Imaging device

Abstract

The present invention addresses the problem of providing an imaging device capable of correctly correcting parallax deviation during traveling. An imaging device (10) according to the present invention images a first object (21) and a second object (22) of a preceding vehicle (200) and corrects parallax deviation, the imaging device (10) comprising: a first imaging unit (11); a second imaging unit (12); a first parallax calculating unit (13) that calculates a first parallax on the basis of an image of the first object captured by the first imaging unit and an image of the first object captured by the second imaging unit; a second parallax calculating unit (14) that calculates a second parallax on the basis of an image of the first object captured by the first imaging unit and an image of the second object captured by the second imaging unit; an imaginary baseline length calculating unit (15) that calculates an imaginary baseline length by means of the first parallax, the second parallax, and a baseline length; and a correcting unit (16) that corrects the parallax deviation on the basis of the imaginary baseline length.

Description

Imaging device

[0001] The present invention relates to an imaging device.

[0002] Patent Document 1 states that "This image processing device recognizes three-dimensional position information of a subject based on a pair of image data obtained by capturing images of the vehicle's traveling direction using a stereo camera, and includes a vehicle detection processing unit 1d that detects the positions of a pair of brake lights on the pair of image data, and a parallax offset value correction processing unit 1e that corrects a calculation formula for calculating the distance to the subject based on the parallax of the brake lights for two points at different distances in the traveling direction and the number of pixels separating the pair of left and right brake lights."

[0003] Japanese Patent Application Laid-Open No. 2017-009388

[0004] When applying stereo camera parallax correction technology in emerging countries such as Southeast Asia, there is a risk that the technology, which corrects parallax shift while driving, may not work properly due to poor road conditions. Conventional parallax correction technology uses the presence of white lines on both sides of the road or the brake lights (BL) of a preceding vehicle, but this information cannot be fully utilized in the road environments and congested driving environments of Asia. The present invention has been made in consideration of the above-mentioned points, and its purpose is to provide an imaging device that can properly correct parallax shift while driving.

[0005] The imaging device of the present invention that solves the above problem is an imaging device that corrects parallax displacement by capturing images of a first object and a second object at positions equidistant from the first object and a second object, and includes: a first imaging unit; a second imaging unit that is paired with the first imaging unit and disposed at a position a fixed distance away in a direction intersecting a direction toward the object; a first parallax calculation unit that calculates a first parallax of the first object based on the position of the first object on the image captured by the first imaging unit and the position of the first object on the image captured by the second imaging unit; a second parallax calculation unit that calculates a second parallax that is the parallax between the first object and the second object based on the position of the first object on the image captured by the first imaging unit and the position of the second object on the image captured by the second imaging unit; and a virtual baseline length calculation unit that calculates a virtual baseline length that is the sum of the separation distance from the first object to the baseline length, using the first parallax, the second parallax, and a baseline length that is the separation distance between the first imaging unit and the second imaging unit. a correction unit that corrects parallax displacement between the first imaging unit and the second imaging unit based on the virtual base line length.

[0006] According to the present invention, it is possible to obtain an imaging device that can correctly correct parallax displacement while driving.

[0007] Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. In addition, problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0008] 1 is a functional block diagram of an imaging device according to a first embodiment of the present invention; FIG. 2 is a flowchart illustrating a parallax displacement correction method used by the imaging device according to the first embodiment; FIG. 3 is a diagram illustrating a first parallax; FIG. 4 is a diagram illustrating a second parallax; FIG. 5 is a diagram illustrating the distance measurement principle of a stereo camera and a virtual baseline length; FIG. 6 is a schematic diagram illustrating an example of mounting an imaging device on a vehicle; FIG. 7 is a diagram illustrating the positional relationship between the subject vehicle and a preceding vehicle at the time of the first imaging and the time of the second imaging; FIG. 8 is a flowchart illustrating a parallax displacement correction method used by the imaging device according to the second embodiment;

[0009] Next, an embodiment of the present invention will be described with reference to the drawings.

[0010] First Embodiment FIG. 1 is a functional block diagram of an imaging device according to a first embodiment of the present invention, and FIG. 6 is a schematic diagram showing an example of mounting the imaging device on a vehicle.

[0011] As shown in Fig. 1, the imaging device 10 includes an imaging unit 10a and a control unit 10b that processes images captured by the imaging unit 10a. As shown in Fig. 6, the imaging device 10 is mounted on a vehicle 100 (hereinafter also referred to as the host vehicle 100). The imaging device 10 is a stereo camera that captures images of the area ahead of the vehicle through the windshield, and includes a pair of left and right imaging units 11 and 12 as the imaging unit 10a. The pair of left and right imaging units 11 and 12 are arranged spaced apart from each other in the vehicle width direction, and can capture images of overlapping areas ahead of the host vehicle 100, and can capture images of objects such as a leading vehicle 200 traveling in the same direction ahead of the vehicle.

[0012] In this embodiment, the first imaging unit 11 of the pair of left and right imaging units 11, 12 is disposed in the center in the vehicle width direction, and the second imaging unit 12 is disposed in a direction intersecting the direction toward the object, specifically, at a position a certain distance away from the first imaging unit 11 to the left in the vehicle width direction.

[0013] The control unit 10b of the imaging device 10 detects objects and recognizes object types by, for example, processing images captured by the pair of left and right imaging units 11 and 12. Information on the results of object detection and object type recognition is used for AD (autonomous driving) and ADAS (advanced driver assistance system) of the vehicle 100.

[0014] The control unit 10b of the imaging device 10 then corrects the parallax shift while the vehicle is moving. Parallax shift is an error in parallax that occurs, for example, due to a misalignment of the optical axes of the left and right imaging units 11 and 12. The control unit 10b of the imaging device 10 corrects the parallax shift between the left and right imaging units 11 and 12 by capturing images of the first and second objects at positions equally distant from the first and second objects. The correction of the parallax shift is performed by control processing within the control unit 10b using a correction value.

[0015] The imaging device 10 images the preceding vehicle 200 as an object, and images the left and right rear tires 21, 22 of the preceding vehicle 200 as a first target and a second target, and corrects parallax displacement. The left and right rear tires 21, 22 of the preceding vehicle 200 are considered to be located at equal distances from the vehicle 100. In this embodiment, the imaging device 10 images the same preceding vehicle 200 twice with a time difference, and corrects parallax displacement based on the difference between the images.

[0016] The control unit 10b of the imaging device 10 has an electronic control unit (ECU) equipped with a central processing unit (CPU), a memory unit (RAM and ROM), and an input / output unit (I / O interface), and realizes various functions by reading dedicated software programs installed in the memory unit and executing them on the central processing unit. The control unit 10b has, as internal functions, a first parallax calculation unit 13, a second parallax calculation unit 14, a virtual baseline length calculation unit 15, and a correction unit 16.

[0017] The first parallax calculation unit 13 performs a process of calculating a first parallax w1 [pix] of the first object based on the position of the first object on the image captured by the first imaging unit 11 and the position of the first object on the image captured by the second imaging unit 12. The first parallax calculation unit 13 calculates the first parallax w1 [pix] of the first object by template matching, using, for example, the image captured by the first imaging unit 11 as a base image and the image captured by the second imaging unit 12 as a reference image.

[0018] 3 is a diagram illustrating the first parallax. When the imaging device 10 captures images of the preceding vehicle 200 using the first imaging unit 11 and the second imaging unit 12 as shown in FIG. 3(1), the position of the preceding vehicle 200 captured in the image 31 of the first imaging unit 11 shown in FIG. 3(2) and the image 32 of the second imaging unit 12 shown in FIG. 3(3) is shifted left and right. A first parallax w1 [pix] occurs between the position of the left rear tire (hereinafter also referred to as the first target) 21 captured in the image 31 of the first imaging unit 11 and the position of the first target 21 captured in the image 32 of the second imaging unit 12.

[0019] The second parallax calculation unit 14 performs processing to calculate a second parallax w2 [pix], which is the parallax between the first object 21 and the second object 22, based on the position of the first object 21 on the image captured by the first imaging unit 11 and the position of a right rear tire (hereinafter also referred to as the second object) 22 on the image captured by the second imaging unit 12. For example, the second parallax calculation unit 14 calculates the second parallax w2 [pix] by template matching using the image captured by the first imaging unit 11 as a base image and the image captured by the second imaging unit 12 as a reference image.

[0020] Fig. 4 is a diagram illustrating the second parallax. When the imaging device 10 captures images of the preceding vehicle 200 using the first imaging unit 11 and the second imaging unit 12 as shown in Fig. 4(1), the position of the preceding vehicle 200 captured in the image 41 of the first imaging unit 11 shown in Fig. 4(2) and the image 42 of the second imaging unit 12 shown in Fig. 4(3) is shifted left and right. The second parallax w2 [pix] is defined as the distance between the position of the left rear tire 21, which is the first object captured in the image 41 of the first imaging unit 11, and the position of the right rear tire 22, which is the second object captured in the image 42 of the second imaging unit 12.

[0021] The virtual baseline length calculation unit 15 calculates a virtual baseline length B2 [mm], which is the sum of the baseline length B1 [mm] and the separation distance x [mm] from the first object 21 to the second object 22, using the first parallax w1 [pix], the second parallax w2 [pix], and the baseline length B1 [mm], which is the separation distance between the first image capture unit 11 and the second image capture unit 12. The value of the baseline length B1 [mm] is stored in advance in the control unit 10b.

[0022] The correction unit 16 corrects the parallax shift between the first imaging unit 11 and the second imaging unit 12 based on the virtual baseline length B2 [mm]. In this embodiment, the correction unit 16 calculates a correction value α for correcting the parallax shift based on a first virtual baseline length B2_t1 [mm] that is the virtual baseline length at the first imaging time point t1 and a second virtual baseline length B2_t2 [mm] that is the virtual baseline length at the second imaging time point t2.

[0023] FIG. 5 is a diagram showing the distance measurement principle of a stereo camera and the virtual base line length.

[0024] In this embodiment, it is assumed that the distances z [mm] in the optical axis direction from the first imaging unit 11 and the second imaging unit 12 to the left and right rear tires 21, 22 of the preceding vehicle 200 are the same. As shown in Fig. 5 , the imaging unit 10a of the imaging device 10 has lenses 111, 121 of the first imaging unit 11 and the second imaging unit 12, and imaging elements (e.g., CMOS elements) 112, 122 of the first imaging unit 11 and the second imaging unit 12. In the imaging unit 10a, the focal length of the lenses 111, 121 is f [mm].

[0025] 5, the distance in the optical axis direction from the lenses 111, 121 to the left and right rear tires 21, 22 of the preceding vehicle 200 is z [mm], and the distance between the left rear tire 21 and the right rear tire 22 of the preceding vehicle 200 is x [mm]. The distance between the left and right imaging units 11, 12 is a baseline length B1 [mm], and the length obtained by adding the baseline length B1 [mm] to the distance x [mm] between the left and right rear tires 21, 22 is a virtual baseline length B2 [mm].

[0026] When no parallax displacement occurs, the first parallax w1 [pix] and the second parallax w2 [pix] can be calculated by the following equations (1) and (2): w1 = (f × B1) / (z) (1) w2 = (f × B2) / (z) (2)

[0027] From equation (1), z = (f × B1) / (w1) ... (3) From equation (2), B2 = (w2 × z) / (f) ... (4)

[0028] Substituting equation (3) into equation (4), we get B2 = (w2) × ((f × B1) / (w1)) / (f) = B1 × (w1 / w2) (5)

[0029] The baseline length B1 [mm] and the virtual baseline length B2 [mm] are fixed values ​​and do not change even if the distance z [mm] between the host vehicle 100 and the preceding vehicle 200 changes depending on the time of image capture. Therefore, if no parallax displacement occurs, the virtual baseline length B2 [mm] does not change even if the distance z [mm] changes. For example, the result (6) of calculating the virtual baseline length B2 (10 m) when z = 10 meters [m] and the result (7) of calculating the virtual baseline length B2 (20 m) when z = 20 meters [m] have the relationship (6) B2 (10 m) = (7) B2 (20 m).

[0030] On the other hand, when a parallax displacement Δw occurs, the first parallax w1 [pix] and the second parallax w2 [pix] are expressed by the following equations (1') and (2'): w1 + Δw = (f × B1) / (z) (1') w2 + Δw = (f × B2) / (z) (2')

[0031] From equation (1'), z = (f × B1) / (w1 + Δw) ... (3') From equation (2'), B2 = ((w2 + Δw) × z) / (f) ... (4')

[0032] Substituting equation (3') into equation (4'), we get B2 = (w2 + Δw) × ((f × B1) / (w1 + Δw)) / (f) = B1 × ((w1 + Δw) / (w2 + Δw)) ... (5')

[0033] In a situation where parallax shift occurs, for example, the result (6') of calculating the virtual baseline length B2 (10 m) when z = 10 meters [m] and the result (7') of calculating the virtual baseline length B2 (20 m) when z = 20 meters [m] are not the same, and the relationship (6') ≠ (7') holds.

[0034] In this embodiment, when the relationship (6')≠(7') holds, the correction value α is set using the following equation (8) so that the relationship (6')=(7') holds: B2=B1×(w2+Δw+α) / (w1+Δw+α) (8)

[0035] Fig. 2 is a flowchart illustrating the parallax displacement correction method using the imaging device of the first embodiment, and Fig. 7 is a diagram showing the positional relationship between the host vehicle and the preceding vehicle at the time of the first and second imaging. Reference numeral 700 in Fig. 7 illustrates a situation in which the host vehicle 100 is traveling in a driving lane sandwiched between a white line 71 and a shoulder strip 72, with the preceding vehicle 200 traveling ahead of the host vehicle 100 at a distance z1. Reference numeral 701 illustrates a situation in which the preceding vehicle 200 is traveling ahead of the host vehicle 100 at a distance z2.

[0036] The processes of S101 to S103 in Fig. 2 are for processing an image captured at a first imaging time t1 under the conditions indicated by reference numeral 700 in Fig. 7, and the processes of S105 to S108 are for processing an image captured at a second imaging time t2 under the conditions indicated by reference numeral 701 in Fig. 7. At the first imaging time t1, the distance between the host vehicle 100 and the preceding vehicle 200 is z1, but at the second imaging time t2, the distance between the host vehicle 100 and the preceding vehicle 200 has increased to z2.

[0037] The first parallax calculation unit 13 calculates the first parallax w1_t1 [pix] at the first imaging time t1 (S101), and the second parallax calculation unit 14 calculates the second parallax w2_t1 [pix] at the first imaging time t1 (S102).

[0038] The virtual baseline length calculation unit 15 calculates the first virtual baseline length B2_t1 [mm] using the first parallax w1_t1 [pix] at the first imaging time t1, the second parallax w2_t1 [pix] at the first imaging time t1, and the baseline length B1 [mm] (S103).

[0039] Then, it is determined whether a predetermined time has elapsed since the first image capture time t1 (S104), and if it is determined that the predetermined time has elapsed (YES in S104), the process proceeds to processing the image captured at the second image capture time t2. In this embodiment, the elapse of the predetermined time from the first image capture time t1 is considered to indicate that the inter-vehicle distance between the host vehicle 100 and the preceding vehicle 200 has changed, and image capture is performed at this time point as the second image capture time t2. Note that the distance z to the preceding vehicle 200 may be measured, and the time point at which a change in the inter-vehicle distance from the first image capture time t1 is detected may also be set as the second image capture time t2.

[0040] The first parallax calculation unit 13 calculates the first parallax w1_t2 [pix] at the second imaging time t2 (S105), and the second parallax calculation unit 14 calculates the second parallax w2_t2 [pix] at the second imaging time t2 (S106).

[0041] The virtual baseline length calculation unit 15 calculates the second virtual baseline length B2_t2 [mm] using the first parallax w1_t2 [pix] at the second imaging time t2, the second parallax w2_t2 [pix] at the second imaging time t2, and the baseline length B1 [mm] (S107).

[0042] The correction unit 16 calculates a correction value α that makes the first imaginary base line length B2_t1 [mm] equal to the second imaginary base line length B2_t2 [mm] (S108).

[0043] In the imaging device 10 of this embodiment, the first parallax calculation unit 13 calculates the first parallax w1_t1 [pix] at the first imaging time t1 and the first parallax w1_t2 [pix] at the second imaging time t2, and the second parallax calculation unit 14 calculates the second parallax w2_t1 [pix] at the first imaging time t1 and the second parallax w2_t2 [pix] at the second imaging time t2.

[0044] The virtual baseline length calculation unit 15 calculates a first virtual baseline length B2_t1 [mm] using the first parallax w1_t1 [pix] at the first imaging time point t1, the second parallax w2_t1 [pix] at the first imaging time point t1, and the baseline length B1 [mm], and calculates a second virtual baseline length B2_t2 [mm] using the first parallax w1_t2 [pix] at the second imaging time point t2, the second parallax w2_t2 [pix] at the second imaging time point t2, and the baseline length B1 [mm]. The correction unit 16 sets a value that makes the first virtual baseline length B2_t1 [mm] and the second virtual baseline length B2_t2 [mm] equal as the correction value α.

[0045] Next, the effects of the image capture device 10 of this embodiment will be described. For example, as shown in Fig. 5, the base line length B1, which is the distance between the first image capture unit 11 and the second image capture unit 12, is a fixed value, and the distance x between the left and right rear tires 21, 22 is also a fixed value. Therefore, when no parallax shift occurs in the image capture device 10, the virtual base line length B2 calculated by equation (5) using the first parallax w1, the second parallax w2, and the base line length B1 is also invariant regardless of the distance z from the preceding vehicle 200.

[0046] On the other hand, when parallax shift occurs in the imaging device 10, the virtual baseline length B2 calculated by equation (5') using the first parallax w1, the second parallax w2, the parallax error Δw, and the baseline length B1 changes depending on the distance z to the preceding vehicle 200.

[0047] Therefore, the virtual baseline lengths B2 are calculated for different separation distances, and the first virtual baseline length B2_t1, which is the virtual baseline length B2 at the first image capturing time point t1, is compared with the virtual baseline length B2_t2 at the second image capturing time point t2. If they do not match, the correction value α that satisfies the relationship (6') = (7') is set using the above-described equation (8), and the parallax displacement can be corrected using this correction value α. According to the image capturing device 10 of this embodiment, for example, the parallax displacement can be corrected by capturing an image of the preceding vehicle 200 while the host vehicle 100 is traveling.

[0048] 8 to 10 are flowcharts illustrating a parallax displacement correction method used by an image capture device according to a second embodiment, Fig. 11 is a diagram illustrating a method for calculating the number of pixels corresponding to the baseline length in S205, and Fig. 12 is a diagram illustrating a method for calculating the distance between the left and right rear tires of a preceding vehicle in the image capture device according to the second embodiment. Note that components similar to those in the first embodiment are assigned the same reference numerals, and detailed descriptions thereof will be omitted.

[0049] A distinctive feature of this embodiment is that it is configured to be able to calculate a correction value by measuring the parallax only once, rather than measuring the parallax multiple times at time intervals as in the first embodiment.

[0050] As in the first embodiment, the first imaging unit 11 and the second imaging unit 12 capture images of the first target 21 and the second target 22 of the preceding vehicle 200 at positions that are equal distances away from the first target 21 and the second target 22.

[0051] The first parallax calculation unit 13 calculates the first parallax w1 [pix] based on the position of the first object 21 on the image captured by the first imaging unit 11 and the position of the first object 21 on the image captured by the second imaging unit 12 (S201). In this embodiment, the first parallax w1 [pix] is calculated using the position of one tire of the preceding vehicle 200 as the position of the first object 21.

[0052] Then, the second parallax calculation unit 14 calculates the second parallax w2 [pix] based on the position of the first object 21 on the image captured by the first imaging unit 11 and the position of the second object 22 on the image captured by the second imaging unit 12 (S202). In this embodiment, the second parallax w2 [pix] is calculated by setting the position of one tire of the preceding vehicle 200 as the position of the first object 21 and the position of the other tire as the position of the second object 22.

[0053] The virtual baseline length calculation unit 15 calculates a third virtual baseline length B23 (mm) using the first parallax w1 (pix), the second parallax w2 (pix), and the baseline length B1 (mm) (S203). The third virtual baseline length B23 (mm) is the sum of the baseline length B1 (mm) and the separation distance x (mm) from the first object 21 to the second object 22. The processes from S201 to S203 are the same as the processes from S101 to S103 in the first embodiment.

[0054] Next, the virtual baseline length calculation unit 15 calculates the number of pixels w3 [pix] corresponding to the baseline length B1 on the image captured by the first imaging unit 11 or the second imaging unit 12 (S205). The virtual baseline length calculation unit 15 calculates the number of pixels w3 corresponding to the baseline length B1 based on the distance by which an object captured at the center position of the image captured by either the first imaging unit 11 or the second imaging unit 12 is separated from the center position of the image captured by the other of the first imaging unit 11 and the second imaging unit 12.

[0055] 12, for example, both a third object 23 and a fourth object 24 of the preceding vehicle 200 are captured by capturing images of the preceding vehicle 200 using the first imaging unit 11 and the second imaging unit 12. As shown in FIG. 11, the third object 23 is captured at the center of the image captured by the second imaging unit 12, and the fourth object 24 is captured at the center of the image captured by the first imaging unit 11.

[0056] In the image of the first imaging unit 11, the third object 23 is imaged at a position a predetermined distance to the left of the fourth object 24 at the center position. This predetermined distance is the parallax of the third object 23, and is the number of pixels w3 [pix] corresponding to the base line length B1 on the image captured by the first imaging unit 11 or the second imaging unit 12. In other words, by calculating the parallax of the third object 23, the number of pixels w3 [pix] corresponding to the base line length B1 can be calculated. The parallax of the third object 23 is found by detecting the third object 23 imaged in the image of the first imaging unit 11 from the third object 23 imaged in the image of the second imaging unit 12, and then calculating the number of pixels of a width indicating the distance between the center position of the image of the first imaging unit 11 and the third object 23.

[0057] Next, the virtual baseline length calculation unit 15 calculates the number of pixels between the first object 21 and the second object 22 on the image captured by the first imaging unit 11 or the second imaging unit 12, that is, the number of pixels w4 [pix] between the left and right rear tires 21, 22 of the preceding vehicle 200 (S206).

[0058] Then, by proportional calculation using the number of pixels w3 [pix] corresponding to the base line length B1, the number of pixels w4 between the first object 21 and the second object 22, and the base line length B1, the separation distance between the first object 21 and the second object 22, i.e., the distance x [mm] between the left and right rear tires 21, 22 of the preceding vehicle 200, is calculated by proportional calculation of the following equation (9) (S207). Number of pixels w3 [pix]: Number of pixels w4 [pix] = Base line length B1 [mm]: Distance x [mm] ... (9)

[0059] Then, a fourth virtual base line length B24, which is the length obtained by adding the base line length B1 [mm] to the separation distance x [mm] between the first object 21 and the second object 22, is calculated (S208).

[0060] The correction unit 16 calculates a correction value α that makes the third imaginary base line length B23 [mm] equal to the fourth imaginary base line length B24 (S209). The correction unit 16 uses the correction value α to correct the parallax shift between the first imaging unit 11 and the second imaging unit 12.

[0061] According to the imaging device 10 of this embodiment, it is possible to calculate a correction value by measuring the parallax only once, without the need to measure the parallax multiple times with a time difference as in the first embodiment, and it is possible to correct the parallax shift while the vehicle 100 is traveling.

[0062] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments, and various design modifications can be made without departing from the spirit of the present invention as defined in the claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0063] 10: Imaging device, 11: First imaging unit, 12: Second imaging unit, 13: First parallax calculation unit, 14: Second parallax calculation unit, 15: Virtual baseline length calculation unit, 16: Correction unit, 100: Host vehicle, 200: Leading vehicle (object), B1: Baseline length, B2: Virtual baseline length, x: Separation distance between left and right rear tires 21, 22, α: Correction value, w1: First parallax, w2: Second parallax, z: Distance

Claims

1. An imaging device that images a first object and a second object at positions equidistant from the first object and the second object and corrects parallax shift, comprising: a first imaging unit; a second imaging unit disposed at a fixed distance in a direction intersecting the direction toward the object, paired with the first imaging unit; a first parallax calculation unit that calculates a first parallax of the first object based on the position of the first object in the image captured by the first imaging unit and the position of the first object in the image captured by the second imaging unit; a second parallax calculation unit that calculates a second parallax serving as the parallax between the first object and the second object based on the position of the first object in the image captured by the first imaging unit and the position of the second object in the image captured by the second imaging unit; a virtual baseline length calculation unit that calculates a virtual baseline length, which is the length obtained by adding the baseline length, which is the separation distance between the first imaging unit and the second imaging unit, to the separation distance from the first object to the second object, using the first parallax, the second parallax, and the baseline length; and a correction unit that corrects the parallax shift between the first imaging unit and the second imaging unit based on the virtual baseline length. The imaging device is characterized by comprising the above components.

2. The imaging device according to claim 1, wherein the correction unit calculates a correction value for correcting the parallax shift based on a first virtual baseline length, which is the virtual baseline length at a first imaging time point, and a second virtual baseline length, which is the virtual baseline length at a second imaging time point.

3. The first parallax calculation unit calculates the first parallax at the first imaging time point and the first parallax at the second imaging time point. The second parallax calculation unit calculates the second parallax at the first imaging time point and the second parallax at the second imaging time point. The virtual baseline length calculation unit calculates the first virtual baseline length using the first parallax at the first imaging time point, the second parallax at the first imaging time point, and the baseline length, and calculates the second virtual baseline length using the first parallax at the second imaging time point, the second parallax at the second imaging time point, and the baseline length. The correction unit sets, as a correction value, a value that makes the first virtual baseline length equal to the second virtual baseline length. The imaging device according to claim 2 is characterized by the above configuration.

4. The first parallax calculation unit uses the image captured by the first imaging unit as a reference image, the image captured by the second imaging unit as a reference image, and calculates the first parallax by template matching. The second parallax calculation unit uses the image captured by the first imaging unit as a reference image, the image captured by the second imaging unit as a reference image, and calculates the second parallax by template matching. The imaging device according to claim 1, characterized in that.

5. The virtual baseline length calculation unit calculates a third virtual baseline length, which is the length obtained by adding the baseline length to the separation distance from the first object to the second object, using the first parallax, the second parallax, and the baseline length. Calculate the number of pixels corresponding to the baseline length and the number of pixels between the first object and the second object in the image captured by the first imaging unit or the second imaging unit. By proportional calculation using the number of pixels corresponding to the baseline length, the number of pixels between the first object and the second object, and the baseline length, calculate the separation distance from the first object to the second object. Calculate a fourth virtual baseline length, which is the length obtained by adding the baseline length to the separation distance from the first object to the second object. The correction unit calculates a correction value for correcting the parallax shift based on the third virtual baseline length and the fourth virtual baseline length. The imaging device according to claim 1, characterized in that.

6. The virtual baseline length calculation unit calculates the number of pixels corresponding to the baseline length based on the distance by which the object imaged at the center position of the image captured by either one of the first imaging unit and the second imaging unit is separated from the center position of the image captured by the other of the first imaging unit and the second imaging unit. The imaging device according to claim 5, characterized in that.

Images