Image processing device, image processing method, and program

Abstract

Provided is an image processing device comprising: a detection unit that detects another vehicle present in an image captured by an image sensor mounted to an own vehicle; and a correction unit that corrects a vanishing point of the image. The correction unit corrects the vanishing point on the basis of a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

Description

Image processing device, image processing method, and program

[0001] The present invention relates to an image processing device, an image processing method, and a program.

[0002] Conventionally, there are known techniques for extracting vanishing points from images captured by a camera mounted on a vehicle. For example, Patent Document 1 discloses a technique for calculating motion vectors between multiple images captured by a camera and determining vanishing point coordinates within the images using the calculated motion vectors.

[0003] JP 2012-123751 A

[0004] The technology described in Patent Document 1 calculates a motion vector for each pixel from multiple images captured by a camera using a block matching method, a gradient method, etc. However, such conventional technology may require a large calculation cost and battery consumption in order to improve the accuracy of the vanishing point.

[0005] The present invention has been made in consideration of these circumstances, and one of its objects is to provide an image processing device, an image processing method, and a program that can determine the vanishing point with high accuracy while keeping calculation costs and battery consumption low.

[0006] The image processing device, image processing method, and program according to the present invention employ the following configuration: (1): An image processing device according to one aspect of the present invention includes a detection unit that detects other vehicles present in an image captured by an image sensor mounted on a vehicle, and a correction unit that corrects a vanishing point of the image, and the correction unit corrects the vanishing point based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

[0007] (2): In the above aspect (1), when the first width of the other vehicle traveling in either the left or right direction relative to the host vehicle is smaller than the second width, the correction unit corrects the vanishing point in that direction.

[0008] (3): In the aspect (1) above, when the first width of the other vehicle traveling in either the left or right direction relative to the host vehicle is greater than the second width, the correction unit corrects the vanishing point in the opposite direction to the first width.

[0009] (4): In the above aspect (2) or (3), the correction unit performs the correction when the distance between the subject vehicle and the same other vehicle detected in time series is getting closer.

[0010] (5) In the above aspect (2) or (3), the correction unit performs the correction when the distance between the host vehicle and the other vehicle is equal to or less than a threshold value.

[0011] (6): Another aspect of the image processing method of the present invention is a method in which a computer detects other vehicles present in an image captured by an image sensor mounted on the vehicle and corrects the vanishing point of the image, the correction being based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

[0012] (7): Another aspect of the present invention is a program that causes a computer to detect other vehicles present in an image captured by an image sensor mounted on the vehicle, and correct the vanishing point of the image, the correction being based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

[0013] According to (1) to (7), the vanishing point can be determined with high accuracy while keeping the calculation cost and battery consumption low.

[0014] 1 is a diagram showing an example of a usage environment of a terminal device 100 mounted on a host vehicle M. FIG. 2 is a diagram showing an example of a configuration of the terminal device 100. FIG. 3 is a diagram showing an example of a scene in which a distance estimation unit 120 estimates a distance between the host vehicle M and another vehicle M1. FIG. 4 is a diagram for explaining a method in which the distance estimation unit 120 estimates a distance in a vertical direction between the host vehicle M and another vehicle M1. FIG. 5 is a diagram for explaining a method in which the distance estimation unit 120 estimates a distance in a horizontal direction of another vehicle M1. FIG. 6 is a diagram showing an example of a bird's-eye view and a camera image displayed on a display unit 20. FIG. 7 is a diagram showing another example of a bird's-eye view and a camera image displayed on a display unit 20. FIG. 8 is a diagram for explaining a method in which a vanishing point correction unit 140 determines a vertical position of a vanishing point V. FIG. 9 is a diagram showing an example of a scene in which the vanishing point correction unit 140 corrects a horizontal position of a vanishing point V. FIG. 10 is a diagram for explaining a method in which a vanishing point correction unit 140 corrects a horizontal position of a vanishing point V. FIG. 11 is a diagram for explaining a method in which a vanishing point correction unit 140 corrects a horizontal position of a vanishing point V. FIG. 12 is a flowchart showing an example of a processing flow executed by the terminal device 100.

[0015] Hereinafter, embodiments of an image processing device, an image processing method, and a program of the present invention will be described with reference to the drawings. In this embodiment, the image processing device is, for example, a terminal device 100 such as a smartphone having a camera (image sensor) and a display (display unit). However, the present invention is not limited to such a configuration. The image processing device may be at least a computer device having a computing function, excluding the camera and display, from the configuration of the terminal device 100 described below. In this case, the camera, the display, and the image processing device cooperate to realize the functions of the present invention.

[0016] 1 is a diagram showing an example of a usage environment of a terminal device 100 mounted on a host vehicle M. The host vehicle M may be, for example, a two-wheeled, three-wheeled, or four-wheeled vehicle, and its drive source may be an internal combustion engine such as a diesel engine or a gasoline engine, an electric motor, or a combination of these. The electric motor operates using power generated by a generator connected to the internal combustion engine, or discharged power from a secondary battery or a fuel cell.

[0017] 1, the terminal device 100 is installed on the host vehicle M so as to be able to capture an image of the area ahead of the host vehicle M in the traveling direction of the host vehicle M. The terminal device 100 is held, for example, by an in-vehicle holder (not shown) attached to the dashboard of the host vehicle M, and captures an image of the area ahead of the host vehicle M. As will be described later, a user of the terminal device 100 aligns the terminal device 100 to a predetermined height by following guide lines displayed on the display unit 20. FIG. 1 shows how, as a result of the terminal device 100 being aligned, an image of the vicinity of the top end of another vehicle M1 of the host vehicle M is captured in a substantially horizontal direction with respect to the road surface.

[0018] FIG. 2 is a diagram illustrating an example of the configuration of the terminal device 100. As illustrated in FIG. 2, the terminal device 100 includes, for example, a camera 10, a display unit 20, an object detection unit 110, a distance estimation unit 120, a bird's-eye view generation unit 130, and a vanishing point correction unit 140. The object detection unit 110, the distance estimation unit 120, the bird's-eye view generation unit 130, and the vanishing point correction unit 140 are realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program (software). Some or all of these components may be realized by hardware (including circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a GPU (Graphics Processing Unit), or may be realized by a combination of software and hardware. The program may be stored in advance in a storage device (a storage device having a non-transitory storage medium) such as a hard disk drive (HDD) or flash memory, or may be stored in a removable storage medium (a non-transitory storage medium) such as a DVD or CD-ROM, and installed by inserting the storage medium into a drive device. In the following description, the functions of the object detection unit 110, the distance estimation unit 120, the bird's-eye view generation unit 130, and the vanishing point correction unit 140 may be collectively referred to as the "bird's-eye view application." The bird's-eye view application is installed in the terminal device 100 and is activated, for example, when the user of the terminal device 100 starts driving the vehicle M. The camera 10 is, for example, a digital camera using a solid-state image sensor such as a charge-coupled device (CCD) or a complementary metal oxide semiconductor (CMOS). The display unit 20 is, for example, a display device such as a touch panel or a liquid crystal display.

[0019] [Distance Estimation] The object detection unit 110 detects objects captured in images captured by the camera 10. More specifically, for example, the object detection unit 110 detects objects using a trained model that has been trained to output information such as the presence, position, and type of an object when an image captured by the camera 10 is input. For example, the object detection unit 110 can use this trained model to detect the presence and position of a vehicle or the presence and position of road dividing lines.

[0020] When the object detected by the object detection unit 110 is a vehicle (in this case, a vehicle means a vehicle with two, three, four, or other wheels), the distance estimation unit 120 estimates the distance between the detected vehicle and the camera 10. Fig. 3 is a diagram showing an example of a scene in which the distance estimation unit 120 estimates the distance between the host vehicle M and another vehicle M1. In Fig. 3, the symbol hA indicates the height from the road position corresponding to the bottom edge of the display unit 20 to the vanishing point of the image, and the symbol hB indicates the height from the road position corresponding to the bottom edge of the detected other vehicle M1 to the vanishing point of the image. A method for determining the vanishing point will be described later.

[0021] 4 is a diagram illustrating a method by which distance estimation unit 120 estimates the longitudinal distance between host vehicle M and another vehicle M1. In Fig. 4, symbol IS denotes an image sensor included in camera 10, symbol D denotes a display (an end of camera 10) included in camera 10, symbol O denotes a center of the image sensor, symbol A denotes a position on display D corresponding to road position F shown on the bottom of display D, symbol B denotes a position on display D corresponding to road position G at the rear end of another vehicle M1, symbol C denotes an intersection between the imaging direction of the image sensor and display D, symbol H denotes the height of camera 10 relative to the road surface, symbol DA denotes the distance from the position of image sensor IS to road position F shown on the bottom of display D, and symbol DB denotes the distance from the position of image sensor IS to road position G at the rear end of another vehicle M1.

[0022] In FIG. 4 , triangles OAC and OEF are similar to each other, and triangles OBC and OEG are similar to each other. That is, since L:hA = DA:H and L:hB = DB:H hold true for distance, by rearranging the equations, DA = L×H / hA and DB = L×H / hB are obtained. Therefore, the distance estimation unit 120 can calculate the distance to the road position G of the rear end of the other vehicle M1 using the formula DB = DA×hA / hB. Here, the heights hA and hB to the vanishing point are calculated in advance based on the image captured by the camera 10, and the distance DA, which is independent of the position of the other vehicle M1, can be calculated in advance based on the installation position of the terminal device 100. Note that the above calculation can be performed using only the height information of the vanishing point, without requiring all coordinate information of the vanishing point.

[0023] If the object detected by the object detection unit 110 is a vehicle, the distance estimation unit 120 further estimates the lateral distance to the detected vehicle. Fig. 5 is a diagram for explaining a method for estimating the lateral distance to another vehicle M1 by the distance estimation unit 120. In Fig. 5, the symbol V indicates the vanishing point of the image, the symbol Wb indicates the number of pixels in the lateral direction based on the vanishing point V of the other vehicle M1, and the symbol Wa indicates the number of pixels when the number of pixels Wb is moved to the bottom edge of the display D.

[0024] In Figure 5, triangle VT'T and triangle VS'S are similar to each other. That is, since Wa:hA == Wb:hB holds true for distance, by rearranging, we obtain Wa = Wb x hA / hB. Here, assuming that the total number of pixels Wsc at the bottom edge of display D and the width Wrd of the road on which vehicle M is traveling are known, distance estimation unit 120 can calculate the actual lateral distance W corresponding to the number of pixels Wa using the formula W = Wrd x Wa / Wsc. In this way, distance estimation unit 120 calculates the longitudinal distance and the lateral distance between vehicle M and other vehicle M1.

[0025] [Generation of Bird's-Eye View] The bird's-eye view generation unit 130 generates a bird's-eye view that shows the surrounding situation of the host vehicle M, based on the vertical distance and the horizontal distance between the host vehicle M and the other vehicle M1 estimated by the distance estimation unit 120. The bird's-eye view generation unit 130 displays the generated bird's-eye view on the display unit 20 together with the camera image captured by the camera 10.

[0026] FIG. 6 is a diagram showing an example of a bird's-eye view and a camera image displayed on the display unit 20. The left part of FIG. 6 represents a bird's-eye view, and the right part of FIG. 6 represents a camera image. In FIG. 6, the symbol GL indicates guide lines that prompt the user to set the position of the terminal device 100 when the bird's-eye view app is launched (or when the host vehicle M starts moving). The user places the terminal device 100 inside the vehicle so that the intersection of the displayed guide lines coincides with the vanishing point V (i.e., the point where the road becomes invisible). The intersection point set at this time is loaded into the bird's-eye view app as the initial value of the vanishing point V. FIG. 6 shows an example in which only the host vehicle M is displayed on the bird's-eye view immediately after the bird's-eye view app is launched. Alternatively, the bird's-eye view app may automatically set the vanishing point V, as the intersection of the guide lines, to the center position of the display unit 20.

[0027] 7 is a diagram showing another example of a bird's-eye view and a camera image displayed on the display unit 20. Fig. 7 shows an example in which the distance estimation unit 120 estimates the vertical distances and horizontal distances relative to the host vehicle M for four other vehicles and one motorcycle detected by the object detection unit 110, and the bird's-eye view generation unit 130 generates a bird's-eye view based on the estimated distances. As shown in the left part of Fig. 7, by referring to the bird's-eye view, the driver of the host vehicle M can more reliably recognize other vehicles in relation to the host vehicle M and use this information in his or her driving.

[0028] [Determining the Vanishing Point] As described above, the vanishing point V of the image captured by the camera 10 is utilized to estimate the distance between the host vehicle M and other vehicles. Immediately after the bird's-eye view app is launched, the vanishing point V is automatically calibrated as the center position of the display unit 20 or is manually determined by the user using guide lines. However, once the host vehicle M starts traveling, the vanishing point V of the image may change depending on, for example, the road conditions. Therefore, in order to accurately estimate the distance between the host vehicle M and other vehicles and generate a bird's-eye view, it is necessary to accurately determine and correct the vanishing point V. The method of determining and correcting the vanishing point V executed by the vanishing point correction unit 140 will be described below.

[0029] [Determining the Vertical Position of the Vanishing Point] The vanishing point correction unit 140 sets a bounding box for the other vehicle M1 detected by the object detection unit 110 from the camera image, and determines the vertical position of the vanishing point V as a position a predetermined length (a predetermined percentage of the length) below the upper end of the set bounding box. This is because, in this embodiment, the terminal device 100 including the camera 10 is installed above the front window or rear window of the host vehicle M, as shown in FIG. 1, and therefore, it is assumed that the height of the vanishing point corresponds to the height at which the terminal device 100 is installed.

[0030] FIG. 8 is a diagram illustrating a method for determining the vertical position of the vanishing point V. As shown in FIG. 8, the vanishing point correction unit 140 sets a bounding box BB surrounding the detected other vehicle M1, and determines the vertical position VL of the vanishing point V as a position a predetermined length l below the top end of the set bounding box BB. When multiple other vehicles M1 are detected from the camera image, the vanishing point correction unit 140 selects the other vehicle M1 closest to the host vehicle M as the target vehicle for setting the vertical position VL of the vanishing point V. Note that the vanishing point correction unit 140 may change the predetermined length l depending on whether the host vehicle M on which the terminal device 100 is installed is a two-wheeled vehicle or a four-wheeled vehicle. In this manner, the vertical position of the vanishing point V is determined.

[0031] [Determining the Lateral Position of the Vanishing Point] Meanwhile, with regard to the lateral position of the vanishing point, the vanishing point correction unit 140 corrects the lateral position of the vanishing point V based on the results of comparing the width (actual measurement value) of the bounding box BB of the detected other vehicle M1 with the width (estimated value) of the other vehicle M1 estimated using the previously determined vanishing point VP. Fig. 9 is a diagram showing an example of a situation in which the vanishing point correction unit 140 corrects the lateral position of the vanishing point V. Fig. 9 shows, as an example, a case in which another vehicle M1 has been detected in the lane to the left of the lane in which the host vehicle M is traveling. In Fig. 9, x represents the width of the bounding box BB, and y represents the height of the bounding box BB.

[0032] 9, the vanishing point correction unit 140 performs the lateral position correction of the vanishing point V, which will be described below, when another vehicle M1 traveling to the left or right of the host vehicle M is detected. "When another vehicle M1 traveling to the left or right of the host vehicle M is detected" may mean, for example, when it is detected that the other vehicle M1 is traveling in a lane different from the lane in which the host vehicle M is traveling, or when it is detected that the lateral distance from the host vehicle M estimated based on the camera image is equal to or greater than a predetermined distance.

[0033] Furthermore, the vanishing point correction unit 140 performs the lateral position correction of the vanishing point V, which will be described below, when the distance between the host vehicle M and the same other vehicle M1 detected in time series is decreasing (in other words, when the length of the relative coordinate of the other vehicle M1 with the host vehicle M as the reference is decreasing). This is because the lateral position correction of the vanishing point V, which will be described below, is effective when the host vehicle M and the other vehicle M1 are not too far apart. Therefore, alternatively, the vanishing point correction unit 140 may perform the correction, which will be described below, when the distance between the host vehicle M and the other vehicle M1 is equal to or less than a threshold. Furthermore, for example, the vanishing point correction unit 140 may perform the correction, which will be described below, when the side of the other vehicle M1 is recognized to have an area equal to or greater than a predetermined area.

[0034] Fig. 10 is a diagram for explaining a method for correcting the lateral position of the vanishing point V by the vanishing point correction unit 140. Fig. 10 shows a bird's-eye view obtained by the bird's-eye view generation unit 130 performing bird's-eye view conversion on the image shown in Fig. 9 using the previously determined vanishing point VP. In other words, the vanishing point correction unit 140 estimates the width of the other vehicle M1 from the bird's-eye view obtained using the previously determined vanishing point VP, compares it with the width of the other vehicle M1 actually measured from the bounding box BB, thereby verifying the previously determined vanishing point VP, and corrects the lateral position of the vanishing point VP in accordance with the comparison result.

[0035] 10, angle α represents the angle of the diagonal line P1-P2 of the other vehicle M1, which is formed as a rectangle on the bird's-eye view, with respect to the lane, and angle β represents the angle of the other vehicle M1 when the other vehicle M1 is viewed from the host vehicle M. Furthermore, length a represents the vehicle width of the other vehicle M1, and length b represents the vehicle length of the other vehicle M1, which are set as the average vehicle width and vehicle length, respectively.

[0036] At this time, the vanishing point correction unit 140 can calculate the angles α and β as α = arctan(a / b) and β = arctan(x / y), respectively. Using the above values, the vanishing point correction unit 140 can estimate the width w of the other vehicle M1 as w = L sin(α + β) = √(a2 + b2 sin(α + β). In other words, the estimated width w represents the width of the other vehicle M1 estimated based on the vanishing point VP determined previously.

[0037] Next, the vanishing point correction unit 140 compares the estimated width w of the other vehicle M1 with the width x of the bounding box BB actually measured from the camera image, and corrects the lateral position of the vanishing point VP in accordance with the comparison result. Figure 11 is another diagram for explaining the method of correcting the lateral position of the vanishing point V by the vanishing point correction unit 140. Pattern (a) of the comparison result shown in Figure 11 represents the case where the estimated width w of the other vehicle M1 is greater than the actually measured width x (i.e., w = x + Δ), and pattern (b) of the comparison result shown in Figure 11 represents the case where the estimated width w of the other vehicle M1 is smaller than the actually measured width x (i.e., x = w + Δ).

[0038] As shown in pattern (a), if the estimated width w of the other vehicle M1 is larger than the measured width x, this means that the other vehicle M1 is actually facing more to the left than estimated. In other words, the vanishing point VP used to generate the bird's-eye view should be located further to the left, so the vanishing point correction unit 140 corrects the vanishing point VP to the left from the previously determined vanishing point VP. On the other hand, as shown in pattern (b), if the estimated width w of the other vehicle M1 is smaller than the measured width x, this means that the other vehicle M1 is actually facing more to the right than estimated. In other words, the vanishing point VP used to generate the bird's-eye view should be located further to the right, so the vanishing point correction unit 140 corrects the vanishing point VP to the right from the previously determined vanishing point VP.

[0039] The vanishing point correction unit 140 repeats the determination of the vertical position and correction of the horizontal position of the vanishing point VP described above in a predetermined control cycle (e.g., every few seconds). The determination and correction of the vanishing point according to this embodiment does not require communication with an external server or the like, and does not require complex calculations associated with the use of a trained model or the like. This allows the vanishing point to be determined with high accuracy while keeping calculation costs and battery consumption low. In particular, in this embodiment, since the terminal device 100 is a terminal held by a user, such as a smartphone, keeping calculation costs and battery consumption low can improve convenience for the user.

[0040] 9 to 11 have been described with reference to an example in which the other vehicle M1 is traveling ahead of the host vehicle M on the left side. However, the vanishing point correction unit 140 can also correct the lateral position of the vanishing point VP when the other vehicle M1 is traveling ahead of the host vehicle M on the right side. More specifically, if the estimated width w of the other vehicle M1 traveling ahead of the host vehicle M on the right side is greater than the measured width x, the vanishing point correction unit 140 determines that the other vehicle M1 is actually facing more to the right than estimated. Therefore, the vanishing point correction unit 140 corrects the vanishing point VP to the right from the previously determined vanishing point VP. If the estimated width w of the other vehicle M1 is smaller than the measured width x, this means that the other vehicle M1 is actually facing more to the left than estimated. Therefore, the vanishing point correction unit 140 corrects the vanishing point VP to the left from the previously determined vanishing point VP.

[0041] Next, the flow of processing executed by the terminal device 100 will be described with reference to Fig. 12. Fig. 12 is a flowchart showing an example of the flow of processing executed by the terminal device 100. The processing of the flowchart shown in Fig. 12 is executed, for example, in a predetermined control cycle while the host vehicle M is traveling. The processing of the flowchart shown in Fig. 12 is performed on the premise that the terminal device 100 has been started up and the initial value of the vanishing point VP has been calibrated.

[0042] First, the terminal device 100 acquires a camera image captured by the camera 10 and detects other vehicles (step S100). Next, the terminal device 100 sets a bounding box for the detected other vehicles and sets the vertical position of the vanishing point to a position a predetermined length below the top end of the set bounding box (step S102). Next, the terminal device 100 generates a bird's-eye view based on the camera image and the vanishing point (step S104).

[0043] Next, the terminal device 100 calculates the relative coordinates of the other vehicle relative to the host vehicle on the bird's-eye view in chronological order (step S106). Next, the terminal device 100 determines whether the distance between the host vehicle M and the other vehicle is decreasing based on the relative coordinates calculated in chronological order (step S108). If it is determined that the distance between the host vehicle M and the other vehicle is not decreasing, the terminal device 100 ends the processing of the flowchart. In this case, a vanishing point is obtained in which only the vertical position is corrected.

[0044] On the other hand, if it is determined that the distance between the subject vehicle and the other vehicle is decreasing, the terminal device 100 compares the estimated width w of the other vehicle calculated on the bird's-eye view with the actual measured width w of the bounding box set for the other vehicle (step S110). Next, the terminal device 100 corrects the lateral position of the vanishing point based on the comparison result between the estimated width w and the actual measured width w (step S112). This ends the processing of this flowchart.

[0045] In step S108 of the above flowchart, the terminal device 100 determines whether the distance between the host vehicle M and the other vehicle M1 is decreasing based on the relative coordinates calculated in time series. However, the present invention is not limited to such a configuration, and the terminal device 100 may determine in step S108 whether the distance between the host vehicle M and the other vehicle M1 is equal to or less than a threshold value.

[0046] According to the present embodiment described above, other vehicles present in an image captured by a camera mounted on the vehicle are detected, and the lateral position of the detected other vehicle is corrected based on the results of a comparison between a first width, which is an actual measurement value of the bounding box of the other vehicle, and a second width, which is an estimated value of the width of the other vehicle estimated using the previous vanishing point. This makes it possible to determine the vanishing point with high accuracy while keeping calculation costs and battery consumption low.

[0047] The above-described embodiment can be expressed as follows: An image processing device comprising: a storage medium storing computer-readable instructions; and a processor connected to the storage medium, wherein the processor is configured to: detect another vehicle present in an image captured by an image sensor mounted on a host vehicle by executing the computer-readable instructions; and correct a vanishing point of the image, wherein the correction corrects the vanishing point based on a result of comparing a first width of the detected other vehicle with a second width of the other vehicle estimated using the vanishing point.

[0048] The above describes the form for carrying out the present invention using an embodiment, but the present invention is not limited to such an embodiment, and various modifications and substitutions can be made within the scope that does not deviate from the gist of the present invention.

[0049] REFERENCE SIGNS LIST 10 camera 20 display unit 110 object detection unit 120 distance estimation unit 130 bird's-eye view generation unit 140 vanishing point correction unit

Claims

1. An image processing device comprising: a detection unit that detects other vehicles present in an image captured by an image sensor mounted on a vehicle; and a correction unit that corrects a vanishing point of the image, wherein the correction unit corrects the vanishing point based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

2. The image processing device according to claim 1, wherein the correction unit corrects the vanishing point to the direction when the first width of the other vehicle traveling in either the left or right direction relative to the host vehicle is smaller than the second width.

3. The image processing device according to claim 1, wherein the correction unit corrects the vanishing point to a direction opposite to the first direction when the first width of the other vehicle traveling in either the left or right direction relative to the host vehicle is greater than the second width.

4. The image processing device according to claim 2 or 3, wherein the correction unit performs the correction when the distance between the host vehicle and the same other vehicle detected in time series is decreasing.

5. The image processing device according to claim 2 or 3, wherein the correction unit performs the correction when the distance between the host vehicle and the other vehicle is equal to or less than a threshold value.

6. An image processing method in which a computer detects another vehicle present in an image captured by an image sensor mounted on the vehicle, and corrects the vanishing point of the image, the correction being based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

7. A program that causes a computer to detect other vehicles present in an image captured by an image sensor mounted on a vehicle, and correct the vanishing point of the image, the correction being based on a comparison result between a first width of the detected other vehicle and a second width of the other vehicle estimated using the vanishing point.

Images