JP4062145B2 - Imaging device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an imaging apparatus for grasping the state of a traveling path on which a moving body moves.
[0002]
[Prior art]
Conventionally, on the basis of an imaging output signal obtained by performing an imaging operation on a traveling road on which the vehicle is traveling and its surroundings by a camera mounted on the vehicle, various information on the traveling road is obtained and the information is obtained. Various travel path recognition systems are known which are used for vehicle control and assistance to a vehicle driver (see, for example, Patent Document 1 and Patent Document 2).
[0003]
In the travel path recognition system described in Patent Document 1, a travel lane on a travel path is recognized by performing a zooming operation at a zoom speed faster than the shutter speed by an optical zooming mechanism.
[0004]
In the image processing apparatus described in Patent Document 2, the zoom magnification is changed in each part of one image by the digital zoom mechanism, the zoom magnification is increased in the distant part, and the zoom magnification is decreased in the closest part. The driving lane is recognized.
[0005]
[Patent Document 1]
Japanese Patent No. 3229687
[Patent Document 2]
Japanese Patent No. 3084208
[0006]
[Problems to be solved by the invention]
However, in the above-described traveling path recognition system of Patent Document 1, since optical zooming is performed optically by an optical zooming mechanism, there is a problem in durability, and the number of parts increases, so that cost reduction and downsizing are difficult. .
[0007]
In the image processing apparatus disclosed in Patent Document 2, a zooming operation is performed by a digital zoom mechanism. This zooming operation increases the zoom magnification in a distant portion of a captured image and decreases the zoom magnification in a close portion. is doing. In this case, there is a problem that the object needs to be aligned at positions with different zoom magnifications, and the image processing becomes complicated.
[0008]
In general, there is a device that captures the lane of the road ahead of the vehicle with a single camera, but in such a device, if a configuration with a wide-angle lens is used to recognize a wide range, a far lane is recognized. However, when a configuration including a telephoto lens for recognizing a far lane is used, there is a problem that a recognizable field of view becomes narrow and it is difficult to recognize a near lane.
[0009]
The present invention has been made to solve the above-described problem, and an object of the present invention is to provide an imaging apparatus capable of ensuring a wide field of view and accurately recognizing the state of a distant traveling path. It is.
[0010]
[Means for Solving the Problems]
An imaging apparatus according to the present invention includes an imaging sensor that photoelectrically converts an optical image formed on an imaging surface in which a plurality of pixels are two-dimensionally arranged into an electrical signal at each pixel, and a central portion of the captured image An image portion obtained by the imaging sensor, and an imaging unit that forms an optical image in the moving direction of the moving body on the imaging surface so that the image height is large at the peripheral portion and smaller than the central portion at the peripheral portion. A recognizing unit that recognizes a state of a traveling path on which the moving body travels at a central portion having a large height.
[0011]
According to this configuration, the light image in the moving direction of the moving object is formed on the imaging surface so that the image height is large in the center portion of the captured image and the image height is smaller in the peripheral portion than the center portion. Since the state of the traveling road on which the moving body travels is recognized at the central portion where the image height of the image obtained by the sensor is large, it is possible to ensure a wide field of view and accurately recognize the state of the far traveling road.
[0012]
The image pickup apparatus may further include a drive unit that shifts one of the image sensor and the image forming unit in a direction perpendicular to the optical axis, and the image captured by the image sensor. It is preferable to shift the portion having a large image height from the center portion.
[0013]
According to this configuration, either one of the imaging sensor and the imaging unit is relatively displaced in the direction perpendicular to the optical axis by the driving unit, and the image height of the image captured by the imaging sensor is large. Since the portion is shifted from the center portion, a region having a large image height in the captured image can be moved, and image data can be used effectively without displaying unnecessary information.
[0014]
In the imaging apparatus, the moving body is a vehicle traveling on a road, and the recognition unit determines a lane of the road on which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large. Based on the lane recognition unit that recognizes, the vehicle state detection unit that detects the traveling state of the vehicle, the lane recognized by the lane recognition unit, and the traveling state of the vehicle detected by the vehicle state detection unit, A risk determination unit that determines whether or not the vehicle is in a dangerous state. When the risk determination unit determines that the vehicle is in a dangerous state, the vehicle is dangerous to the driver of the vehicle. It is preferable to further include a notification unit that notifies that the vehicle is in a state.
[0015]
According to this configuration, the moving body is a vehicle that travels on the road, and the lane recognition unit recognizes the lane of the road on which the vehicle travels at the central portion where the image height of the image obtained by the imaging sensor is large. Then, the vehicle running state is detected by the vehicle state detecting unit, and the vehicle is dangerous based on the lane recognized by the lane recognition unit by the danger determining unit and the running state of the vehicle detected by the vehicle state detecting unit. It is determined whether or not it is in the correct state. When the notification unit determines that the vehicle is in a dangerous state, the notification unit notifies the driver of the vehicle that the vehicle is in a dangerous state.
[0016]
Therefore, the driver of the vehicle can drive the vehicle safely, and even if the vehicle is in a dangerous state, that is, a state that may cause an accident, this can be prevented in advance.
[0017]
In the imaging apparatus, the lane recognition unit recognizes a lane in which the vehicle travels in a central portion where an image height of an image obtained by the imaging sensor is large, and the lane recognized by the lane recognition unit is It is preferable to further include a vehicle motion control unit that controls the motion of the vehicle so that the image portion obtained by the imaging sensor is always recognized at a central portion where the image height is large.
[0018]
According to this configuration, the lane recognizing unit recognizes the lane in which the vehicle travels at the central portion where the image height of the image obtained by the imaging sensor is large, and the lane recognized by the lane recognizing unit by the vehicle motion control unit is Since the movement of the vehicle is controlled so that it is always recognized at the central portion where the image height of the image obtained by the imaging sensor is large, the vehicle can be guided autonomously.
[0019]
Further, in the above-described imaging device, the pan driving amount calculation unit that calculates the driving amount in the pan direction of the imaging device, and the panning of the imaging device based on the driving amount calculated by the pan driving amount calculation unit. A pan driving unit for driving in the direction, a tilt driving amount calculating unit for calculating a driving amount in the tilt direction of the imaging device, and the driving amount calculated by the tilt driving amount calculating unit. And a tilt driving unit that drives the mobile phone in a tilt direction, and the recognition unit includes a guide marker provided on a traveling path on which the moving body travels in a peripheral portion where an image height of an image obtained by the imaging sensor is small. The pan driving amount calculation unit recognizes the position, and moves the position of the guidance marker recognized by the recognition unit to a central portion where the image height of the image obtained by the imaging sensor is large. The tilt driving amount calculation unit calculates the driving amount in the pan direction for causing the position of the guide marker recognized by the recognition unit to be a central portion where the image height of the image obtained by the imaging sensor is large. It is preferable to calculate the amount of drive in the tilt direction for moving to the right.
[0020]
According to this configuration, the position of the guidance marker provided on the travel path on which the mobile object travels is recognized in the peripheral portion where the image height of the image obtained by the imaging sensor is small. Then, the pan driving amount calculation unit calculates the driving amount in the pan direction for moving the position of the guidance marker recognized by the recognition unit to the center portion where the image height of the image obtained by the imaging sensor is large, The drive amount calculation unit for tilt calculates the drive amount in the tilt direction for moving the position of the guidance marker recognized by the recognition unit to the center portion where the image height of the image obtained by the imaging sensor is large. The imaging device is driven in the pan direction by the pan driving unit based on the driving amount calculated by the pan driving amount calculating unit, and based on the driving amount calculated by the tilt driving amount calculating unit by the tilt driving unit. Thus, the imaging device is driven in the tilt direction.
[0021]
Therefore, when the guidance marker is recognized in the peripheral portion where the image height of the image obtained by the imaging sensor is small, the guidance marker is recognized in the central portion where the image height is large by driving the imaging device in the pan direction and the tilt direction. Therefore, the guidance marker can be recognized with high accuracy.
[0022]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments according to the present invention will be described below with reference to the drawings. In the present embodiment, the moving body provided with the imaging device is described as a vehicle. However, the present invention is not particularly limited thereto, and may be another moving body that travels on a predetermined traveling path.
[0023]
(First embodiment)
An imaging apparatus according to a first embodiment of the present invention will be described. FIG. 1 is a block diagram illustrating the configuration of the imaging apparatus according to the first embodiment. As shown in FIG. 1, the imaging apparatus includes an imaging unit 1, an imaging sensor 2, an imaging sensor driving unit (corresponding to a driving unit) 3, an A / D conversion unit 4, a first image memory 5, and a lane recognition unit. 6, vehicle state sensor (corresponding to vehicle state detection unit) 7, vehicle state determination unit 8, danger determination unit 9, alarm device (corresponding to notification unit) 10, vehicle motion control device 11, switch 12, image processing unit 13 and a second image memory 14 and an image display device 15.
[0024]
The imaging unit 1 has the highest visual acuity in a range called a fovea on the retina corresponding to the center of the field of view, and a lens that imitates the characteristics of a human eyeball (hereinafter, the visual acuity rapidly decreases as the distance from the fovea increases). The optical image in the moving direction of the moving object is formed so that the image height is large in the central part of the captured image and the image height is smaller in the peripheral part than in the central part. An image is formed on the imaging surface of the imaging sensor 2 disposed behind the unit 1. On the imaging sensor 2, an optical image in which the object in the central part is enlarged and the object in the peripheral part is reduced is formed on the imaging surface. The imaging unit 1 is provided so that the light image of the traveling path in the moving direction of the moving body is always captured at the center portion. An image captured using the foveal lens has a large image height in the central portion and a lower image height in the peripheral portion than in the central portion, and distortion occurs from the central portion to the peripheral portion.
[0025]
FIG. 2 is a diagram illustrating lens characteristics of the foveal lens. In FIG. 2, the horizontal axis represents the field angle θ, and the vertical axis represents the image height y. As shown in FIG. 2, in the region where the angle of view θ is small, the image height y is substantially linear with respect to the angle of view θ, and the amount of change with respect to the unit change of the angle of view θ is large. On the other hand, in the region where the angle of view θ is large, the image height y is non-linear with respect to the angle of view θ, and the amount of change with respect to the unit change of the angle of view θ gradually decreases as the angle of view θ increases. The image height y is saturated to a substantially constant value. That is, in an imaging sensor with a fixed pixel size and pixel pitch, the resolution is high in a region where the angle of view θ is small, and the resolution is low in a region where the angle of view θ is large. Accordingly, since the image height y of the light image in the portion with the small angle of view θ is large, the object is enlarged in the central portion of the image, and the object is compressed in the peripheral portion of the image.
[0026]
Returning to FIG. 1, the image sensor 2 includes a CCD color area sensor (hereinafter referred to as a CCD) having a rectangular image pickup region, and an optical image of the subject imaged by the image forming unit 1 is represented by R ( It is photoelectrically converted into an image signal (a signal composed of a signal train of image signals received by each pixel) of red, G (green), and B (blue) color components and output.
[0027]
The imaging sensor driving unit 3 drives the imaging sensor 2 and drives the imaging sensor 2 so as to output an image signal of a predetermined imaging area by setting an imaging area on the CCD of the imaging sensor 2.
[0028]
The A / D converter 4 converts each pixel signal (analog signal) of the image signal output from the image sensor 2 into a digital signal. The first image memory 5 stores the image signal converted into a digital signal by the A / D conversion unit 4.
[0029]
The lane recognition unit 6 recognizes a portion corresponding to an image of a white line display provided on a travel path (in this case, a road) on which the vehicle travels in the image captured by the image sensor 2. Specifically, the lane recognition unit 6 extracts pixels corresponding to the preset extraction color condition from the image captured by the imaging sensor 2 and determines the left and right sides based on the position coordinate data for the extracted pixels. The lane is recognized, and the recognition result is output to the danger determining unit 9 as lane information. The extracted color condition is data relating to a specific luminance and color difference. For example, data relating to the luminance and color difference of the traveling road excluding the white line display is stored in a memory (not shown) and is read out as necessary. . The lane recognition unit 6 recognizes the lane by detecting a boundary between a pixel corresponding to the extraction color condition and a pixel not corresponding to the extraction color condition.
[0030]
The vehicle state sensor 7 is a sensor that detects the traveling state of the vehicle, and the detected traveling state of the vehicle is output to the vehicle state determination unit 8 as vehicle state information. The vehicle state sensor 7 includes a vehicle speed sensor that detects a vehicle speed, a sensor that detects a steering angle, a sensor that detects a braking state, and the like. The information is information relating to the traveling state of the vehicle obtained from the vehicle speed, the steering angle, the braking state, and the like. The vehicle state determination unit 8 determines the traveling state of the host vehicle based on the vehicle state information input from the vehicle state sensor 7, and outputs the determination result to the risk determination unit 9.
[0031]
The danger determination unit 9 determines whether or not the host vehicle is in a dangerous state based on the lane information input from the lane recognition unit 6 and the determination result input from the vehicle state determination unit 8. For example, the danger determination unit 9 calculates the curvature of the lane input from the lane recognition unit 6, determines whether or not the vehicle can be bent at the speed detected by the vehicle state sensor 7, and the vehicle cannot be bent. If it judges, the alarm device 10 and the vehicle quantity movement control apparatus 11 will be driven noting that the own vehicle is in a dangerous state. Further, for example, when the position of the lane input from the lane recognition unit 6 is not detected within a predetermined range within a predetermined period, the danger determination unit 9 determines that the vehicle is meandering and the own vehicle is dangerous. In this state, the alarm device 10 and the vehicle quantity movement control device 11 are driven.
[0032]
When the danger determination unit 9 determines that the vehicle is in a dangerous state, the warning device 10 notifies the driver that the vehicle is in a dangerous state by voice or a visible display. When notifying by voice, for example, a warning sound such as a buzzer sound is generated to notify the driver that the vehicle is in a dangerous state. Moreover, when notifying by the visually recognizable display, for example, the driver is notified that the vehicle is in a dangerous state by causing the LED to emit light.
[0033]
The vehicle motion control device 11 controls vehicle motion, such as a drive system, a braking system, and a steering system. For example, when the risk determination unit 9 determines that the vehicle is in a dangerous state, the vehicle motion control device 11 automatically controls the vehicle speed to decelerate the vehicle. Note that the vehicle motion control device 11 may automatically operate the brake when the danger judgment unit 9 determines that the vehicle is in a dangerous state.
[0034]
In addition, although the imaging device in this embodiment is a structure provided with the alarm device 10 and the vehicle motion control apparatus 11, this invention is not specifically limited to this, Only either the alarm device 10 or the vehicle motion control device 11 is used. May be provided.
[0035]
The switch 12 switches the connection between the first image memory 5 and the image processing unit 13 and the connection between the first image memory 5 and the second image memory 14. In the present embodiment, the connection is switched by the switch 12, but the present embodiment is not particularly limited to this, and the first image memory 5 includes the image processing unit 13 and the second image memory 14. The structure connected with only any one may be sufficient.
[0036]
The image processing unit 13 performs a process of removing distortion from a distorted image captured using a foveal lens. When an image is taken using a foveal lens, distortion occurs from the center portion to the peripheral portion of the image as described above, and thus processing for removing this distortion is necessary. Therefore, the image processing unit 13 removes distortion of the image captured using the foveal lens and creates an image without distortion.
[0037]
When the first image memory 5 and the image processing unit 13 are connected by the switch 12, the second image memory 14 stores an image from which distortion has been removed by the image processing unit 13. Further, when the first image memory 5 and the second image memory 14 are connected by the switch 12, the second image memory 14 reads and stores an image stored in the first image memory 5. To do.
[0038]
The image display device 15 displays the state of the moving direction of the vehicle to the driver, and reads and displays the image data stored in the image memory 14. Examples of the image display device 15 include a monitor device such as a car navigation provided in a vehicle, a head-up display device provided on a windshield of the vehicle, and the like. Note that when the first image memory 5 and the second image memory 14 are connected by the switch 12, the image displayed on the image display device 15 is distorted, and in particular, the peripheral portion of the displayed image. Increases the distortion.
[0039]
Next, the operation of the imaging apparatus in the first embodiment shown in FIG. 1 will be described. FIG. 3 is a flowchart showing an example of the operation of the imaging apparatus according to the first embodiment shown in FIG.
[0040]
In step S <b> 1, the imaging sensor 2 captures a light image in the moving direction of the vehicle imaged by the imaging unit 1. The captured light image is photoelectrically converted and input to the A / D converter 4 as an image signal. The A / D conversion unit 4 converts the input analog image signal into digital and outputs it to the first image memory 5. In step S2, the first image memory 5 temporarily stores the image data.
[0041]
In step S <b> 3, the lane recognition unit 6 recognizes a portion representing the lane in the image captured by the image sensor 2. The lane information recognized by the lane recognition unit 6 is output to the danger determination unit 9.
[0042]
In step S4, the vehicle state sensor 7 detects the traveling state of the vehicle. The traveling state of the vehicle detected by the vehicle state sensor 7 is output to the vehicle state determination unit 8.
[0043]
In step S <b> 5, the vehicle state determination unit 9 determines the traveling state of the vehicle detected by the vehicle state sensor 7. The determination result determined by the vehicle state determination unit 9 is output to the danger determination unit 9.
[0044]
In step S6, the danger determination unit 9 determines whether or not the host vehicle is currently in a dangerous state based on the lane recognized by the lane recognition unit 6 and the determination result determined by the vehicle state determination unit 8. Judging. If it is determined that the host vehicle is in a dangerous state (YES in step S7), the process proceeds to step S8. If it is determined that the host vehicle is not in a dangerous state (NO in step S7), step S10 is performed. Migrate to
[0045]
In step S <b> 8, the danger determination unit 9 drives the alarm device 10 when determining that the host vehicle is in a dangerous state. For example, the alarm device 10 notifies the driver that the host vehicle is in a dangerous state by sounding a warning sound.
[0046]
In step S <b> 9, the risk determination unit 9 drives the vehicle motion control device 11 when determining that the host vehicle is in a dangerous state. The vehicle motion control device 11 controls the motion of the vehicle by, for example, automatically controlling the speed of the vehicle and decelerating the vehicle.
[0047]
Here, when the first image memory 5 and the image processing unit 13 are connected (YES in step S10), the process proceeds to step S11, and the first image memory 5 and the second image memory 14 are connected. If yes (NO in step S10), the process proceeds to step S12.
[0048]
In step S <b> 11, the image processing unit 13 reads image data captured by the foveal lens from the first image memory 5, and performs image processing for removing distortion of the image.
[0049]
In step S <b> 12, when the first image memory 5 and the image processing unit 13 are connected, the second image memory 14 stores the image data from which the distortion has been removed by the image processing unit 13. When the image memory 5 and the second image memory 14 are connected, the image data captured by the foveal lens is read from the first image memory 5 and stored. In step S <b> 13, the image display device 15 displays the image data stored in the second image memory 14.
[0050]
FIG. 4 is a diagram illustrating an example of a captured image in a case where a front state of the vehicle is captured using a single-focus wide-angle lens. The vehicle travels on the road R, and the road R is sandwiched between the white line display LH on the left side of the vehicle and the white line display RH on the right side. As shown in FIG. 4, an image P <b> 1 captured using a single-focus wide-angle lens in the imaging unit 1 has a wide field of view, but it is difficult to recognize a distant road state. Therefore, in this embodiment, a wide field of view is obtained by imaging the front of the vehicle using the above-mentioned foveal lens, and an image capable of accurately recognizing the state of a distant traveling path is captured.
[0051]
FIG. 5 is a diagram illustrating an example of a captured image in a case where the front side of the vehicle is imaged using the foveal lens. The vehicle travels on the road R, and the road R is sandwiched between the white line display LH on the left side of the vehicle and the white line display RH on the right side. The image in the circular area S shown in FIG. 5 is an image enlarged by the foveal lens, and the image outside the area S is an image reduced by the foveal lens. In the actually captured image, the range enlarged by the foveal lens and the range reduced by the foveal lens are not represented by clear boundaries as shown in the region S in FIG. It gradually shifts from a range enlarged by the foveal lens to a range reduced by the foveal lens. As shown in FIG. 5, when the front of the vehicle is imaged using the foveal lens, the captured image P2 is displayed with the central portion enlarged and the peripheral portion reduced, so that a distant white line is displayed. The LH, RH and the road R are displayed in an enlarged manner than the nearby white line display LH, RH and the road R. Therefore, when the foveal lens is used, it is possible to accurately recognize the state of the distant traveling path as compared with the case where the single focus wide-angle lens as shown in FIG. 4 is used. Further, when the foveal lens is used, it is possible to pick up an image at a wide angle, so that it is possible to recognize the state of the near traveling path as compared with the case where the telephoto lens is used.
[0052]
As described above, the light image in the moving direction of the vehicle is formed on the imaging surface so that the image height is large in the region S of the central portion of the captured image and the image height is smaller in the peripheral portion than the region S of the central portion. Since the state of the traveling road on which the vehicle travels is recognized in the central region S where the image height of the image obtained by the image sensor 2 is large, a wide field of view is ensured and the state of the far traveling road is accurately determined. Can be recognized.
[0053]
Further, the lane recognition unit 6 recognizes the lane of the road on which the vehicle travels in the central region S where the image height of the image obtained by the imaging sensor 2 is large. Then, the vehicle state sensor 7 detects the traveling state of the vehicle, and the risk determination unit 9 recognizes the lane recognized by the lane recognition unit 6 and the vehicle traveling state detected by the vehicle state sensor 7. Is determined to be in a dangerous state. When the warning device 10 determines that the vehicle is in a dangerous state, the warning device 10 notifies the driver of the vehicle that the vehicle is in a dangerous state. Therefore, the driver of the vehicle can drive the vehicle safely, and even if the vehicle is in a dangerous state, that is, a state that may cause an accident, this can be prevented in advance.
[0054]
Furthermore, since no optical zooming mechanism is used, durability can be improved, and the size and cost of the apparatus can be reduced.
[0055]
(Second Embodiment)
An imaging apparatus according to a second embodiment of the present invention will be described. As shown in FIG. 5, when the front of the vehicle is imaged using the foveal lens, the image of the central portion of the captured image P2 is enlarged. As a result, unnecessary information such as sky is included in the upper part of the captured image P2, and the image data may not be used effectively. Therefore, in the second embodiment, in order to effectively use the image data, the state of the nearby traveling road is imaged over a wide range by shifting the region S having a large image height upward on the screen.
[0056]
In addition to the function in the first embodiment, the image sensor driving unit 3 shown in FIG. 1 further physically moves the image sensor 2 downward. That is, the imaging sensor driving unit 3 moves the imaging sensor 2 downward, so that the optical axis of the imaging unit 1 moves above the imaging surface of the imaging sensor 2, and as a result, a region where the image height is large. S moves to the upper part of the screen.
[0057]
FIG. 6 is a diagram illustrating an example of a captured image when the position of the imaging sensor is shifted downward with respect to the optical axis of the imaging unit. As shown in FIG. 6, by shifting the position of the image sensor 2 downward with respect to the optical axis of the imaging unit 1, the region S having a large image height moves from the central portion to the upper portion of the captured image P <b> 3, and is far away. The white line displays LH and RH and the road R are displayed larger than the nearby white line displays LH and RH and the road R. Furthermore, when the region S having a large image height moves from the central portion to the upper portion of the captured image P3, the nearby white line displays LH and RH and the road R are displayed in a wider range in the lower portion of the captured image P3. The Rukoto.
[0058]
As described above, by relatively shifting the positions of the imaging unit 1 and the imaging sensor 2, the region S having a large image height in the captured image can be moved upward, without displaying unnecessary information. Image data can be used effectively.
[0059]
In the present embodiment, the image sensor 2 is moved by the image sensor driving unit 3 to relatively shift the positions of the image forming unit 1 and the image sensor 2. However, the present invention is particularly limited to this. Instead, the positions of the imaging unit 1 and the imaging sensor 2 may be relatively shifted by physically moving the imaging unit 1. That is, the imaging apparatus further includes a lens driving unit that shifts the position of the imaging unit relative to the imaging sensor, and the lens driving unit sets the position of the lens constituting the imaging unit 1 in the direction perpendicular to the optical axis. Is moved upward. That is, the lens driving unit moves the imaging unit 1 upward, so that the optical axis of the imaging unit 1 moves above the imaging surface of the imaging sensor 2, and as a result, the region S having a large image height. Will move to the top of the screen.
[0060]
In the present embodiment, since unnecessary information is included in the upper part of the captured image P2 captured by the image sensor 2, the region S having a large image height is moved above the imaging surface of the image sensor 2. However, the present invention is not particularly limited to this. For example, when unnecessary information is included in the lower part of the captured image P2 captured by the image sensor 2, the image sensor 2 is moved upward. The region S having a large image height may be moved below the imaging surface of the imaging sensor 2. That is, either one of the imaging unit 1 and the imaging sensor 2 may be relatively shifted in accordance with the position where unnecessary information is included in the captured image P2 captured by the imaging sensor 2.
[0061]
In addition, since the imaging apparatus according to the present invention uses a foveal lens that is capable of enlarging a distant area and imaging a near area at a wide angle, a plurality of adjacent vehicles in addition to a traveling lane in which the host vehicle is traveling The traveling lane can be imaged. FIG. 7 is a diagram illustrating an example of a captured image when a plurality of traveling lanes are captured. As shown in FIG. 7, in the captured image P4, the first travel lane R1, the second travel lane R2, the third travel lane R3, the fourth travel lane R4, and the fifth travel lane R5 are captured. The first travel lane R1 is a travel lane sandwiched between the first white line display H1 and the second white line display H2, and the second travel lane R2 is a travel lane sandwiched between the second white line display H2 and the third white line display H3. Yes, the third travel lane R3 is a travel lane sandwiched between the third white line display H3 and the fourth white line display H4, and the fourth travel lane R4 is a travel sandwiched between the fourth white line display H4 and the fifth white line display H5. The fifth traveling lane R5 is a traveling lane sandwiched between the fifth white line display H5 and the sixth white line display H6.
[0062]
As shown in FIG. 7, since the foveal lens has a wide field of view, it is possible to recognize the surrounding traveling lanes R1, R2, R4, and R5 in addition to the traveling lane R3 in which the host vehicle is traveling. That is, the lane recognition unit 6 recognizes a plurality of lanes imaged by the imaging sensor 2 and recognizes a travel lane in which the host vehicle is traveling. The lane recognition unit 6 outputs the travel lane recognized by the lane recognition unit 6 to the vehicle guidance device (so-called known car navigation system) as travel lane information. The vehicle guidance device sets the host vehicle to an appropriate driving lane based on traffic jam information, construction information, accident information, and the like of the driving lane in which the host vehicle is driving, and the driving lane information input from the lane recognition unit 6. Induce.
[0063]
In this way, the imaging device using the foveal lens capable of enlarging the distant and imaging the near at a wide angle can capture the state of the traveling path in the moving direction of the vehicle in a wide range. A plurality of lanes can be recognized, and furthermore, the traveling lane in which the host vehicle is currently traveling can be recognized.
[0064]
(Third embodiment)
An imaging apparatus according to a third embodiment of the present invention will be described. Conventionally, a technique for autonomously guiding a vehicle by detecting one white line display by an imaging device is known, and in this conventional imaging device, a single focus lens is mainly used. Therefore, the imaging apparatus according to the third embodiment recognizes the white line display provided on the travel path by the foveal lens and autonomously guides the vehicle.
[0065]
FIG. 8 is a diagram illustrating an example of a captured image in a case where a lane (white line display) of a traveling road is captured using a foveal lens. The arrow YS represents the traveling direction of the vehicle. In the captured image P5 shown in FIG. 8, the white line display HN is enlarged and captured in the region S having a large image height as current information, and the lane recognition unit 6 can accurately recognize the white line display. In addition, since the foveal lens can capture images at a wide angle, the peripheral image in the captured image P5 includes a front and rear of the white line display HN recognized in the region S having a large image height in the central portion. The state of can be detected. That is, the lane recognition unit 6 can recognize the white line display HF on the traveling direction side as future information and the white line display HP on the opposite side to the traveling direction as past information. The vehicle motion control device 11 controls the motion of the vehicle so that the lane recognized by the lane recognition unit 6 is always recognized at the central portion where the image height of the image obtained by the imaging sensor 2 is large. Therefore, the imaging apparatus according to the present embodiment can accurately recognize current information, and can also recognize future information and past information, thereby enabling more advanced guidance control.
[0066]
According to this configuration, the lane recognizing unit 6 recognizes the lane in which the vehicle travels in the central portion where the image height of the image obtained by the imaging sensor 2 is large, and the vehicle motion control device 11 recognizes the lane. Since the movement of the vehicle is controlled so that the lane is always recognized at the central portion where the image height of the image obtained by the imaging sensor 2 is large, the vehicle can be guided autonomously.
[0067]
In this embodiment, the white line display provided on the road is recognized. However, the present invention is not particularly limited to this, and other objects such as curbs provided on the road may be recognized. Good.
[0068]
(Fourth embodiment)
An imaging apparatus according to a fourth embodiment of the present invention will be described. In the fourth embodiment, the imaging apparatus further has a function of driving in the pan direction and the tilt direction.
[0069]
FIG. 9 is a schematic diagram illustrating an appearance of an imaging apparatus that is driven in the pan direction and the tilt direction. The imaging apparatus 100 further includes a pan actuator (corresponding to a pan driving unit) 110 and a tilt actuator (corresponding to a tilt driving unit) 120. The imaging device 100 is configured by any of the imaging devices described in the first to third embodiments, and the pan actuator 110 moves the imaging device 100 in the pan direction (in the arrow Y1 in FIG. 9). In the direction shown). The tilt actuator 120 drives the imaging device 100 in the tilt direction (the direction indicated by the arrow Y2 in FIG. 9).
[0070]
FIG. 10 is a block diagram illustrating a main configuration related to the pan / tilt mechanism of the imaging apparatus. As shown in FIG. 10, the imaging apparatus 100 includes a lane recognition unit 6, a pan / tilt control unit (corresponding to a pan driving amount calculation unit and a tilt driving amount calculation unit) 101, a pan motor driver 102, and a tilting unit. A motor driver 103, a pan driving motor 104, and a tilt driving motor 105 are provided. The pan actuator 110 is composed of a pan motor driver 102 and a pan drive motor 104, and the tilt actuator 120 is composed of a tilt motor driver 103 and a tilt drive motor 105.
[0071]
The pan / tilt control unit 101 calculates the driving amount in the pan direction for capturing the white line display detected in the peripheral portion of the captured image at the central portion of the captured image, and drives for pan with the calculated driving amount. A control signal for driving the motor 104 is output. Further, the pan / tilt control unit 103 calculates a driving amount in the tilt direction for capturing the white line display detected in the peripheral portion of the captured image at the central portion of the captured image, and tilts the calculated driving amount. A control signal for driving the drive motor 105 is output.
[0072]
The pan motor driver 102 is a driver for driving the pan drive motor 106, and drives the pan drive motor 104 based on a control signal output from the pan / tilt control unit 101. The tilt motor driver 103 is a driver for driving the tilt drive motor 105, and drives the tilt drive motor 105 based on a control signal output from the pan / tilt control unit 101.
[0073]
The pan driving motor 104 is a motor for driving the imaging apparatus 100 in the pan direction, and rotates the imaging apparatus 100 left and right about a predetermined axis. The tilt drive motor 105 is a motor for driving the imaging apparatus 100 in the tilt direction, and rotates the imaging apparatus 100 up and down around a predetermined axis.
[0074]
When the guidance marker (in this case, white line display) necessary for autonomous guidance of the vehicle is not captured in the region S having a large image height at the center portion in the captured image captured by the imaging sensor 2, the white line display is accurately recognized. In order to do this, it is necessary to capture the white line display detected in the peripheral portion of the captured image at the central portion of the captured image. In addition, during the autonomous guidance control, when the white line display deviates from the center portion of the captured image captured by the image sensor 2, it is necessary to reposition the white line display to the center portion of the captured image. Therefore, the pan / tilt control unit 101 calculates the driving amount in the pan direction and the tilt direction for capturing the white line display recognized in the peripheral part of the captured image by the lane recognition unit 6 in the central part of the captured image. The calculated drive amount is output to the pan actuator 110 and the tilt actuator 120. The pan actuator 110 rotates the imaging device 100 in the pan direction based on the calculated driving amount, and the tilt actuator 120 rotates the imaging device 100 in the tilt direction based on the calculated driving amount.
[0075]
As described above, the white line display recognized in the peripheral portion of the captured image captured by the image sensor 2 is moved to the central portion of the captured image with a high image height by the pan actuator 110 and the tilt actuator 120. The white line display can always be accurately recognized.
[0076]
Next, a case where a road sign is recognized using the imaging device according to the present invention will be described. FIG. 11 is a diagram illustrating an example of a captured image in the case where a road sign is present in the peripheral portion of the image captured by the image sensor 2, and FIG. 12 illustrates a road in the central portion of the image captured by the image sensor 2. It is a figure which shows an example of the captured image in case a label | marker exists. The road sign includes a regulation sign, an instruction sign, a warning sign, a prohibition sign, a guidance sign, and a guidance sign.
[0077]
In the picked-up image P6 shown in FIG. 11, an intersection name display board HS that represents an intersection name that is a guide sign and a traffic light SG are imaged in the peripheral portion. The intersection name display plate HS displays a place name near the road on which the vehicle is traveling, and in FIG. 11, the place name “Hundred Bird” is displayed as an example. The traffic lights SG are arranged so as to be blinkable in a predetermined color arrangement pattern. In FIG. 11, as an example, the traffic lights SG are arranged so as to be blinkable in blue, yellow, and red color arrangement patterns from the left. In the picked-up image P6 shown in FIG. 11, the intersection name display board SH is picked up in the peripheral portion of the picked-up image P6 having a small image height, so that it is difficult for the driver to visually recognize it. Therefore, as shown in FIG. 12, the intersection name display plate HS is displayed so as to be visible by capturing an image in the region S in the center of the captured image P6 having a large image height.
[0078]
The pan / tilt control unit 101 calculates the driving amount in the pan direction and the tilt direction for capturing the intersection name display plate SH recognized in the peripheral portion of the captured image P6 in the central portion of the captured image, and is calculated. The drive amount is output to the pan actuator 110 and the tilt actuator 120. Note that the method for recognizing the intersection name display board SH in the peripheral part of the captured image is to store the shape and color of the intersection name display board SH in advance in the memory as the extraction condition, and then store each pixel of the captured image P6 and the extraction condition. The intersection name display board SH is recognized by comparing. Moreover, when the intersection name display board SH is installed with the traffic light SG, the traffic light SG may be recognized. In other words, the color arrangement pattern of the traffic light SG is stored in advance in the memory as an extraction condition, and the intersection name display board SH is recognized by comparing each pixel of the captured image P6 with the extraction condition. The pan actuator 110 rotates the imaging device 100 in the pan direction based on the calculated driving amount, and the tilt actuator 120 rotates the imaging device 100 in the tilt direction based on the calculated driving amount.
[0079]
In this way, by capturing the intersection name display plate SH in the region S where the image height of the captured image is large, as shown in FIG. 12, the intersection name display plate SH is enlarged and displayed at the center of the captured image P7. Thus, the driver can easily visually recognize the driver.
[0080]
In the present embodiment, the case of recognizing the intersection name display board SH as an example of a guide sign has been described. However, the present invention is not particularly limited to this, for example, a regulation sign, an instruction sign, a warning sign, a prohibition sign Various road signs such as guidance signs and other guidance signs may be recognized.
[0081]
Next, a case where the above-described imaging device is applied to an autonomous transport vehicle will be described. 2. Description of the Related Art Conventionally, an autonomous guided vehicle that travels along a rail, a magnetic sheet, or the like provided on a traveling path is known. Autonomous transport vehicles using rails and magnetic sheets have a problem that it takes time to change the layout. Therefore, an imaging device using a foveal lens is provided in the autonomous conveyance vehicle, and the imaging device recognizes a guide line such as a white line display provided on the traveling path by the foveal lens, thereby allowing the autonomous conveyance vehicle to self-run. .
[0082]
FIG. 13 is a schematic diagram showing an external appearance of an autonomous guided vehicle to which the imaging apparatus according to the present invention is applied. The autonomous transport vehicle 200 includes two imaging devices 100 in front of it. The two imaging devices 100 are provided with a predetermined convergence angle. Moreover, the white line display H is intermittently provided in the traveling path R on which the autonomous guided vehicle 200 travels. Since the imaging device 100 captures the front white line display H by the foveal lens capable of capturing a wide range, it can recognize the white line display H provided intermittently without missing the autonomous transport vehicle. It can be guided with high accuracy. Also, in general, since the guide marker provided intermittently is easier to change the line than the guide marker provided continuously, the intermittent guide marker (in this case, the white line in this case) By using the display H), it is possible to respond more flexibly to a change in the production line. Furthermore, since the two imaging devices 100 are provided with a predetermined convergence angle, the distance to the guidance marker can be calculated by using the triangulation method, and stereoscopic viewing can be performed. That is, the autonomous guided vehicle 200 can detect an obstacle existing in front of the traveling path R by performing a stereoscopic view, and can avoid the obstacle by detecting the obstacle. Advanced guidance control can be performed.
[0083]
In the present embodiment, the two imaging devices 100 are provided in the autonomous transport vehicle 200, but the present invention is not particularly limited thereto, and only one imaging device 100 may be provided. In this case, the number of parts can be reduced, and the cost for the parts can be reduced.
[0084]
In the present embodiment, the white line display H is intermittently provided on the travel path R. However, the present invention is not particularly limited thereto, and the white line display H may be continuously provided on the travel path R.
[0085]
Further, the imaging device can be applied to a biped walking mobile robot that moves by performing bipedal walking. In other words, the foveal lens provided in the imaging device imitates the function of the human eye, so that the operation closer to a human is realized by applying this imaging device to a biped walking mobile robot. It can be widely applied to autonomous guidance of robots including stereoscopic vision.
[0086]
In the present embodiment, the lane and the guidance marker are recognized using the white line display, but the present invention is not particularly limited to this, and can be applied to lanes of colors other than white. In other words, the lane and the guide marker may be any color as long as the lane and the guide marker can be recognized on the travel path on which the mobile body travels.
[0087]
Furthermore, in this embodiment, the imaging unit 1 forms an optical image in front of the moving body on the imaging surface of the imaging sensor 2, but the present invention is not particularly limited to this, and the rear side of the moving body. An optical image may be formed on the imaging surface of the imaging sensor 2.
[0088]
The specific embodiments described above mainly include inventions having the following configurations.
[0089]
(1) an imaging sensor that photoelectrically converts a light image formed on an imaging surface in which a plurality of pixels are two-dimensionally arranged into an electrical signal at each pixel;
An imaging unit that forms a light image in the moving direction of the moving body on the imaging surface so that the image height is large in the center portion of the captured image and the image height is smaller in the peripheral portion than the center portion;
An image pickup apparatus comprising: a recognition unit that recognizes a state of a traveling path on which the moving body travels at a central portion where an image height of an image obtained by the image pickup sensor is large.
[0090]
(2) The state of the travel path on which the moving body travels is a lane provided on the travel path on which the vehicle travels, and the recognizing unit is the center portion where the image height of the image obtained by the imaging sensor is large. The imaging apparatus according to (1), wherein a lane is recognized.
[0091]
According to this configuration, the state of the travel path on which the moving body travels is the lane provided on the travel path on which the vehicle travels, and the lane is in the center portion where the image height of the image obtained by the imaging sensor is large by the recognition unit. Is recognized.
[0092]
(3) The state of the travel path on which the mobile body travels is a road sign provided on the travel path on which the vehicle travels, and the recognition unit is a central portion where the image height of the image obtained by the imaging sensor is large. The imaging apparatus according to (1), wherein the road sign is recognized.
[0093]
According to this configuration, the state of the travel path on which the moving body travels is a road sign provided on the travel path on which the vehicle travels, and the recognition unit performs a central portion where the image height of the image obtained by the imaging sensor is large. A road sign is recognized.
[0094]
(4) The state of the travel path on which the mobile body travels is a plurality of lanes provided on the travel path on which the vehicle travels, and the recognition unit is a central portion where the image height of the image obtained by the imaging sensor is large. The above-described (1) is characterized by recognizing the plurality of lanes to recognize a currently traveling lane among a plurality of lanes sandwiched between two adjacent lanes of the plurality of lanes. The imaging device described.
[0095]
According to this configuration, the state of the travel path on which the moving body travels is a plurality of lanes provided on the travel path on which the vehicle travels, and the central portion where the image height of the image obtained by the imaging sensor is large by the recognition unit. Thus, a plurality of lanes are recognized, and a currently traveling lane is recognized among a plurality of lanes sandwiched between two adjacent lanes of the plurality of lanes.
[0096]
(5) The moving body is a vehicle that travels on a road, and the recognition unit recognizes a lane of the road on which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large. The vehicle is dangerous based on a vehicle state detection unit that detects a traveling state of the vehicle, a lane recognized by the lane recognition unit, and a vehicle traveling state detected by the vehicle state detection unit. A risk determination unit that determines whether the vehicle is in a state, and further includes a vehicle motion control unit that controls the motion of the vehicle when the risk determination unit determines that the vehicle is in a dangerous state. The imaging apparatus according to (1), characterized in that:
[0097]
According to this configuration, the moving body is a vehicle that travels on the road, and the lane recognition unit recognizes the lane of the road on which the vehicle travels in the central portion where the image height of the image obtained by the imaging sensor is large, and the vehicle state The vehicle driving state is detected by the detection unit, and the vehicle is in a dangerous state based on the lane recognized by the lane recognition unit by the danger determination unit and the vehicle driving state detected by the vehicle state detection unit. If the vehicle motion control unit determines that the vehicle is in a dangerous state, the vehicle motion control unit controls the motion of the vehicle.
[0098]
(6) The moving body is a vehicle traveling on a road, and the recognition unit recognizes a curvature of a lane of the road on which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large. The vehicle is based on a recognition unit, a vehicle speed detection unit that detects the speed of the vehicle, a curvature of the lane recognized by the lane recognition unit, and a vehicle speed detected by the vehicle speed detection unit. A risk determination unit that determines whether or not the lane can be turned, and the vehicle is in a dangerous state for a driver of the vehicle when the risk determination unit determines that the vehicle cannot turn the lane. The imaging apparatus according to claim 1, further comprising a notification unit that notifies
[0099]
According to this configuration, the moving body is a vehicle that travels on the road, and the lane recognition unit recognizes the curvature of the lane of the road on which the vehicle travels in the central portion where the image height of the image obtained by the imaging sensor is large. The vehicle speed is detected by the vehicle speed detection unit, and the vehicle determines the lane based on the curvature of the lane recognized by the lane recognition unit by the danger determination unit and the vehicle speed detected by the vehicle speed detection unit. It is determined whether or not the vehicle can turn, and the notification unit notifies the driver of the vehicle that the vehicle is in a dangerous state when the danger determination unit determines that the vehicle cannot turn the lane.
[0100]
(7) The moving body is a vehicle traveling on a road, and the recognition unit recognizes a curvature of a lane of the road on which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large. The vehicle is based on a recognition unit, a vehicle speed detection unit that detects the speed of the vehicle, a curvature of the lane recognized by the lane recognition unit, and a vehicle speed detected by the vehicle speed detection unit. A risk determination unit that determines whether or not the lane can be turned, and further includes a vehicle motion control unit that decelerates the vehicle when the risk determination unit determines that the vehicle cannot turn the lane. The imaging device according to (1), characterized in that
[0101]
According to this configuration, the moving body is a vehicle that travels on the road, and the lane recognition unit recognizes the curvature of the lane of the road on which the vehicle travels in the central portion where the image height of the image obtained by the imaging sensor is large. The vehicle speed is detected by the vehicle speed detection unit, and the vehicle determines the lane based on the curvature of the lane recognized by the lane recognition unit by the danger determination unit and the vehicle speed detected by the vehicle speed detection unit. It is determined whether or not the vehicle can turn, and when the vehicle motion control unit determines that the vehicle cannot turn the lane, the vehicle is decelerated.
[0102]
【The invention's effect】
According to the first aspect of the present invention, the light image in the moving direction of the moving body is formed on the imaging surface so that the image height is large at the central portion of the captured image and the image height is smaller at the peripheral portion than at the central portion. Since the state of the traveling road on which the moving body travels is recognized at the central part where the image height obtained by the imaging sensor is large, the wide field of view is ensured and the state of the far traveling road is accurately recognized. can do.
[0103]
According to the second aspect of the present invention, an image picked up by the image pickup sensor in which one of the image pickup sensor and the image forming portion is relatively shifted in the direction perpendicular to the optical axis by the drive portion. Since the portion having a large image height is shifted from the center portion, a region having a large image height in the captured image can be moved, and image data can be used effectively without displaying unnecessary information.
[0104]
According to the invention described in claim 3, the driver of the vehicle can drive the vehicle safely, and even if the vehicle is in a dangerous state, that is, a state that may cause an accident, This can be prevented beforehand.
[0105]
According to the fourth aspect of the present invention, the lane recognizing unit recognizes the lane in which the vehicle travels at the central portion where the image height of the image obtained by the imaging sensor is large, and the vehicle motion control unit recognizes the lane recognizing unit. Since the movement of the vehicle is controlled so that the lane thus detected is always recognized at the central portion where the image height of the image obtained by the imaging sensor is large, the vehicle can be guided autonomously.
[0106]
According to the fifth aspect of the present invention, when the guidance marker is recognized in the peripheral portion where the image height of the image obtained by the imaging sensor is small, the image height is increased by driving the imaging device in the pan direction and the tilt direction. Since the guidance marker can be recognized at the large central portion, the guidance marker can be recognized with high accuracy.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of an imaging apparatus according to a first embodiment.
FIG. 2 is a diagram illustrating lens characteristics of a foveal lens.
FIG. 3 is a flowchart illustrating an example of the operation of the imaging apparatus according to the first embodiment illustrated in FIG.
FIG. 4 is a diagram illustrating an example of a captured image in a case where a front state of a vehicle is captured using a single-focus wide-angle lens.
FIG. 5 is a diagram illustrating an example of a captured image in a case where an image of the front of the vehicle is captured using a foveal lens.
FIG. 6 is a diagram illustrating an example of a captured image when the position of the imaging sensor is shifted downward with respect to the optical axis of the imaging unit.
FIG. 7 is a diagram illustrating an example of a captured image when a plurality of traveling lanes are captured.
FIG. 8 is a diagram illustrating an example of a captured image when a white line display of a traveling road is captured using a foveal lens.
FIG. 9 is a schematic diagram illustrating an appearance of an imaging apparatus that is driven in a pan direction and a tilt direction.
FIG. 10 is a block diagram illustrating a main configuration related to a pan / tilt mechanism of the imaging apparatus.
FIG. 11 is a diagram illustrating an example of a captured image in a case where a road sign is present in the peripheral portion of the image captured by the image sensor.
FIG. 12 is a diagram illustrating an example of a captured image in a case where a road sign is present at the center of the image captured by the image sensor.
FIG. 13 is a schematic diagram showing an external appearance of an autonomous guided vehicle to which an imaging apparatus according to the present invention is applied.
[Explanation of symbols]
1 Imaging unit
2 Imaging sensor
3 Imaging sensor driver
4 A / D converter
5 First image memory
6 Lane recognition part
7 Vehicle condition sensor
8 Vehicle state judgment part
9 Risk judgment department
10 Alarm device
11 Vehicle motion control device
12 switch
13 Image processing unit
14 Second image memory
15 Image display device

Claims (5)

An imaging sensor that photoelectrically converts a light image formed on an imaging surface in which a plurality of pixels are two-dimensionally arranged into an electrical signal at each pixel;
An imaging unit that forms a light image in the moving direction of the moving body on the imaging surface so that the image height is large in the center portion of the captured image and the image height is smaller in the peripheral portion than the center portion;
An image pickup apparatus comprising: a recognition unit that recognizes a state of a traveling path on which the moving body travels at a central portion where an image height of an image obtained by the image pickup sensor is large. A drive unit that relatively shifts one of the imaging sensor and the imaging unit in a direction perpendicular to the optical axis;
The image pickup apparatus according to claim 1, wherein a portion where an image height of an image picked up by the image pickup sensor is large is shifted from a center portion. The moving body is a vehicle traveling on a road,
The recognition unit is a lane recognition unit that recognizes a lane of a road on which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large
A vehicle state detection unit for detecting a traveling state of the vehicle;
A risk determination unit that determines whether or not the vehicle is in a dangerous state based on a lane recognized by the lane recognition unit and a running state of the vehicle detected by the vehicle state detection unit;
2. The information processing apparatus according to claim 1, further comprising: a notifying unit that notifies the driver of the vehicle that the vehicle is in a dangerous state when the danger determining unit determines that the vehicle is in a dangerous state. 2. The imaging device according to 2. The lane recognition unit recognizes a lane in which the vehicle travels at a central portion where an image height of an image obtained by the imaging sensor is large,
The vehicle further includes a vehicle motion control unit that controls the motion of the vehicle so that the lane recognized by the lane recognition unit is always recognized at a central portion where an image height of an image obtained by the imaging sensor is large. The imaging device according to claim 3. A pan driving amount calculating unit that calculates a driving amount in the pan direction of the imaging device;
A pan driving unit that drives the imaging apparatus in the pan direction based on the driving amount calculated by the pan driving amount calculating unit;
A drive amount calculation unit for tilt that calculates a drive amount in the tilt direction of the imaging device;
A tilt drive unit that drives the imaging device in a tilt direction based on the drive amount calculated by the tilt drive amount calculation unit;
The recognizing unit recognizes a position of a guide marker provided on a traveling path on which the moving body travels in a peripheral portion where an image height of an image obtained by the imaging sensor is small;
The pan driving amount calculation unit calculates a driving amount in the pan direction for moving the position of the guidance marker recognized by the recognition unit to a central portion where an image height of an image obtained by the imaging sensor is large. And
The tilt drive amount calculation unit calculates a drive amount in the tilt direction for moving the position of the guidance marker recognized by the recognition unit to a central portion where an image height of an image obtained by the imaging sensor is large. The imaging apparatus according to claim 1, wherein:

Images