JP5938940B2 - Three-dimensional object detection device - Google Patents

Description

  The present invention relates to a three-dimensional object detection device.

  2. Description of the Related Art Conventionally, a technique for imaging a tire at a constant imaging interval is known in order to inspect the state of a tread groove of a tire (see Patent Document 1).

JP 2006-242673 A

  The predetermined area is imaged at a constant imaging interval, the images are aligned based on the two captured images captured in succession, the difference between the two images after alignment is taken, and the portion where the difference has occurred When detecting as a three-dimensional object, there are the following problems. That is, when the host vehicle is traveling on a straight road, the host vehicle hardly moves in the vehicle width direction, so two images obtained continuously even if the captured images are captured at a constant interval. The position of these two images can be aligned by simply shifting the position along the traveling direction. However, when the host vehicle is traveling along a curve, the host vehicle also moves in the vehicle width direction. Therefore, when the captured images are captured at a constant interval, two images obtained in succession are obtained. The amount of displacement in the vehicle width direction between the images becomes large, and there is a problem that the position of these two images cannot be properly aligned only by shifting these two images along the traveling direction. .

  The problem to be solved by the present invention is to provide a three-dimensional object detection device capable of appropriately aligning the positions of images based on continuously captured images and, as a result, appropriately detecting a three-dimensional object. It is to be.

The present invention calculates a travel distance in which a shift amount in the vehicle width direction of each imaging target region in continuously obtained captured images is equal to or less than a predetermined number of pixels in the captured image, and calculates the calculated travel distance as an imaging interval. By setting as , the above problem is solved.

  According to the present invention, for example, when the host vehicle is traveling on a curve, the imaging interval can be set short. Therefore, even when the host vehicle moves in the vehicle width direction during curve traveling, images are continuously captured. The amount of shift in the vehicle width direction between images based on the captured images can be reduced, and by simply shifting the images based on the captured images captured continuously along the traveling direction, these two images Can be aligned with high accuracy.

It is a schematic block diagram of the vehicle carrying the solid object detection apparatus which concerns on 1st Embodiment. It is a top view which shows the driving state of the vehicle of FIG. It is a block diagram which shows the detail of the computer which concerns on 1st Embodiment. It is a figure for demonstrating the outline | summary of the process of the position alignment part which concerns on 1st Embodiment, (a) is a top view which shows the movement state of a vehicle, (b) is an image which shows the outline | summary of position alignment. It is the schematic which shows the mode of the production | generation of the difference waveform by the solid-object detection part which concerns on 1st Embodiment. It is a figure which shows the small area | region divided | segmented by the solid-object detection part which concerns on 1st Embodiment. It is a figure which shows an example of the histogram obtained by the solid-object detection part which concerns on 1st Embodiment. It is a figure which shows the weighting by the solid-object detection part which concerns on 1st Embodiment. It is a figure which shows the other example of the histogram obtained by the solid-object detection part which concerns on 1st Embodiment. It is a flowchart which shows the adjacent vehicle detection method which concerns on 1st Embodiment. It is a figure for demonstrating the setting method of the imaging interval of the camera which concerns on this embodiment. It is a schematic block diagram of the vehicle carrying the solid-object detection apparatus which concerns on 2nd Embodiment. It is a block diagram which shows the detail of the computer which concerns on 2nd Embodiment. It is a flowchart which shows the adjacent vehicle detection method which concerns on 2nd Embodiment.

<< First Embodiment >>
FIG. 1 is a schematic configuration diagram of a vehicle equipped with a three-dimensional object detection device 1 according to the first embodiment. The three-dimensional object detection device 1 according to the present embodiment is intended to detect another vehicle (hereinafter also referred to as an adjacent vehicle V2) existing in an adjacent lane that may be contacted when the host vehicle V1 changes lanes. To do. As shown in FIG. 1, the three-dimensional object detection device 1 according to the present embodiment includes a camera 10, a vehicle speed sensor 20, a calculator 30, and a yaw rate sensor 40.

  As shown in FIG. 1, the camera 10 is attached to the vehicle V <b> 1 so that the optical axis is at an angle θ from the horizontal to the lower side at the position of the height h behind the host vehicle V <b> 1. The camera 10 captures an image of a predetermined area in the surrounding environment of the host vehicle V1 from this position. The vehicle speed sensor 20 detects the traveling speed of the host vehicle V1, and calculates the vehicle speed from the wheel speed detected by, for example, a wheel speed sensor that detects the rotational speed of the wheel. The computer 30 detects an adjacent vehicle existing in an adjacent lane behind the host vehicle. The yaw rate sensor 40 detects the amount of change per unit time of the rotation angle (yaw angle) around the vertical axis passing through the center of gravity of the vehicle, that is, the yaw rate (yaw angular velocity).

  FIG. 2 is a plan view showing a traveling state of the host vehicle V1 of FIG. As shown in the figure, the camera 10 images the vehicle rear side at a predetermined angle of view a. At this time, the angle of view a of the camera 10 is set to an angle of view at which the left and right lanes (adjacent lanes) can be imaged in addition to the lane in which the host vehicle V1 travels.

  FIG. 3 is a block diagram showing details of the computer 30 of FIG. In FIG. 3, the camera 10, the vehicle speed sensor 20, and the yaw rate sensor 40 are also illustrated in order to clarify the connection relationship.

  As shown in FIG. 3, the computer 30 includes an imaging interval setting unit 31, a viewpoint conversion unit 32, a positioning unit 33, a three-dimensional object detection unit 34, and a three-dimensional object determination unit 35. Below, each structure is demonstrated.

  The imaging interval setting unit 31 sets the imaging interval of the camera 10 based on the vehicle speed of the host vehicle detected by the vehicle speed sensor 20 and the yaw rate detected by the yaw rate sensor 40. The details of the method for setting the imaging interval by the imaging interval setting unit 31 will be described later.

  The viewpoint conversion unit 32 inputs captured image data of a predetermined area captured by the camera 10 and converts the input captured image data into a bird's-eye image data in a bird's-eye view state. The state viewed from a bird's-eye view is a state viewed from the viewpoint of a virtual camera looking down from above, for example, vertically downward. This viewpoint conversion can be executed as described in, for example, Japanese Patent Application Laid-Open No. 2008-219063. The viewpoint conversion of captured image data to bird's-eye view image data is based on the principle that a vertical edge peculiar to a three-dimensional object is converted into a straight line group passing through a specific fixed point by viewpoint conversion to bird's-eye view image data. This is because a planar object and a three-dimensional object can be distinguished if used.

  The alignment unit 33 sequentially inputs the bird's-eye view image data obtained by the viewpoint conversion of the viewpoint conversion unit 32 and aligns the positions of the inputted bird's-eye view image data at different times. 4A and 4B are diagrams for explaining the outline of the processing of the alignment unit 33. FIG. 4A is a plan view showing the moving state of the host vehicle V1, and FIG. 4B is an image showing the outline of the alignment.

As shown in FIG. 4 (a), the host vehicle V1 of the current time is located in P _1, one unit time before the vehicle V1 is located in the P _{1 '.} Further, there is a parallel running state with the vehicle V1 is located is adjacent vehicle V2 laterally after the vehicle V1, located in P ₂ adjacent vehicle V2 is the current time, one unit time before the adjacent vehicle V2 is P ₂ Suppose it is located at '. Furthermore, it is assumed that the host vehicle V1 has moved a distance d at one time. Note that “one hour before” may be a past time for a predetermined time (for example, one control cycle) from the current time, or may be a past time for an arbitrary time.

In this state, the bird's-eye view image PB _t at the current time is as shown in Figure 4 (b). In the bird's-eye view image PB _t, becomes a rectangular shape for the white line drawn on the road surface, but a relatively accurate is a plan view state, the adjacent vehicle V2 (position P ₂₎ is tilting occurs. Similarly, for the bird's-eye view image PB _{t-1 one} hour before, the white line drawn on the road surface has a rectangular shape and is in a state of being relatively accurately viewed in plan, but the adjacent vehicle V2 (position P _2). ') Will fall down. As described above, the vertical edges of solid objects (including the edges that rise in the three-dimensional space from the road surface in addition to the vertical edges in the strict sense) are straight lines along the collapse direction by the viewpoint conversion processing to bird's-eye view image data. This is because the plane image on the road surface does not include a vertical edge, but such a fall does not occur even when the viewpoint is changed.

The alignment unit 33 performs alignment of the bird's-eye view images PB _t and PB _t−1 as described above on the data. At this time, the alignment unit 33 offsets the bird's-eye view image PB _t-1 at the previous time and matches the position with the bird's-eye view image PB _{t at the} current time. The image on the left side and the center image in FIG. 4B show a state that is offset by the movement distance d ′. This offset amount d ′ is a movement amount on the bird's-eye view image data corresponding to the actual movement distance d of the host vehicle V1 shown in FIG. 4 (a). It is determined based on the time until the time.

In addition, after the alignment, the alignment unit 33 takes the difference between the bird's-eye view images PB _t and PB _t−1 and generates data of the difference image PD _t . Here, in the present embodiment, the alignment unit 33 converts the pixel value difference between the bird's-eye view images PB _t and PB _t−1 to an absolute value in order to cope with a change in the illuminance environment, and the absolute value is a predetermined value. By setting the pixel value of the difference image PD _t to “1” when the threshold value is equal to or greater than the threshold th, and setting the pixel value of the difference image PD _t to “0” when the absolute value is less than the predetermined threshold th, Data of the difference image PD _t as shown on the right side of FIG. 4B can be generated.

Returning to FIG. 3, the three-dimensional object detection unit 34 detects a three-dimensional object based on the data of the difference image PD _t shown in FIG. At this time, the three-dimensional object detection unit 34 also calculates the movement distance of the three-dimensional object in the real space. In detecting the three-dimensional object and calculating the movement distance, the three-dimensional object detection unit 34 first generates a differential waveform.

In generating the difference waveform, the three-dimensional object detection unit 34 sets a detection region in the difference image PD _t . The three-dimensional object detection device 1 of the present example is intended to calculate a movement distance for an adjacent vehicle that may be contacted when the host vehicle V1 changes lanes. For this reason, in this example, as shown in FIG. 2, rectangular detection areas A1, A2 are set on the rear side of the host vehicle V1. Such detection areas A1, A2 may be set from a relative position with respect to the host vehicle V1, or may be set based on the position of the white line. When setting the position of the white line as a reference, the three-dimensional object detection device 1 may use, for example, an existing white line recognition technique.

  Further, as shown in FIG. 2, the three-dimensional object detection unit 34 recognizes the sides (sides along the traveling direction) of the set detection areas A1 and A2 on the own vehicle V1 side as the ground lines L1 and L2. In general, the ground line means a line in which the three-dimensional object contacts the ground. However, in the present embodiment, the ground line is set as described above, not a line in contact with the ground. Even in this case, from experience, the difference between the ground line according to the present embodiment and the ground line obtained from the position of the original adjacent vehicle V2 is not too large, and there is no problem in practical use.

FIG. 5 is a schematic diagram illustrating how the three-dimensional object detection unit 34 generates a differential waveform. As illustrated in FIG. 5, the three-dimensional object detection unit 34 calculates a differential waveform from a portion corresponding to the detection areas A <b> 1 and A <b> 2 in the difference image PD _t (right diagram in FIG. 4B) calculated by the alignment unit 33. DW _t is generated. At this time, the three-dimensional object detection unit 34 generates a differential waveform DW _t along the direction in which the three-dimensional object falls due to viewpoint conversion. In the example shown in FIG. 5, only the detection area A1 is described for convenience, but the difference waveform DW _t is generated for the detection area A2 in the same procedure.

Specifically, the three-dimensional object detection unit 34 first defines a line La on the direction the three-dimensional object collapses on data of the difference image PD _t. Then, the three-dimensional object detection unit 34 counts the number of difference pixels DP indicating a predetermined difference on the line La. In the present embodiment, the difference pixel DP indicating the predetermined difference is expressed by the pixel value of the difference image PD _t as “0” and “1”, and the pixel indicating “1” is counted as the difference pixel DP. .

  The three-dimensional object detection unit 34 counts the number of difference pixels DP and then obtains an intersection CP between the line La and the ground line L1. Then, the three-dimensional object detection unit 34 associates the intersection point CP with the count number, determines the horizontal axis position based on the position of the intersection point CP, that is, the position on the vertical axis in the right diagram of FIG. The axis position, that is, the position on the left-right axis in the right diagram of FIG. 5 is determined and plotted as the number of counts at the intersection CP.

Similarly, the three-dimensional object detection unit 34 defines lines Lb, Lc... In the direction in which the three-dimensional object falls, counts the number of difference pixels DP, and determines the horizontal axis position based on the position of each intersection CP. Then, the vertical axis position is determined from the count number (number of difference pixels DP) and plotted. The three-dimensional object detection unit 34 generates the differential waveform DW _t as shown in the right diagram of FIG.

Here, the difference pixel PD on the data of the difference image PD _t is a pixel that has changed in the images at different times, in other words, a location where a three-dimensional object exists. For this reason, the difference waveform DW _t is generated by counting the number of pixels along the direction in which the three-dimensional object collapses and performing frequency distribution at the location where the three-dimensional object exists. In particular, since the number of pixels is counted along the direction in which the three-dimensional object falls, the differential waveform DW _t is generated from the information in the height direction for the three-dimensional object.

As shown in the left diagram of FIG. 5, the line La and the line Lb in the direction in which the three-dimensional object collapses have different distances overlapping the detection area A1. For this reason, if the detection area A1 is filled with the difference pixels DP, the number of difference pixels DP is larger on the line La than on the line Lb. For this reason, when the three-dimensional object detection unit 34 determines the vertical axis position from the count number of the difference pixels DP, the three-dimensional object detection unit 34 is normalized based on the distance at which the lines La and Lb in the direction in which the three-dimensional object falls and the detection area A1 overlap. Turn into. As a specific example, in the left diagram of FIG. 5, there are six difference pixels DP on the line La, and there are five difference pixels DP on the line Lb. Therefore, in determining the vertical axis position from the count number in FIG. 5, the three-dimensional object detection unit 34 normalizes the count number by dividing it by the overlap distance. Thus, as shown in the difference waveform DW _t, the line La on the direction the three-dimensional object collapses, the value of the differential waveform DW _t corresponding to Lb is substantially the same.

After the generation of the differential waveform DW _t , the three-dimensional object detection unit 34 detects the adjacent vehicle existing in the adjacent lane based on the generated differential waveform DW _t . Three-dimensional object detection unit 34 calculates the moving distance in comparison with the differential waveform DW _t-1 of the previous differential waveform DW _t and a time instant at the current time. That is, the three-dimensional object detection unit 34 calculates the movement distance from the time change of the differential waveforms DW _t and DW _t−1 .

More specifically, the three-dimensional object detection unit 34 divides the differential waveform DW _t into a plurality of small regions DW _{t1 to} DW _tn (n is an arbitrary integer equal to or greater than 2) as shown in FIG. FIG. 6 is a diagram illustrating the small areas DW _{t1 to} DW _tn divided by the three-dimensional object detection unit 34. The small regions DW _{t1 to} DW _tn are divided so as to overlap each other, for example, as shown in FIG. For example, the small area DW _t1 and the small area DW _t2 overlap, and the small area DW _t2 and the small area DW _t3 overlap.

Next, the three-dimensional object detection unit 34 obtains an offset amount (amount of movement of the differential waveform in the horizontal axis direction (vertical direction in FIG. 6)) for each of the small regions DW _{t1 to} DW _tn . Here, the offset amount is determined from the difference between the differential waveform DW _t in the difference waveform DW _t-1 and the current time before one unit time (distance in the horizontal axis direction). At this time, three-dimensional object detection unit 34, for each small area _DW t1 _{~DW tn,} when moving the differential waveform _{DW t1} before one unit time in the horizontal axis direction, the differential waveform DW _t at the current time The position where the error is minimized (the position in the horizontal axis direction) is determined, and the amount of movement in the horizontal axis between the original position of the differential waveform DW _t-1 and the position where the error is minimized is obtained as an offset amount. Then, the three-dimensional object detection unit 34 counts the offset amount obtained for each of the small regions DW _{t1 to} DW _tn and forms a histogram.

FIG. 7 is a diagram illustrating an example of a histogram obtained by the three-dimensional object detection unit 34. As shown in FIG. 7, the offset amount, which is the amount of movement that minimizes the error between each of the small areas DW _{t1 to} DW _tn and the differential waveform DW _t-1 one time before, has some variation. For this reason, the three-dimensional object detection unit 34 forms a histogram of offset amounts including variations, and calculates a movement distance from the histogram. At this time, the three-dimensional object detection unit 34 calculates the movement distance of the adjacent vehicle from the maximum value of the histogram. That is, in the example illustrated in FIG. 7, the three-dimensional object detection unit 34 calculates the offset amount indicating the maximum value of the histogram as the movement distance τ ^* . As described above, in this embodiment, even if the offset amount varies, it is possible to calculate a more accurate movement distance from the maximum value. The moving distance τ ^* is a relative moving distance of the adjacent vehicle with respect to the own vehicle. For this reason, when calculating the absolute movement distance, the three-dimensional object detection unit 34 calculates the absolute movement distance based on the obtained movement distance τ ^* and the signal from the vehicle speed sensor 20.

Thus, in the present embodiment, a one-dimensional waveform is obtained by calculating the moving distance of the three-dimensional object from the offset amount of the differential waveform DW _t when the error of the differential waveform DW _t generated at different times is minimized. Thus, the movement distance is calculated from the offset amount of the information, and the calculation cost can be suppressed in calculating the movement distance. In addition, by dividing the differential waveform DW _t generated at different times into a plurality of small regions DW _{t1 to} DW _tn , a plurality of waveforms representing the respective locations of the three-dimensional object can be obtained. Since the offset amount can be obtained for each of the positions, and the movement distance can be obtained from a plurality of offset amounts, the calculation accuracy of the movement distance can be improved. Further, in the present embodiment, by calculating the moving distance of the three-dimensional object from the time change of the differential waveform DW _t including the information in the height direction, compared with a case where attention is paid only to one point of movement, Since the detection location before the time change and the detection location after the time change are specified including information in the height direction, it is likely to be the same location in the three-dimensional object, and the movement distance is calculated from the time change of the same location, and the movement Distance calculation accuracy can be improved.

Incidentally, the three-dimensional object detection unit 34 Upon histogram is weighted for each of a plurality of small areas _DW t1 _{~DW tn,} the offset amount determined for each small area _DW t1 _{~DW tn} histogram of counts in response to the weight May be. FIG. 8 is a diagram illustrating weighting by the three-dimensional object detection unit 34.

As shown in FIG. 8, the small region DW _m (m is an integer of 1 to n−1) is flat. That is, in the small area DW _m , the difference between the maximum value and the minimum value of the number of pixels indicating a predetermined difference is small. Three-dimensional object detection unit 34, to reduce weight for such small area DW _m. This is because the flat small area DW _m has no characteristics and is likely to have a large error in calculating the offset amount.

On the other hand, the small area DW _{m + k} (k is an integer equal to or less than nm) is rich in undulations. That is, in the small area DW _m , the difference between the maximum value and the minimum value of the number of pixels indicating a predetermined difference is large. Three-dimensional object detection unit 34 increases the weight for such small area DW _m. This is because the small region DW _{m + k} rich in undulations is characteristic and there is a high possibility that the offset amount can be accurately calculated. By weighting in this way, the calculation accuracy of the movement distance can be improved.

Although dividing the differential waveform DW _t into a plurality of small areas DW _t1 ~DW _tn in the above embodiment in order to improve the calculation accuracy of the moving distance, if the calculation accuracy of the moving distance is not less required small regions DW _t1 It is not necessary to divide into ~ DW _tn . In this case, the three-dimensional object detection unit 34 calculates the movement distance from the offset amount of the differential waveform DW _t when the error between the differential waveform DW _t and the differential waveform DW _t−1 is minimized. That is, the method for obtaining the offset amount of the difference waveform DW _t in the difference waveform DW _t-1 and the current time before one unit time is not limited to the above disclosure.

  In the present embodiment, the three-dimensional object detection unit 34 obtains the moving speed of the host vehicle V1 (camera 10), and obtains the offset amount for the stationary object from the obtained moving speed. After obtaining the offset amount of the stationary object, the three-dimensional object detection unit 34 ignores the offset amount corresponding to the stationary object among the maximum values of the histogram and calculates the moving distance of the adjacent vehicle.

  FIG. 9 is a diagram illustrating another example of a histogram obtained by the three-dimensional object detection unit 34. When there is a stationary object in addition to the adjacent vehicle within the angle of view of the camera 10, two maximum values τ1 and τ2 appear in the obtained histogram. In this case, one of the two maximum values τ1, τ2 is the offset amount of the stationary object. Therefore, the three-dimensional object detection unit 34 obtains the offset amount for the stationary object from the moving speed, ignores the maximum value corresponding to the offset amount, and calculates the moving distance of the three-dimensional object by using the remaining maximum value. To do. Thereby, the situation where the calculation accuracy of the moving distance of a solid object falls by a stationary object can be prevented.

  Even if the offset amount corresponding to the stationary object is ignored, when there are a plurality of maximum values, it is assumed that there are a plurality of adjacent vehicles within the angle of view of the camera 10. However, it is extremely rare that a plurality of adjacent vehicles exist in the detection areas A1 and A2. For this reason, the three-dimensional object detection unit 34 stops calculating the movement distance. Thereby, in the present embodiment, it is possible to prevent a situation in which an erroneous movement distance having a plurality of maximum values is calculated.

  The three-dimensional object determination unit 35 determines whether the three-dimensional object detected by the three-dimensional object detection unit 34 is another vehicle that exists in the adjacent lane. The details of the determination method by the three-dimensional object determination unit 35 will be described later.

  Next, the adjacent vehicle detection process according to the first embodiment will be described. FIG. 10 is a flowchart illustrating the adjacent vehicle detection process according to the first embodiment.

  As shown in FIG. 10, first, in step S <b> 101, acquisition of vehicle speed information of the host vehicle is performed by the imaging interval setting unit 31. Specifically, the vehicle speed of the host vehicle is detected by the vehicle speed sensor 20, and information on the detected vehicle speed of the host vehicle is acquired by the imaging interval setting unit 31. In step S102, the imaging interval setting unit 31 acquires the yaw rate information of the host vehicle. Specifically, the yaw rate of the host vehicle is detected by the yaw rate sensor 40, and information on the detected yaw rate is acquired by the imaging interval setting unit 31.

  Next, in step S103, the imaging interval setting unit 31 calculates the amount of movement of the host vehicle in the vehicle width direction within a predetermined time. Specifically, the imaging interval setting unit 31 moves in the vehicle width direction of the host vehicle within a predetermined time based on the vehicle speed information of the host vehicle acquired in step S101 and the yaw rate information of the host vehicle acquired in step S102. Is calculated as information representing the turning state of the host vehicle. As the yaw rate of the host vehicle increases, the moving speed of the host vehicle in the vehicle width direction increases and the amount of movement of the host vehicle in the vehicle width direction during a predetermined time increases. For example, the host vehicle travels on a curve. In some cases, when turning right or left, or when changing lanes, the amount of movement of the host vehicle in the vehicle width direction tends to increase.

  In step S <b> 104, the imaging interval setting unit 31 sets the imaging interval of the camera 10. Specifically, the imaging interval setting unit 31 captures each of the two bird's-eye view images obtained continuously based on the amount of movement of the host vehicle in the vehicle width direction within the predetermined time calculated in step S103. A travel distance in which the shift amount in the vehicle width direction of the target region is equal to or less than a predetermined number of pixels in the resolution in the bird's-eye view image is calculated, and the calculated travel distance is set as an imaging interval of the camera 10. Note that the predetermined number of pixels is the number of pixels set in advance through experiments or the like, and when two bird's-eye view images obtained at different times successively are aligned, the two bird's-eye view images If the shift amount in the vehicle width direction of the imaging target area is within the shift amount range corresponding to the predetermined number of pixels, the two bird's-eye view images are simply shifted in the traveling direction and aligned. The number of pixels that can match the positions of these two bird's-eye view images with high accuracy. That is, in order to perfectly align the positions of the two bird's-eye view images obtained in succession, the two bird's-eye view images are not only shifted in the traveling direction, but the two bird's-eye view images are Although it is necessary to shift and rotate in the direction, if the shift amount in the vehicle width direction of the imaging target region of these two bird's-eye view images is within the range of the shift amount corresponding to this predetermined number of pixels, practically, A solid object can be detected based on the difference image obtained by shifting the positions of the two bird's-eye view images in the traveling direction.

Here, FIG. 11 is a diagram for explaining a method of setting the imaging interval by the imaging interval setting unit 31, and FIG. 11A shows a scene where the host vehicle is traveling on a straight road. ), (C) each illustrate a scene in which the host vehicle is traveling along a curve. For example, as shown in FIG. 11A, when the host vehicle is traveling on a straight road, the yaw rate of the host vehicle is small, and therefore the amount of movement of the host vehicle in the vehicle width direction within a predetermined time is also small. Therefore, as shown in FIG. 11 (A), the imaging interval setting unit 31 is used when the host vehicle shown in FIG. 11 (B) is running on a curve when the host vehicle is running on a straight road. Compared to the above, the imaging interval of the camera 10 is set to a relatively large equidistant interval d ₁ . Specifically, the imaging interval setting unit 31 includes an imaging target region of the n-th bird's-eye view image obtained based on the amount of movement of the host vehicle in the vehicle width direction within a predetermined time and the vehicle speed of the host vehicle. , travel distance deviation amount in the vehicle width direction between the imaging target region of the bird's-eye view image obtained in the (n + 1) th becomes the resolution of the bird's-eye view image a predetermined number of pixels (e.g. the number of pixels corresponding to a predetermined distance w ₁₎ less d ₁ is calculated, and the imaging interval of the camera 10 is set to the equidistant interval d ₁ . Thus, in the scene shown in FIG. 11 (A), so that the imaging captured image is performed for each shooting distance d ₁ by the camera 10.

Further, for example, as shown in FIG. 11B, when the host vehicle is traveling along a curve, the yaw rate of the host vehicle increases, and thus the amount of movement of the host vehicle in the vehicle width direction within a predetermined time period. Also grows. Therefore, as shown in FIG. 11B, the imaging interval setting unit 31 sets the imaging interval of the camera 10 to an equidistant interval d smaller than the equidistant interval d ₁ when the host vehicle is traveling on a curve. Set to ₂ . Specifically, the imaging interval setting unit 31 is based on the amount of movement of the host vehicle in the vehicle width direction within a predetermined time, and the nth bird's-eye view obtained based on the nth bird's-eye view image. The amount of shift in the vehicle width direction between the imaging target area of the visual image and the imaging target area of the (n + 1) th bird's-eye-view image is a pixel corresponding to a predetermined number of pixels (for example, a predetermined distance w ₁₎ in the resolution of the bird's-eye-view image The traveling distance d ₂ that is equal to or less than the number) is calculated, and the imaging interval of the camera 10 is set to an equidistant interval d ₂ that is smaller than the equidistant interval d ₁ . Thus, in the scene shown in FIG. 11 (B), so that the imaging captured image is performed for each shooting distance d ₂ by the camera 10.

Next, in step S105, the captured image is captured by the camera 10 at the capturing interval set in step S104, and data of the captured captured image is acquired by the viewpoint conversion unit 32. Subsequently, at step S106, the viewpoint conversion unit 32, based on data of the acquired captured image data of the bird's-eye view image PB _t is generated.

Next, in step S107, the alignment unit 33 aligns the data of the bird's-eye view image PB _{t and} the data of the bird's-eye view image PB _{t-1 one} hour before, and generates data of the difference image PD _t. Is done.

Here, in the first embodiment, the imaging interval is set according to the yaw rate and the vehicle speed of the host vehicle, so that the host vehicle is traveling on a straight road as shown in FIG. As shown in FIG. 11B, even if the host vehicle is traveling on a curve, the vehicle is out of the shift amount of the imaging target area of two bird's-eye view images obtained continuously. shift amount in the width direction (Y axis direction), a predetermined number of pixels (e.g. the number of pixels corresponding to a predetermined distance w ₁₎ below. Therefore, the alignment unit 33 was obtained continuously by shifting two bird's-eye view images obtained in succession along the advancing direction (X-axis direction) by a movement amount corresponding to the imaging interval. The alignment of the two bird's-eye view images can be performed with high accuracy, whereby the difference image PD _t suitable for detecting the three-dimensional object can be generated.

FIG. 11C illustrates a scene in which a vehicle equipped with a conventional three-dimensional object detection device is traveling on a curve. In the scene example shown in FIG. 11C, since the imaging interval of the camera 10 is not set according to the turning state of the vehicle, the imaging interval of the camera 10 is longer than that in the scene example shown in FIG. As a result, two bird's-eye view images obtained in succession can be obtained simply by shifting the bird's-eye view images obtained in succession in the advancing direction (X-axis direction) by the amount of movement corresponding to the imaging interval. In some cases, image alignment cannot be performed with high accuracy. That is, in the scene shown in FIG. 11 (C), in order to perform the alignment of two bird's-eye view images obtained in succession with an accuracy capable of appropriately detecting a three-dimensional object, the images were obtained continuously. By capturing two bird's-eye view images by rotating the two bird's-eye view images or shifting them in the vehicle width direction (Y-axis direction), the two obtained bird's-eye view images are matched. It is necessary to estimate the amount of deviation of the target area, which may increase the processing load required for alignment. On the other hand, in the present embodiment, as shown in FIG. 11B, when the travel amount of the host vehicle in the vehicle width direction in a predetermined time increases, such as when the host vehicle travels on a curve, Since the imaging interval of the camera 10 can be set short, by shifting the _two bird's-eye view images obtained in succession along the advancing direction (X-axis direction) by a movement amount corresponding to the imaging interval d2. In addition, the position of the two bird's-eye view images obtained in succession can be aligned with accuracy capable of appropriately detecting a three-dimensional object, and as a result, compared to the scene shown in FIG. The processing load required for alignment can be reduced.

Next, in step S108, the three-dimensional object detection unit 34 counts the number of difference pixels DP having a pixel value “1” from the data of the difference image PD _t generated in step S107, and generates a difference waveform DW _t. .

In the present embodiment, the three-dimensional object detection unit 34, when generating a differential waveform DW _t based on the difference image PD _t, based on the plurality of difference images PD _t, to generate a difference waveform DW _t construction It is good. For example, three-dimensional object detection unit 34, by integrating the differential data on a plurality of difference images PD _t, the pixel value on the difference image difference data is accumulated counts the number of difference pixels DP "1" The differential waveform DW _t may be generated. Alternatively, three-dimensional object detection unit 34, by a plurality of difference images PD _t plurality of the differential waveform DW _t generated respectively, integrating the plurality of differential waveform DW _t that generated the difference based on the plurality of difference images PD _t Waveform DW _t can be generated. Thereby, even when the imaging interval is set short in step S104 and the difference due to the three-dimensional object is difficult to occur on the difference image PD _t , the differential waveform DW _t suitable for detecting the three-dimensional object can be generated.

In step S109, the three-dimensional object determination unit 35 determines whether or not the peak of the differential waveform DW _t is greater than or _{equal to} a predetermined threshold value α. When the peak of the difference waveform DW _t is not equal to or greater than the threshold value α, that is, when there is almost no difference, it is considered that there is no three-dimensional object in the captured image. For this reason, when it is determined that the peak of the differential waveform DW _t is not equal to or greater than the threshold value α (step S109 = No), the three-dimensional object determination unit 35 determines that there is no three-dimensional object and no other vehicle exists (step S109). S117).

On the other hand, when it is determined that the peak of the difference waveform DW _t is equal to or greater than the threshold value α (step S109 = Yes), the three-dimensional object determination unit 35 determines that a three-dimensional object exists in the adjacent lane. In step S110, the three-dimensional object detection unit 34 divides the differential waveform DW _t into a plurality of small areas DW _{t1 to} DW _tn . Next, the three-dimensional object detection unit 34 performs weighting for each of the small areas DW _{t1 to} DW _tn (Step S111), calculates an offset amount for each of the small areas DW _{t1 to} DW _tn (Step S112), and adds the weight. A histogram is generated (step S113).

  Then, the three-dimensional object detection unit 34 calculates a relative movement distance that is a movement distance of the three-dimensional object with respect to the host vehicle based on the histogram, and calculates an absolute movement speed of the three-dimensional object from the relative movement distance (step S114). At this time, the three-dimensional object detection unit 34 calculates the relative movement speed by differentiating the relative movement distance with respect to time, and adds the own vehicle speed detected by the vehicle speed sensor 20 to calculate the absolute movement speed of the three-dimensional object.

  Thereafter, the three-dimensional object determination unit 35 determines whether or not the absolute movement speed of the three-dimensional object is 10 km / h or more and the relative movement speed of the three-dimensional object with respect to the host vehicle is +60 km / h or less (step S115). When both are satisfied (step S115 = Yes), the three-dimensional object determination unit 35 determines that the detected three-dimensional object is an adjacent vehicle existing in the adjacent lane, and an adjacent vehicle exists in the adjacent lane (step S116). Then, the process shown in FIG. 10 ends. On the other hand, when either one is not satisfied (step S115 = No), the three-dimensional object determination unit 35 determines that there is no adjacent vehicle in the adjacent lane (step S117). Then, the process shown in FIG. 10 ends.

  In the present embodiment, the rear side of the host vehicle is set as the detection areas A1 and A2, and emphasis is placed on whether or not there is a possibility of contact when the host vehicle changes lanes. For this reason, the process of step S115 is performed. That is, assuming that the system according to the present embodiment is operated on a highway, if the speed of the adjacent vehicle is less than 10 km / h, even if the adjacent vehicle exists, the host vehicle is required to change the lane. Because it is located far behind, there are few problems. Similarly, when the relative moving speed of the adjacent vehicle to the own vehicle exceeds +60 km / h (that is, when the adjacent vehicle is moving at a speed higher than 60 km / h than the speed of the own vehicle), when changing the lane Is less likely to be a problem because it is moving in front of the host vehicle. For this reason, it can be said that the adjacent vehicle which becomes a problem at the time of lane change is judged in step S115.

  In step S115, it is determined whether the absolute moving speed of the adjacent vehicle is 10 km / h or more and the relative moving speed of the adjacent vehicle with respect to the own vehicle is +60 km / h or less. For example, depending on the mounting error of the camera 10, the absolute moving speed of the stationary object may be detected to be several km / h. Therefore, by determining whether the speed is 10 km / h or more, it is possible to reduce the possibility that the stationary object is determined to be an adjacent vehicle. Further, depending on the noise, the relative speed of the adjacent vehicle to the host vehicle may be detected as a speed exceeding +60 km / h. Therefore, the possibility of erroneous detection due to noise can be reduced by determining whether the relative speed is +60 km / h or less.

  Furthermore, instead of the process in step S115, it may be determined that the absolute moving speed of the adjacent vehicle is not negative or not 0 km / h. Further, in this embodiment, since an emphasis is placed on whether or not there is a possibility of contact when the host vehicle changes lanes, a warning sound is sent to the driver of the host vehicle when an adjacent vehicle is detected in step S116. Or a display corresponding to a warning may be performed by a predetermined display device.

As described above, in the first embodiment, two images obtained at different times are converted into bird's-eye view images, and a difference image PD _t is generated based on the difference between the two bird's-eye view images. When a three-dimensional object is detected from the difference data on PD _t , the amount of movement of the host vehicle in the vehicle width direction within a predetermined time is calculated based on the vehicle speed and yaw rate of the host vehicle, and continuously obtained 2 The imaging interval of the camera 10 is set so that the movement amount in the vehicle width direction between the two bird's-eye view images is equal to or less than a predetermined value (a predetermined number of pixels in the resolution of the bird's-eye view image). Thereby, in this embodiment, as shown to FIG. 11 (A) and (B), the advancing direction (X-axis direction) of the two bird's-eye view images obtained continuously by the movement amount according to the imaging interval It is possible to perform positioning of two bird's-eye view images obtained in succession with high accuracy suitable for detection of a three-dimensional object, and thus a difference image suitable for detection of a three-dimensional object. DP _t can be generated. Furthermore, in this embodiment, by simply shifting two bird's-eye view images obtained in succession along the traveling direction (X-axis direction) by a movement amount corresponding to the imaging interval, the two bird's-eye view images are obtained. Since the positions can be aligned, the processing load required for the alignment can be reduced.

<< Second Embodiment >>
Next, the three-dimensional object detection device 1a according to the second embodiment will be described. As illustrated in FIG. 12, the three-dimensional object detection device 1 a according to the second embodiment includes a camera 10, a vehicle speed sensor 20, a calculator 30 a, a map database 50, and a GPS unit 60. Here, different points will be mainly described in order to avoid redundant description. FIG. 12 is a schematic configuration diagram of a vehicle equipped with the three-dimensional object detection device 1a according to the second embodiment.

  The map database 50 stores map information including road shapes such as road curvature. Map information stored in the map database 50 is acquired by the computer 30a as necessary.

  The GPS unit 60 detects radio waves transmitted from a plurality of satellite communications, and periodically detects position information of the host vehicle. The position information of the host vehicle detected by the GPS unit 60 is transmitted to the computer 30a.

  FIG. 13 is a block diagram showing details of the computer 30a according to the second embodiment. In FIG. 12, the camera 10, the vehicle speed sensor 20, the map database 50, and the GPS unit 60 are also illustrated in order to clarify the connection relationship.

  As shown in FIG. 13, the computer 30a according to the second embodiment includes an imaging interval setting unit 31a, a viewpoint conversion unit 32, a positioning unit 33, a three-dimensional object detection unit 34, and a three-dimensional object determination unit 35. Yes. Again, in order to avoid redundant description, the description will focus on the different points.

  The imaging interval setting unit 31 a acquires map information from the map database 50 and acquires position information of the host vehicle from the GPS unit 60. Then, the imaging interval setting unit 31a identifies the road on which the host vehicle is traveling based on the acquired position information and map information of the host vehicle, and identifies the road shape of the road on which the host vehicle is traveling. . Specifically, the imaging interval setting unit 31a detects the curvature of the road on which the host vehicle is traveling as the road shape of the road on which the host vehicle is traveling. Then, the imaging interval setting unit 31a calculates the moving speed of the host vehicle in the vehicle width direction based on the curvature of the road on which the host vehicle is traveling and the vehicle speed of the host vehicle acquired from the vehicle speed sensor 20. Thereby, the amount of movement of the host vehicle in the vehicle width direction within a predetermined time is calculated.

  The imaging interval setting unit 31a, like the imaging interval setting unit 31 of the first embodiment, is based on the amount of movement of the host vehicle in the vehicle width direction within a predetermined time, and continuously captures two images. A distance interval in which the amount of movement in the vehicle width direction is equal to or less than a predetermined value between the captured images is set as an imaging interval of the camera 10. Then, the imaging interval setting unit 31a causes the camera 10 to perform imaging at the set imaging interval. The viewpoint conversion unit 32, the alignment unit 33, the three-dimensional object detection unit 34, and the three-dimensional object determination unit 35 operate in the same manner as in the first embodiment.

  Next, the adjacent vehicle detection process according to the second embodiment will be described. FIG. 14 is a flowchart illustrating an adjacent vehicle detection process according to the second embodiment.

  As shown in FIG. 14, first, in step S201, vehicle speed information of the host vehicle is acquired by the imaging interval setting unit 31a. In step S202, the position information and map information of the host vehicle are acquired by the imaging interval setting unit 31a. Specifically, the current position of the host vehicle is detected by the GPS unit 60, and the detected position information of the host vehicle is acquired by the imaging interval setting unit 31a. In addition, the imaging interval setting unit 31a refers to the map database 50 and acquires map information around the current position of the host vehicle.

  Next, in step S203, the amount of movement of the host vehicle in the vehicle width direction within a predetermined time is calculated by the imaging interval setting unit 31a. Specifically, the imaging interval setting unit 31a identifies the road on which the host vehicle is traveling based on the map information acquired in step S202 and the position information of the host vehicle. Then, the imaging interval setting unit 31a obtains the curvature of the road on which the host vehicle is traveling based on the map information, and determines the vehicle width of the host vehicle based on the obtained curvature of the road and the vehicle speed of the host vehicle. Calculate the moving speed in the direction. Further, the imaging interval setting unit 31a uses the calculated movement speed of the host vehicle in the vehicle width direction as information indicating the turning state of the host vehicle as the amount of movement of the host vehicle in the vehicle width direction within a predetermined time. calculate.

  In step S204, the imaging interval setting unit 31a sets the imaging interval of the camera 10 based on the amount of movement of the host vehicle in the vehicle width direction within the predetermined time calculated in step S203. Note that the imaging interval setting method in step S204 can be performed in the same manner as in step S104 of the first embodiment.

In steps S205 to S217, the same processing as in steps S105 to S117 of the first embodiment is performed. That is, a captured image is captured by the camera 10 at the capturing interval set in step S204, and data of the captured captured image is acquired by the viewpoint conversion unit 32 of the computer 30 (step S205). Then, based on the image data of the captured image, data of the bird's-eye view image PB _t is generated (step S206), and the data of the bird's-eye view image PB _{t and} the data of the bird's-eye view image PB _{t-1 one} hour before are obtained. By aligning, data of the difference image PD _t is generated (step S207).

Thereafter, a differential waveform DW _t is generated based on the data of the differential image PD _t (step S208). If the peak of the differential waveform DW _t is not equal to or greater than the threshold value α (step S209 = No), there is no solid object. It is determined that there is no other vehicle (step S217).

On the other hand, if it is determined that the peak of the difference waveform DW _t is equal to or greater than the threshold value α (step S209 = Yes), the difference waveform DW _t is divided into a plurality of small regions DW _{t1 to} DW _tn (step S210). Weighting is performed for each of the areas DW _{t1 to} DW _tn (step S211), an offset amount for each of the small areas DW _{t1 to} DW _tn is calculated (step S212), and a histogram is generated with the weights added (step S213). Then, based on the histogram, a relative moving distance that is a moving distance of the adjacent vehicle with respect to the own vehicle is calculated, and an absolute moving speed and an absolute moving speed of the adjacent vehicle are calculated from the relative moving distance (step S214). When the absolute moving speed of the adjacent vehicle is 10 km / h or more and the relative moving speed of the adjacent vehicle with respect to the own vehicle is +60 km / h or less (step S215 = Yes), the three-dimensional object determination unit 35 enters the adjacent lane. It is determined that there is an adjacent vehicle (step S216). On the other hand, if either one is not satisfied (step S215 = No), the three-dimensional object determination unit 35 determines that there is no adjacent vehicle in the adjacent lane (step S215). S217). Then, the process shown in FIG. 14 ends.

As described above, in the second embodiment, two captured images obtained at different times that are successively obtained are converted into bird's-eye view images, and the two bird's-eye view images are aligned to obtain the difference image PD _t. generate, when detecting the three-dimensional object from the difference data in the difference image PD _t, based on the position information and the map information of the vehicle, calculates the movement amount of the vehicle is moved in the vehicle width direction within a predetermined time Then, the imaging interval of the camera 10 is set so that the amount of movement in the vehicle width direction is equal to or less than a predetermined value (a predetermined number of pixels in the resolution of the bird's-eye view image) between two bird's-eye view images obtained in succession. . Thereby, in 2nd Embodiment, in addition to the effect of 1st Embodiment, also in the solid-object detection apparatus 1a which is not provided with a yaw rate sensor, the alignment of two bird's-eye view images obtained continuously is carried out to a solid object. As a result, it is possible to generate a differential image PD _t suitable for detecting a three-dimensional object.

  The embodiment described above is described for facilitating the understanding of the present invention, and is not described for limiting the present invention. Therefore, each element disclosed in the above embodiment is intended to include all design changes and equivalents belonging to the technical scope of the present invention.

  Note that the camera 10 of the above-described embodiment corresponds to the imaging unit of the present invention, the imaging interval setting unit 31 corresponds to the turning state acquisition unit and the control unit of the present invention, and the viewpoint conversion unit 32 corresponds to the image conversion unit of the present invention. The positioning unit 33 corresponds to the difference image generation unit of the present invention, the three-dimensional object detection unit 34 corresponds to the three-dimensional object detection unit of the present invention, and the three-dimensional object determination unit 35 corresponds to the three-dimensional object determination unit of the present invention. It corresponds to.

DESCRIPTION OF SYMBOLS 1, 1a ... Three-dimensional object detection apparatus 10 ... Camera 20 ... Vehicle speed sensor 30, 30a ... Computer 31, 31a ... Imaging interval setting part 32 ... Viewpoint conversion part 33 ... Position alignment part 34 ... Three-dimensional object detection part 35 ... Three-dimensional object determination part 40 ... yaw rate sensor 50 ... map database 60 ... GPS unit a ... angle A1, A2 ... detection area CP ... intersection DP ... differential pixel _{DW t, DW} t _'... differential waveform _{DW t1 ~DW m, DW m +} k ~DW tn ... small regions L1, L2 ... ground line La, Lb ... line PB t on the direction the three-dimensional object collapses _... bird's-eye view image PD t _... difference image V1 ... vehicle V2 ... adjacent vehicle

Claims (7)

Imaging means for imaging a predetermined area;
Vehicle speed detecting means for detecting the vehicle speed of the vehicle;
A turning state detecting means for detecting a turning state of the vehicle;
A control unit that sets an imaging interval of the imaging unit to a predetermined equidistance interval according to a turning state and a vehicle speed of the vehicle, and causes the imaging unit to capture a captured image at the set imaging interval;
Differential image generation means for generating a differential image by aligning the position of the image based on the captured image obtained continuously at the imaging interval along the traveling direction ;
Pixel number counting means for counting the number of pixels showing a predetermined difference on the difference image;
It generates a difference waveform information by the frequency distribution of counts the number of the pixels, and a three-dimensional object detection means for detecting a three-dimensional object on the basis of the differential waveform information,
The control means calculates a travel distance in which a shift amount in a vehicle width direction of each imaging target region in the captured images obtained continuously is equal to or less than a predetermined number of pixels in the captured image, and the calculated travel A three-dimensional object detection apparatus , wherein a distance is set as the imaging interval . The three-dimensional object detection device according to claim 1,
The turning state detection means calculates a movement amount of the vehicle in the vehicle width direction within a predetermined time based on the yaw rate of the vehicle, and detects the calculated movement amount in the vehicle width direction as the turning state of the vehicle. The three-dimensional object detection apparatus characterized by the above-mentioned. The three-dimensional object detection device according to claim 1,
Position information acquisition means for acquiring vehicle position information;
Map information acquisition means for acquiring map information,
The turning state detecting means predicts a road shape on which the vehicle travels based on the position information of the vehicle and the map information, and in the vehicle width direction within a predetermined time based on the predicted road shape. A three-dimensional object detection device that calculates a movement amount and detects the calculated movement amount in the vehicle width direction as a turning state of the vehicle. The three-dimensional object detection device according to claim 2, wherein
The three-dimensional object detection device, wherein the control unit sets the imaging interval to be shorter as the movement amount of the vehicle in the vehicle width direction within the predetermined time is larger. The three-dimensional object detection device according to any one of claims 1 to 4,
The difference image generation means specifies the amount of deviation of each imaging target region in the captured image obtained continuously based on the imaging interval, and is obtained continuously according to the identified amount of deviation. A three-dimensional object detection apparatus characterized by aligning the positions of images based on the continuously obtained captured images by shifting images based on the captured images. The three-dimensional object detection device according to any one of claims 1 to 5,
Image conversion means for converting the captured image obtained continuously at the imaging interval into a bird's-eye view image;
The three-dimensional object detection device, wherein the difference image generation means generates the difference image by aligning the positions of the bird's eye view images continuously obtained by the image conversion means on the bird's eye view. . The three-dimensional object detection device according to any one of claims 1 to 6,
The three-dimensional object detection means generates the difference waveform information based on the plurality of difference images generated within a predetermined time, and detects the three-dimensional object based on the generated difference waveform information. 3D object detection device.

Images