relative positional relationship between the range 60 and the visual field range 62 when the image B1000 is captured is determined. The relative positional relationship between the traveling direction 50 of the vehicle 12 and the visual field ranges 60 and 62 is determined. Therefore, as shown in FIG. 8, the traveling direction 50 when the images A1, B1 are photographed and the traveling direction 50 when the images A1000, B1000 are photographed are specified from the relative positional relationship between the images A1, B1, A1000, and B1000. can do. Therefore, the amount of change in the traveling direction 50 of the vehicle 12 (angle ψ in FIG. 8) from when the images A1 and B1 are taken to when the images A1000 and B1000 are taken is found. The relative position between the intersection (center point 76) of the images A1 and B1 and the intersection (center point 78) of the images A1000 and B1000 is taken as the vehicle 12 when the images A1 and B1 are taken and the images A1000 and B1000. The relative position with respect to the vehicle 12 (that is, the amount of change in position) is shown. The control device 22 calculates the angle ψ and the position change amount.

  In step S10, the control device 22 calculates a moving path along which the vehicle 12 has moved from the time when the images A1 and B1 are captured to the time when the images A1000 and B1000 are captured, based on the calculated angle ψ and the position change amount. The control device 22 calculates the movement route on the assumption that the route of the vehicle 12 is an arc. For example, when the coincidence points 80 and 82 are specified as shown in FIG. 8, the movement route is calculated as shown by the dotted line 90 in FIG.

  In step S12, the control device 22 arranges the images A1 to A1000 along the calculated movement path 90. The control device 22 arranges the images A1 to A1000 in order from the calculated starting point of the movement route 90. The images A1 to A1000 are arranged such that the center point is located on the movement path 90, and the interval on the movement path 90 is an interval of 1 / sin θ mm (the same interval as the interval at which the images A1 to A1000 are taken). Further, the images A1 to A1000 are arranged at an angle of θ1 with respect to the tangent to the moving path 90 where the center point is located. In this way, as shown in FIG. 10, an image of the entire road surface in the range captured by the line camera 14 from the image A1 to the image A1000 is generated. Note that, in an area where adjacent images overlap (area on the inner circumference side of a curved moving path), an image is generated so that one overlapping image is preferentially displayed. Alternatively, an image obtained by averaging overlapping images may be displayed.

  The control device 22 executes the above-described entire image generation process every time 1000 pulses are input from the encoder 18, and connects the generated image to the image generated by the previous entire image generation process. Therefore, an image of the entire road surface on which the vehicle 12 has traveled is generated. In this road surface image generation vehicle 10, since an image is generated in consideration of a change in the traveling direction of the vehicle 12, an image of the entire road surface can be accurately generated even when the vehicle 12 travels on a curved road surface. it can. In addition, the line cameras 14 and 16 described above can capture the road surface with a period that is extremely shorter than that of the area camera. Therefore, the road surface image generation vehicle 10 can generate a road surface image while traveling at a high speed.

  In the above-described embodiment, the images A1 to A1000 are arranged along the movement path 90, but the images B1 to B1000 may be arranged along the movement path 90. Further, as shown in FIG. 10, an interval is formed between the adjacent images A <b> 1 to A <b> 1000 in the region on the outer periphery side of the curved moving path 90. Therefore, as shown in FIG. 11, the road surface is obtained by superimposing the image in which the images A1 to A1000 are arranged and the image in which the images B1 to B1000 are arranged, or by calculating a pixel value by complementing each image. An entire image may be generated. According to such a configuration, the gaps between the images A1 to A1000 are complemented by the superimposed images B1 to B1000. Therefore, a more precise image can be generated.

  In the above-described embodiment, the angle ψ (change amount in the traveling direction) and the position change amount obtained from the relative positional relationship between the images A1, B1, A1000, and B1000 are used when calculating the movement path 90. However, since the images A1 to A1000 are taken at an interval of 1 / sin θ1 (mm) based on the pulse output from the encoder 18, the distance traveled by the vehicle 12 before taking the images A1 to A1000 is known in advance. (1000 / sin θ1 (mm)). Therefore, the movement path 90 may be calculated based on the angle ψ obtained from the relative positional relationship between the images A1, B1, A1000, and B1000 and the movement distance that is known in advance. Furthermore, when the vehicle speed can be regarded as known while images A1 to A1000 and B1 to B1000 are photographed, an image collation operation of coincident points of images A1, B1, A1000, and B1000 is possible, and a route can be estimated. . In the above-described embodiment, the angle ψ is calculated. However, if the positions of the coincidence points 80 and 82 are determined, the angle ψ is uniquely determined. Therefore, the movement route 90 of the vehicle 12 may be calculated from the positions of the coincidence points 80 and 82 without calculating the angle ψ.

  In the above-described embodiment, the road surface image generation vehicle 10 that generates an image of the entire road surface while shooting the road surface has been described. However, it is not always necessary to simultaneously perform the shooting of the road surface and the entire image generation process described above. For example, an image of the road surface is taken and the images taken in the storage device 20 and data indicating the order thereof are stored, then the stored data is transferred to another computer, and the whole image is generated by the computer. May be. Further, while photographing the road surface, the photographed images and the data indicating the photographing order may be transmitted wirelessly to an external computer, and the entire computer may perform the entire image generation process.

  In order to accurately execute the above-described processing, the angle θ1 between the traveling direction 50 of the vehicle 12 and the visual field range 60 and the angle θ2 between the traveling direction 50 and the visual field range 62 are accurately set. Need to be. Therefore, the road surface image generation vehicle 10 preferably performs the angle setting process described below before photographing the road surface.

  In the angle setting process, the road surface image generation vehicle 10 travels straight on a road surface on which a square having a known side length is drawn. Thus, the square image is taken by the line cameras 14 and 16. As described above, the line cameras 14 and 16 perform imaging for each pulse. Next, as shown in FIG. 12, the images A1 to An photographed by the line camera 14 are arranged along the short direction of the image (the short direction of the visual field range 60). Actually, since the visual field range 60 is inclined by an angle θa with respect to the traveling direction of the vehicle 12, in the image generated by being arranged in the short direction, a subject that should be square as shown in FIG. 12. 95 is displayed as a parallelogram. Next, the aspect ratio of the image is adjusted so that the length and height of the bottom side of the subject 95 moving to the generated image are equal (FIG. 12 shows the image after the aspect ratio adjustment). Then, the inclination θa of the vertical sides of the parallelogram is calculated. The calculated vertical side inclination θa is equal to the angle θ1 between the longitudinal direction of the visual field range 60 and the traveling direction 50 of the vehicle 12. By such processing, the angle θ1 of the visual field range 60 with respect to the traveling direction 50 can be calculated. By adjusting the mounting angle of the line camera 14 based on the calculated angle θ1 or by adjusting the value of the angle θ1 in the program executed by the control device 22, it is possible to accurately photograph the road surface. An angle θ2 between the visual field range 62 of the line camera 16 and the traveling direction 50 of the vehicle 12 can also be calculated by a similar process.

  In order to calculate the movement route accurately with the technique of the present embodiment, as shown in FIG. 8, the image set (A1, B1) and the image set (A1000, B1000) are at two intersections 80, 82. It is necessary to cross. Therefore, the crossing angles θ1 and θ2 of the visual field ranges 60 and 62 need to be set to angles at which two intersections 80 and 82 are obtained. For example, if the curvature of the moving route is even larger in FIG. 8, the intersection 80 cannot be obtained. When such a movement route is assumed, it is better to make the intersection angles θ1 and θ2 smaller so that the intersections 80 and 82 are obtained. In addition, the cycle for executing the movement route calculation process needs to be appropriately set so that two intersections 80 and 82 can be obtained. For example, when the movement route calculation process is performed in a shorter cycle than the above-described embodiment (for example, every time the images A1 to A500 and the images B1 to B500 are photographed), the distance between the points 76 and 78 in FIG. Therefore, the intersection points 80 and 82 are easily obtained. However, if the intersection angles θ1 and θ2 are too small, the inclination of the visual field range with respect to the traveling direction becomes large, so that the distortion of the captured image becomes large. If the cycle of the movement route calculation process is too short, the burden on the control device for the process increases. Therefore, it is necessary to appropriately set the intersection angles θ1 and θ2 and the execution cycle of the movement route calculation process according to the assumed curvature of the movement route of the vehicle 12, and the like.

  In the above-described embodiment, the images A1, A1000, B1, and B1000 are described as one-line images. However, if an attempt is made to photograph the road surface with an appropriate resolution in consideration of the curvature of the moving route of the vehicle 12, the photographing interval may be shortened. Therefore, the image captured by each camera may be a band image of about 100 lines. In this case as well, the image can be processed in the same manner using the crossing angle of the shooting ranges.

  Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above. The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology exemplified in this specification or the drawings can achieve a plurality of objects at the same time, and has technical usefulness by achieving one of the objects.

10: road surface image generation vehicle 12: vehicle 14: line camera 16: line camera 18: encoder 18a: rotating unit 20: storage device 22: control device 50: traveling direction 60: visual field range 62: visual field range 80: coincidence point 82: Matching point 90: travel route

Claims (7)

A road surface image generation vehicle that captures a road surface while driving to generate a continuous image of the road surface,
Road surface photographing means for repeatedly photographing the road surface;
A first line camera that images a road surface within a first field of view that is long in one direction;
A second line camera that images a road surface in a second field of view that is long in one direction and intersects the first field of view;
A first coincidence point that is a coincidence point between the first image photographed by the first line camera at the first timing and the fourth image photographed by the second line camera at the second timing after a predetermined time has elapsed from the first timing. First matching point specifying means for specifying
Second match point specification that specifies a second match point that is a match point between the second image captured by the second line camera at the first timing and the third image captured by the first line camera at the second timing Means,
Image arrangement means for arranging images photographed by the road surface photographing means between the first timing and the second timing based on the positional relationship between the first coincidence point and the second coincidence point;
A road surface image generating vehicle.   The second matching point specifying means specifies the first matching point after the first matching point specifying means specifies the first matching point, while relatively rotating the second image and the third image around the first matching point. The road surface image generating vehicle according to claim 1, wherein the degree of coincidence in a range where the two images and the third image overlap is calculated, and the second coincidence point is specified based on the calculated degree of coincidence.   The road surface image generating vehicle according to claim 1 or 2, wherein the first line camera and the road surface photographing means are constituted by one line camera. Road surface photographing means for repeatedly photographing the road surface;
A first line camera that images a road surface within a first field of view that is long in one direction;
A second line camera that images a road surface in a second field of view that is long in one direction and intersects the first field of view;
A road surface image generation method for generating a continuous image of a road surface by shooting a road surface while a vehicle having a road surface image is traveling,
A road surface photographing step of repeatedly photographing the road surface with a road surface photographing means;
A first control step of photographing a road surface with a first line camera at a first timing and photographing a road surface with a second line camera;
A second control step of photographing the road surface with the first line camera at the second timing after a predetermined time has elapsed from the first timing and photographing the road surface with the second line camera;
First matching point specification that specifies a first matching point that is a matching point between the first image captured by the first line camera at the first timing and the fourth image captured by the second line camera at the second timing Steps,
Second match point specification that specifies a second match point that is a match point between the second image captured by the second line camera at the first timing and the third image captured by the first line camera at the second timing Steps,
An image arranging step for arranging images taken by the road surface photographing means between the first timing and the second timing based on the positional relationship between the first coincidence point and the second coincidence point;
A road surface image generation method comprising: A road surface imaging vehicle for shooting a road surface while traveling,
Road surface photographing means for repeatedly photographing the road surface;
A first line camera that images a road surface within a first field of view that is long in one direction;
A second line camera that images a road surface in a second field of view that is long in one direction and intersects the first field of view;
First control means for photographing a road surface with a first line camera at a first timing and photographing a road surface with a second line camera;
Second control means for photographing the road surface with the first line camera and photographing the road surface with the second line camera at a second timing after a lapse of a predetermined time from the first timing;
A road surface imaging vehicle having A road surface image generation device that generates a continuous image of a road surface from an image captured by the road surface image capturing vehicle according to claim 5,
A first coincidence that identifies a first coincidence point that is a coincidence point between the first image taken by the first line camera at the first timing and the fourth image taken by the second line camera at the second timing Point identification means;
A second coincidence that identifies a second coincidence point that is a coincidence point between the second image photographed by the second line camera at the first timing and the third image photographed by the first line camera at the second timing Point identification means;
Image arrangement means for arranging images photographed by the road surface photographing means between the first timing and the second timing based on the positional relationship between the first coincidence point and the second coincidence point;
A road surface image generating apparatus. A road surface image generation method for generating a continuous image of a road surface from an image captured by a road surface image capturing vehicle according to claim 5,
A first coincidence that identifies a first coincidence point that is a coincidence point between the first image taken by the first line camera at the first timing and the fourth image taken by the second line camera at the second timing A point identification step;
A second coincidence that identifies a second coincidence point that is a coincidence point between the second image photographed by the second line camera at the first timing and the third image photographed by the first line camera at the second timing A point identification step;
An image arranging step for arranging images taken by the road surface photographing means between the first timing and the second timing based on the positional relationship between the first coincidence point and the second coincidence point;
A road surface image generation method comprising:

Images