JP2020027057A - Stereo camera device - Google Patents

Abstract

To provide a stereo camera device capable of discriminating between parallax on a road surface and parallax of an obstacle, discriminating between an obstacle (an obstacle on the road surface) such as a stereoscopic object on the remote road surface or subsidence from noise, and detecting it early.SOLUTION: A stereo camera device estimates parallax on the road surface, tracks a feature point on the road surface different from the parallax with a plurality of frames, analyzes whether or not a change in parallax of the feature point is a change in parallax corresponding to a travel distance of a vehicle, and detects a road obstacle.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an in-vehicle stereo camera device that recognizes an obstacle outside a vehicle using a plurality of cameras mounted on the vehicle.

  In order to improve the running safety of the vehicle, a system that detects obstacles ahead etc. with a sensor mounted on the vehicle and issues a warning to the driver and automatic braking if there is a possibility of collision with the obstacle Has been studied.

  Examples of sensors for monitoring the front of a vehicle include a millimeter wave radar, a laser radar, and a camera. The types of cameras include a monocular camera and a stereo camera using a plurality of cameras. The stereo camera can measure a distance to a photographed object by using a parallax of an overlapping area photographed by two cameras arranged at a predetermined interval. For this reason, it is possible to accurately grasp the risk of collision with an object such as the front.

  The stereo camera obtains a parallax between images captured by the two cameras, and converts the parallax into a distance. The stereo camera is characterized in that the parallax decreases as the measurement distance increases. Therefore, when detecting an obstacle on a distant road, the difference between the parallax of the road surface and the parallax of the obstacle becomes small, and it becomes difficult to distinguish the noise of the parallax of the road surface from the parallax of the obstacle. On a road such as a highway where vehicles travel at high speed, it is necessary to detect obstacles on the road at an early stage to prevent accidents. For that purpose, it is required to distinguish the parallax of a distant road surface and the parallax of an obstacle with high accuracy. Patent Document 1 listed below proposes a method for preventing erroneous calculation of parallax.

JP 2014-85120 A

  When calculating the disparity, the above-described Patent Document 1 assumes the disparity range of the object in the current frame using the disparity of the object calculated in the image of the past frame, and sets the disparity of the object in the current frame to the estimated range. It makes it easy to take a value. However, Patent Literature 1 does not disclose that noise of distant obstacle parallax and road surface parallax is detected separately.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a stereo camera device that distinguishes between parallax of a road surface and parallax of an obstacle to obtain an obstacle such as a three-dimensional object or a depression on a distant road. An object of the present invention is to provide a stereo camera device capable of detecting (an obstacle on the road) at an early stage.

  The features of the present invention for solving the above problems are, for example, as follows. That is, a stereo camera device that recognizes obstacles on the road from images captured by a plurality of cameras mounted on a vehicle, calculates parallax from images of the plurality of cameras, and estimates parallax of a road surface in the images. A feature point on a road surface having a difference is extracted, and when it is determined that a change in parallax between a plurality of frames of the feature point corresponds to a moving distance of the vehicle, the feature point is recognized as a road obstacle.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to distinguish the parallax of a distant road obstacle and the noise of the parallax of a road surface, and it becomes possible to detect a distant road obstacle early.

  Problems, configurations, and effects other than those described above will be apparent from the following description of the embodiments.

FIG. 1 is a system configuration diagram of an embodiment of a stereo camera device of the present invention. It is a figure which shows the relationship between the obstacle on the distant road and road surface parallax, (a) is an example of an image when image | photographed by a right camera, (b) is a schematic diagram of the graph which shows the relationship between distance and road surface parallax. FIG. 4 is an explanatory diagram of a parallax calculation process. 9 is a flowchart of a detection process of a distant road obstacle. It is a schematic diagram of a feature point detection process, (a) is a schematic diagram of a feature point detection process in a first frame image (an image captured earlier), and (b) is a schematic diagram of a feature point detection process in a second frame image (an image captured later). FIG. 4 is a schematic diagram of a feature point detection process. Image example when the road surface is inclined in the horizontal direction. 9 is a flowchart of feature point extraction processing when the road surface is inclined in the horizontal direction. FIG. 10 is a system configuration diagram of another example of the stereo camera device of the present invention, which is compatible with pitching of a vehicle. 9 is a flowchart of a detection process of a distant road obstacle corresponding to the pitching of the vehicle. FIG. 3 is a configuration diagram of a feature point hardware detection unit suitable for hardware processing. Image example when there is a sign on the road surface.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings and the like.

  The following description shows specific examples of the content of the present invention, and the present invention is not limited to these descriptions, and various modifications by those skilled in the art within the technical idea disclosed in the present specification. Changes and modifications are possible. In all the drawings for describing the present invention, components having the same function are denoted by the same reference numerals, and repeated description thereof may be omitted.

  FIG. 1 is a block diagram showing an embodiment of a stereo camera device 10 according to the present invention. The stereo camera device 10 is mounted on a vehicle and is for recognizing obstacles outside the vehicle.The stereo camera device 10 has a built-in electronic circuit mainly composed of a microcomputer including a CPU, a ROM, a RAM, and the like. A parallax calculating unit 30, a road parallax estimating unit 40, a feature point detecting unit 50, a feature point recording unit 60, a moving distance parallax correcting unit 70, and a detecting unit 80.

  The photographing unit 20 is connected to two cameras, a right camera 22 and a left camera 21, which are arranged at a predetermined interval. The photographing unit 20 performs luminance correction, noise elimination, and distortion correction of images captured by the left and right cameras 21 and 22. The parallax calculating unit 30 calculates parallax using left and right camera images (images captured by the left and right cameras 21 and 22). The road surface parallax estimating unit 40 estimates the road surface parallax of the traveling lane in which the vehicle equipped with the stereo camera 10 travels. The feature point detection unit 50 detects, from the road surface parallax, a feature point of a portion different from the road surface parallax (a feature point on a road surface having a difference from the road surface estimated parallax in the image). The feature point recording unit 60 records the position and the parallax of the feature point. The position is not a coordinate on the image but a position on the terrain such as an angle and a distance with respect to the traveling direction of the vehicle or a latitude and longitude. The moving distance parallax correction unit estimates and corrects how the parallax recorded by the feature point recording unit 60 changes when the vehicle moves over time. In other words, the vehicle moves as time elapses, and the distance between the feature point and the vehicle approaches due to the movement distance. Therefore, the amount of change in parallax according to the distance is calculated and corrected. For the feature points of the two frame images having different times, the detection unit 80 corrects the parallax of the feature points of the image captured earlier by the moving distance parallax correction unit 70, and the parallax of the image captured later. Compare the parallax of the feature point (particularly, the parallax of a part where the image of the feature point matches with an image captured later) and if the difference (comparison value) is within a predetermined range, Since it can be determined that the change in the parallax between the plurality of frames of the feature point corresponds to the moving distance of the vehicle, the feature point is detected as an obstacle on the road. In order for the vehicle to take an avoidance action, the detection result (obstacle detection information) of the detection unit 80 is used for controlling the accelerator, brake, steering, and the like of the vehicle.

  FIG. 2 illustrates the relationship between distant road obstacles and road surface parallax. FIG. 2A shows a state in which a road obstacle 300 is located far from the road surface 120 between the left lane 200 and the right lane 201 as an image taken by the right camera 22. FIG. 2B is a schematic diagram of a graph showing the relationship between the distance and the road surface parallax. As shown in the figure, the parallax decreases as the distance increases. Therefore, assuming that the obstacle 300 on the road is at a distance of about 100 m, the point (feature point) has a slight change in the road surface parallax. Therefore, it is necessary to detect the distant road obstacle 300 separately from noise.

但し、Iは前記参照画像212の中の画像ブロック（例：8x8画素）、Tは前記基準ブロック画像221の画像データであり、i,jは画像ブロック内の座標である。１つの視差を算出するために、前記参照画像212の参照位置を１画素ずつずらしながら前記探索幅の分だけ演算を行い、最もSAD値が小さくなる位置211を探索する。 FIG. 3 shows a parallax calculation process of the parallax calculation unit 30. An image 2210 taken by the right camera 22 is used as a reference image, and a reference block image 221 such as 8 × 8 pixels is defined. The size of the block image is not limited to this example. On the other hand, in the image 2110 taken by the left camera 21, the reference image 212 having the search width (for example, 256 pixels) is set based on the same vertical position (Y coordinate) and horizontal position (X coordinate) as the reference block image 221. select. Thereafter, a difference between the reference block image 221 and the reference image 212 is calculated. This difference calculation is called SAD, and calculates the following equation (1).
(Equation 1)

Here, I is an image block (for example, 8 × 8 pixels) in the reference image 212, T is image data of the reference block image 221 and i, j are coordinates in the image block. In order to calculate one parallax, a calculation is performed for the search width while shifting the reference position of the reference image 212 by one pixel, and a position 211 where the SAD value becomes the smallest is searched.

In this way, parallax is obtained for the entire image. Using the parallax d, the distance to the stereo camera device 10 can be measured based on the principle of triangulation. The distance Z is obtained from the parallax d by the following equation (2).
(Equation 2)
Z = (f × B) / (d × a) (2)
Here, f is the focal length of the right and left cameras 22 and 21, B is the distance (base line length) between the right camera 22 and the left camera 21, and a is the horizontal size of one pixel of the left camera 21.

  FIG. 4 shows an operation flow of the stereo camera device 10 for detecting a distant road obstacle 300.

  First, an image is captured by the left camera 21 and the right camera 22. This image is used as the first frame image by the photographing unit 20 (S100). Next, the entire parallax of the first frame image is generated by the parallax calculating unit 30 (S110). The road surface parallax is extracted from the parallax, and the road surface parallax estimating unit 40 estimates the road surface parallax in units of one pixel in the Y direction of the image. As a method of estimating the parallax of the road surface, for example, if the road surface is horizontal, the parallax of the left lane 200 or the right lane 201 is selected (S120). Next, the feature point detection unit 50 detects feature points on the road surface. The processing of this part will be described with reference to FIG. In the first frame image of FIG. 5A, a part having a difference from the estimated road surface parallax is extracted as a feature point while scanning the parallax data in the range of the road surface in the raster direction (that is, the lateral direction). FIG. 5A shows an example in which three rows of raster 301, raster 302, and raster 303 are scanned. Assuming that the road surface disparity of the raster 301 is estimated to be 2.0, looking at the disparity of the raster 301 on the road, the disparity of the positions of the feature points 400, 401, and 402 is 2.1, Since the estimated road surface disparity is different from 2.0, the position and the disparity value are recorded (S130).

  Next, an image is photographed by the left camera 21 and the right camera 22, and the photographing unit 20 sets it as a second frame image (S140). It is desirable that the interval between the two frame images is an interval at which the feature points move by one raster or more on the image. The parallax of the entire second frame image is generated by the parallax calculating unit 30 (S150). Next, the feature points recorded in S130 are tracked by the second frame image (by matching processing), and the parallax of the matched part is extracted and analyzed as A. This extraction method will be described with reference to FIG. In the second frame image of FIG. 5B, the time has passed since the image of FIG. 5A was captured, and thus the road obstacle 300 is approaching the vehicle. Then, the road obstacle 300 moves downward in the image. This phenomenon is based on the fact that the image in front of the vehicle is a perspectively projected image. In perspective projection, an object approaching the vehicle is far from the vanishing point. In the examples of FIGS. 5A and 5B, the vanishing point is the farthest point on the road surface, and the road surface is farther from the vanishing point (closer to the vehicle) as the image is lower than the vanishing point. When an object on the road approaches the vehicle, the object moves to the bottom of the image in order to move away from the vanishing point of the road. In the example of FIG. 5B, it is assumed that the position has been moved to the position of the raster 302. When the parallax on the road surface of the raster 302 is compared with the road surface estimated parallax of 2.1, the feature point 400 is the same as the road surface parallax, and there is a high possibility that the noise was parallax noise. On the other hand, the feature point 401 and the feature point 402 have a disparity of 2.2, and are extracted as a portion having a difference from the road surface estimated disparity 2.1 (S160). Next, the moving distance parallax correction unit 70 calculates the distance that the vehicle has moved between the two frames using the vehicle signal. Since the parallax d and the distance Z have a unique relationship according to the above equation (2), a change amount of the parallax corresponding to the moving distance of the vehicle can be obtained by applying the equation (2).

  The amount of change in the parallax corresponding to the moving distance is added (added) to the parallax 2.1 of the feature points 401 and 402 extracted in the first frame. The amount of change in parallax with respect to the moving distance of the vehicle can be calculated by a modification of equation (2). The result is set to B (S170). Next, the detection unit 80 compares the A and the B, and determines whether or not the difference is within a predetermined value (S180). If the difference is within a predetermined value, it can be determined that the change in the parallax between the plurality of frames of the feature points 401 and 402 corresponds to the moving distance of the vehicle, so that the feature points 401 and 402 are determined to be road obstacles. (S190).

  That is, in the above method, the parallax of the feature point changes in the coordinates and the parallax on the image in accordance with the movement of the vehicle (in other words, the change of the parallax between the plurality of frames of the feature point corresponds to the moving distance of the vehicle. ) Is determined to be a road obstacle. In the case of noise, it is extremely rare to change under such conditions, and this method can prevent the noise from being erroneously detected as an obstacle on the road.

[Processing example when the road surface is inclined in the horizontal direction]
FIG. 6 shows an example of an image when the road surface 120 is inclined left and right (horizontally). When the road surface 120 is inclined left and right, the road surface estimated parallax at both ends on the raster in the road surface is different, and the road surface estimated parallax cannot be defined by one parallax for one raster. FIG. 7 shows a feature point extraction method for that purpose.

  FIG. 7 is an operation flow of the feature point extraction processing when the road surface 120 is inclined in the horizontal direction. This extraction process is performed by the feature point detection unit 50 (see FIG. 1).

  First, the parallax of two points in the left and right lane portions on the same raster of the image is extracted (S200). Next, the number of pixels between the two points is calculated, and the difference between the parallax of the point on the left side (parallax on one end side) and the parallax of the point on the right side (parallax on the other end side) is calculated. Divide. Thus, a change in parallax in the minimum pixel unit of the raster is calculated (S210). Next, as the estimated road surface disparity, the disparity is calculated for each pixel as disparity of the left point (disparity on one end side) + (result of S210 (the change) × the number of pixels from the left point) (S220). The road surface estimated parallax is updated in accordance with the scanning in the right direction of the raster, and a road surface portion different from the road surface estimated parallax is extracted as a feature point (S230).

  In the above method, the estimated road surface parallax can be calculated in pixel units even if the road surface is inclined left and right, so that feature points can be detected with high accuracy.

[Example of stereo camera device corresponding to vehicle pitching]
FIG. 8 is a block diagram of another example of the stereo camera device of the present invention corresponding to pitching of a vehicle. FIG. 8 illustrates a case where the vehicle vibrates during shooting of a plurality of frame images and blur occurs in the image. FIG. 3 is a block diagram showing a feature point tracking method. The difference from FIG. 1 is a portion having a feature point image matching unit 90. The feature point image matching unit 90 performs matching processing on the surrounding image of the feature point detected in the first frame image in the second frame image, and specifies the coordinates of the feature point in the second frame image.

  FIG. 9 shows a flow of a process of detecting a road obstacle 300 by the stereo camera device 10 using the feature point image matching unit 90.

  First, an image is captured by the left camera 21 and the right camera 22. This image is used as the first frame image by the photographing unit 20 (S300). Next, the entire parallax of the first frame image is generated by the parallax calculating unit 30 (S310). The road surface parallax is extracted from the parallax, and the road surface parallax is estimated by the road surface parallax estimating unit 40 for each pixel in the Y direction of the image (S320). Next, the feature points on the road surface are detected by the feature point detection unit 50, and the parallax values and surrounding images are cut out and recorded (S330).

  Next, an image is photographed by the left camera 21 and the right camera 22, and the photographing unit 20 sets it as a second frame image (S340). The parallax of the entire second frame image is generated by the parallax calculating unit 30 (S350). Next, the feature point image matching unit 90 performs matching processing between the surrounding image of the feature point extracted in the first frame image and the second frame image, and detects the coordinates of the feature point on the second frame image (S360). . Next, the parallax of the feature point of the second frame image detected in S360 is extracted as A (S370). Next, the moving distance parallax correction unit 70 calculates the distance that the vehicle has moved between the two frames, and adds (adds) the parallax corresponding to the moving distance of the vehicle to the feature points extracted in the first frame image. . The result is set to B (S380). Next, the detection unit 80 compares the A and the B, and determines whether or not the difference is within a predetermined value (S390). If the difference is within a predetermined value, the feature point is determined to be a road obstacle (S400).

  In other words, in the above method, since the image of the feature point portion is subjected to the matching process between a plurality of frames, it is possible to precisely track the feature points even if the image is blurred.

[Configuration example suitable for hardware processing]
FIG. 10 illustrates a parallel processing of a function of detecting a feature point (a function of the feature point detection unit 50 of FIG. 1) and a function of generating a parallax on the road surface (a function of the road surface parallax estimation unit 40 of FIG. 1) by hardware. FIG. 2 shows a configuration diagram of a feature point hardware detection unit 51 that enables high-speed processing.

  The parallax calculating unit 30 calculates parallax for each pixel of the image captured by the capturing unit 20. For this purpose, the parallax calculating unit 30 includes a one-pixel parallax calculating unit 350 that performs a process of obtaining parallax one pixel at a time, and a coordinate updating unit 351 that generates coordinates for repeating the process for the entire image.

  On the other hand, each time the one-pixel parallax calculating unit 350 generates a parallax, the feature point hardware detection unit 51 determines whether or not the parallax is a feature point. For that purpose, the feature point hard detection unit 51 includes a left lane coordinate register 360, a right lane coordinate register 362, and a road surface estimated parallax register 364, which are means for storing the estimated parallax of the left and right ends of the road surface and the coordinates thereof, and the calculated road surface. It has comparators 361, 363, 365, and a feature point selection unit 366 which are means for comparing the above parallax with the estimated parallax of the road surface and outputting parallax having a predetermined difference and its coordinates. In the left lane coordinate register 360, the X coordinate of the left lane 200 is set and stored by software on a raster basis of the image before the processing is started. In the right lane coordinate register 362, the X coordinate of the right lane 201 is set and stored by software on a raster basis of the image before the processing is started. In the road surface estimated parallax register 364, the road surface estimated parallax is set and stored by software in raster units of the image. In such a state, the one-pixel parallax calculating unit 350 calculates the parallax in the image. The X coordinate of the image at that time is compared with the X coordinate value of the raster in the left lane coordinate register 360 by the comparator 361, and the comparison result (that is, the coordinate having a predetermined difference) is sent to the feature point selection unit 366. At the same time, the X coordinate is compared with the X coordinate value of the raster in the right lane coordinate register 362 by the comparator 363, and the comparison result (that is, the coordinate having a predetermined difference) is sent to the feature point selection unit 366. The disparity data is compared with the data of the road surface estimation disparity register 364 by the comparator 365, and the comparison result (that is, the disparity having a predetermined difference) is sent to the feature point selection unit 366. The feature point selection unit 366 determines that the output of the comparator 361 satisfies the register value ≦ X coordinate, the output of the comparator 363 satisfies the register value ≧ X coordinate, and the result of the comparator 365 is road surface estimated disparity ≠ disparity. When the conditions are satisfied, the coordinate values and the parallax data at that time are sent to the feature point recording unit 60.

  With the above configuration, high-speed processing is possible because parallax estimation and feature point detection processing are performed in parallel.

[Example of processing when there is a sign on the road surface]
FIG. 11 shows an example of an image when a sign is present on the road surface. In the parallax calculation of the stereo camera device 10, matching processing of the left and right camera images is performed as shown in FIG. For that purpose, the texture of the image is required, and the parallax cannot be calculated for a portion having a small texture such as a road surface, which may result in invalid parallax. However, effective parallax is easily obtained in a portion with a road marking. Therefore, when there is parallax on the road, it may be a road marking as well as an obstacle on the road. If the image 310 of the part where the parallax exists is image-matched with the texture of the road marking to recognize the road marking, the processing area of the image matching is limited, so that the processing speed can be increased.

  Furthermore, if the parallax of the portion of the image 310 is tracked between a plurality of frames and a change in the parallax according to the moving distance of the vehicle can be confirmed, the parallax of the portion of the image 310 can be distinguished from noise.

  As described above, according to the present embodiment, when it is determined that the change in the parallax between a plurality of frames of the feature point on the road surface having a difference between the estimated parallax of the road surface in the image and the difference corresponds to the moving distance of the vehicle, Since the feature points are recognized as road obstacles, it is possible to distinguish the noise of the parallax of the distant road obstacle from the parallax noise of the road surface, and it is possible to detect the distant road obstacle at an early stage.

  Note that the present invention is not limited to the above-described embodiment, and includes various modifications. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described above. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of one embodiment can be added to the configuration of another embodiment. Further, for a part of the configuration of each embodiment, it is possible to add, delete, or replace another configuration.

  In addition, each of the above-described configurations, functions, processing units, processing means, and the like may be partially or entirely realized by hardware, for example, by designing an integrated circuit. In addition, the above-described configurations, functions, and the like may be realized by software by a processor interpreting and executing a program that realizes each function. Information such as a program, a table, and a file for realizing each function can be stored in a memory, a hard disk, a storage device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  In addition, control lines and information lines are shown as necessary for the description, and do not necessarily indicate all control lines and information lines on a product. In fact, it can be considered that almost all components are connected to each other.

10 Stereo camera device
20 Shooting unit
21 Left camera
22 Right camera
30 Parallax calculator
40 Road surface parallax estimator
50 Feature point detector
51 Feature point hardware detector
60 Feature point recording section
70 Moving distance parallax correction unit
80 Detector
90 Feature point image matching unit
120 road surface
200 left lane
201 right lane
300 Obstacles on the street
350 1-pixel parallax calculation unit
351 Coordinate update unit
360 Left lane coordinate register
361 comparator
362 Right lane coordinate register
363 comparator
364 Road Estimation Parallax Register
365 comparator
366 Feature point selector

Claims (7)

A stereo camera device that recognizes an obstacle on a road from images taken by a plurality of cameras mounted on a vehicle,
Calculating parallax from the images of the plurality of cameras,
Extract the feature points on the road surface having the estimated parallax and difference of the road surface in the image,
A stereo camera device characterized in that when it is determined that a change in parallax between a plurality of frames of the feature point corresponds to a moving distance of the vehicle, the feature point is recognized as a road obstacle. Correct the parallax of the feature point of the image taken earlier with the moving distance of the vehicle,
The corrected parallax is compared with an image obtained by matching the image of the feature point with an image captured later, and the parallax of the matched portion is compared. When the difference is within a predetermined range, the feature point is determined on the road. The stereo camera device according to claim 1, wherein the stereo camera device is determined to be an obstacle.   The method according to claim 2, wherein a surrounding image of the feature point of the previously shot image is subjected to a matching process in an image shot later, and the coordinates of the feature point in the image shot later are specified. The stereo camera device according to claim 1. When the parallax of both ends on the raster on the road surface in the image is different,
Calculate a change in parallax in the minimum pixel unit of the raster,
The stereo camera device according to claim 1, wherein the estimated parallax of the road surface is calculated based on the variation and the number of pixels from one end of a raster on the road surface. The stereo camera device,
Means for storing the estimated parallax of the left and right ends of the road surface and their coordinates,
The stereo camera device according to claim 1, further comprising: means for comparing the calculated disparity on the road surface with the estimated disparity on the road surface, and outputting a disparity having a predetermined difference and its coordinates. A stereo camera device that recognizes an obstacle on a road from images taken by a plurality of cameras mounted on a vehicle,
A parallax calculating unit that calculates parallax from the images of the plurality of cameras,
A road surface parallax estimating unit that estimates a road surface parallax in the image,
A feature point detection unit that detects a feature point on the road surface having the estimated parallax and the difference of the road surface,
A stereo camera device, comprising: a detection unit that, when it is determined that a change in parallax between a plurality of frames of the feature point corresponds to a moving distance of the vehicle, recognizes the feature point as a road obstacle. A moving distance parallax correction unit that corrects the parallax of the feature point with a moving distance of the vehicle,
The detection unit is configured to correct the parallax of the feature point of the previously captured image by the moving distance parallax correction unit, and the parallax of a portion where the image of the feature point matches a subsequently captured image by matching processing. 7. The stereo camera device according to claim 6, wherein the characteristic point is determined to be a road obstacle when the difference is within a predetermined range.

Images