JP2013085142A - Drive support device for vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a drive support device which can prevent a travel region division line which should be extended in a cross direction of its own vehicle from displayed as being bent in a display in the case of displaying a photographed image of a vehicle in a rear side which is enlarged or compressed in a magnification changed in a horizontal direction.SOLUTION: Image forming means 7 for display generates an image for display to be displayed on a display 5 from a photographed image of a camera 3 photographing a rear side of a vehicle 2. The image forming means 7 generates an image for display by executing horizontal scale modification processing in which each part of the photographed image is modified in scale in a set horizontal scaling magnification; and vertical scale modification processing in which a horizontal scale modification image is partially performed with scale modification in a vertical scale modification set so that an image of a travel region division line 21, bent to be displayed in the horizontal scale modification image generated from the photographed image by the horizontal scale modification processing, extends linearly.

Description

  The present invention relates to a driving support device that displays a captured image of a rear side of a vehicle to a driver.

  Conventionally, in order to support driving of a vehicle, a captured image on the rear side of the vehicle is acquired with a wide angle of view in the lateral direction by a camera mounted on the vehicle, and the acquired captured image is displayed on the front side of the driver's seat. The present applicant has proposed a device that allows the driver of the vehicle to visually confirm the situation on the rear side of the vehicle including the blind spot area that is not reflected in the door mirror by displaying on the display. (For example, refer to Patent Document 1).

  In this type of device, since the angle of view of the captured image of the in-vehicle camera is wide, if the captured image is displayed on the display as it is, the driver can visually recognize the captured image on the door mirror. An object such as another vehicle existing in the area to be obtained becomes too small in the captured image as compared with the size of the object visually recognized on the door mirror, and it is difficult for the driver to visually recognize the object. Alternatively, it is difficult to visually recognize the distance between the object in the captured image and the host vehicle (the distance in the front-rear direction of the host vehicle) based on the size of the object in the captured image.

  For this reason, in the technique shown in Patent Document 1, when an image captured by a camera is displayed on a display device, an image of a region that can be visually recognized on a door mirror is enlarged in the horizontal direction, while the outside of the region (own vehicle) The image of the blind spot area on the opposite side) is compressed in the horizontal direction and displayed on the display. Thus, while maintaining the horizontal angle of view of the display image of the display device at a wide angle, the size of the image of the object existing in the region that can be visually recognized on the door mirror in the display image exists, that is, in the distance. The size of the image of the object is set to be approximately the same as the size of the object visually recognized on the door mirror.

JP 2008-22125 A

  By the way, in the technique as shown in Patent Document 1, the lateral enlargement and compression of the captured image of the camera is performed at a magnification set so as to change according to the position of the captured image in the lateral direction. Become. In the display image displayed on the display, the image of the blind spot area that cannot be visually recognized on the door mirror becomes an image compressed in the lateral direction as compared with the image of the area that can be visually recognized on the door mirror. A traveling area dividing line such as a white line that should extend linearly in substantially the same direction as the front-rear direction of the vehicle is displayed as a curved traveling area dividing line on the display image of the display.

  For this reason, it has been found by the inventor's examination that there is a disadvantage that the driver feels uncomfortable, or that the position of an object such as another vehicle existing in the blind spot area is difficult to understand in the lateral direction with respect to the own vehicle. .

  The present invention has been made in view of such a background, and in the case where an image obtained by enlarging or compressing a captured image of a rear side of a vehicle at a magnification that changes in the lateral direction is displayed on a display device. An object of the present invention is to provide a vehicle driving support device that can prevent a traveling area division line such as a white line that should extend in the front-rear direction of the vehicle from being bent and displayed in an image displayed on a display. To do.

  In order to achieve the above object, the vehicle driving support apparatus of the present invention has a camera that captures the rear side of the vehicle, and a display image generating means that generates a display image that is visually recognized by the driver from the captured image of the camera. And a display for displaying the display image generated by the display image generating means, wherein the display image generating means is a lateral direction of the captured image of the in-vehicle camera. A horizontal scale correction process for correcting the scale of each part of the captured image in the horizontal direction at a horizontal scaling factor set so that the value changes according to the position of the vehicle, and linearly extending in the front-rear direction of the vehicle in real space. An image of a traveling area parting line that is an existing traveling area parting line and is bent and displayed in the horizontal scale corrected image generated from the captured image by the horizontal scale correcting process is displayed in the horizontal scale corrected image. Vertical scale correction processing for partially correcting the horizontal scale correction image in the vertical direction at a vertical scaling factor set so as to be an image extending linearly in the longitudinal direction of the vehicle. By the above, it is means for generating the display image (first invention).

  According to the first aspect of the invention, the display image generating means performs the horizontal scaling process at the vertical scaling factor in addition to the horizontal scale correction process for correcting the scale of each part of the captured image at the horizontal scaling factor. A display image to be displayed on the display is generated by executing a vertical scale correction process in which the scale correction image is partially corrected in the vertical direction.

  Thus, the horizontal scale correction process corrects the distortion of the travel area dividing line that is bent and displayed in the horizontal scale correction image, and the display image is linear in the front-rear direction of the vehicle. It is possible to make an image that extends to

  Therefore, according to the first aspect of the present invention, when an image obtained by enlarging or compressing the captured image of the rear side of the vehicle with the magnification that changes in the lateral direction (the lateral expansion / contraction magnification) is displayed on the display. It is possible to prevent a traveling area dividing line such as a white line that should extend in the front-rear direction of the vehicle from being bent and displayed in an image displayed on the display.

  At the same time, not only the travel line dividing line, but also all straight lines parallel to the longitudinal direction of the vehicle, such as guardrails, curbs, or edges of structures such as buildings (these pass through the vanishing point of the captured image) are displayed. You can correct the bend in the image. For this reason, the uncomfortable feeling given to the driver can be reduced.

  Note that the captured image used to generate the display image by the horizontal scale correction process and the vertical scale correction process does not have to be a captured image of the entire imaging area of the camera, but a part of the imaging area. The captured image may be used.

  In the first aspect of the invention, the horizontal scaling factor in the horizontal scale correction processing can be set so as to change according to the position in the horizontal direction of the captured image in various forms. For example, the horizontal scaling factor in the horizontal scale correction process is, for example, a predetermined constant magnification value for enlargement in a predetermined range in the horizontal direction in an area visually recognized by the side mirror of the vehicle in the captured image of the camera. In addition, in a range outside the predetermined range, the fixed magnification value is set to be changed from the constant magnification value to the compression magnification value in accordance with the lateral position of the captured image.

  By setting the horizontal enlargement / reduction ratio in this way, the horizontal angle of view of the display image is made wider than the area that can be visually recognized by the side mirror, and the area that is visually recognized by the side mirror. Preventing the size of the image of an object such as an existing vehicle from becoming too small, for example, making the size of the image of the object the same size as the object viewed by the side mirror it can.

  When the horizontal scaling factor is set in this way, the vertical scale correction processing is performed so that the image of the travel area dividing line in the display image passes through the vanishing point in the horizontal scale corrected image and the horizontal scale correction processing is performed. The vertical scale set in accordance with the horizontal position of the horizontal scale corrected image so as to be an image extending linearly in the same direction as the image of the travel area dividing line in the predetermined range of the scale corrected image. It is preferable that the processing is to compress the portion outside the predetermined range in the horizontal scale corrected image in the vertical direction at the direction scaling factor (second invention).

  According to the second aspect of the invention, in the horizontal scale corrected image, a traveling area dividing line such as a white line (which passes through the vanishing point) that should extend in the front-rear direction of the host vehicle is outside the predetermined range. However, a portion outside the predetermined range of the horizontal scale corrected image is vertically adjusted at the vertical scaling factor set as described above by the vertical scale correction processing. Compressed.

  Thereby, in the horizontal scale corrected image, the distortion of the travel area dividing line that becomes an image bent in the range outside the predetermined range is appropriately corrected by the vertical scale correction process, and the display image in the display image is corrected. It is possible to appropriately perform the image of the travel region dividing line to be an image extending linearly.

  In the second aspect of the invention, more specifically, the vertical scaling factor is the predetermined one of the horizontal scale correction images when the horizontal scaling factor at the vanishing point is used as a reference magnification. At each position in the horizontal direction in the range portion outside the range, the value of the vertical scaling factor is the ratio of the value of the horizontal scaling factor in the range between the position and the vanishing point to the reference magnification. Is set so as to be a magnification of the average value (third invention).

  Thereby, the said vertical direction compression magnification which can correct | amend the curve of the driving | running | working area division line in a horizontal scale correction image appropriately can be set easily.

  In the second or third aspect of the invention, the processing for compressing the portion of the range outside the predetermined range in the horizontal scale correction image by the vertical scale correction processing in the vertical direction is the horizontal scale correction image. The pixels on the horizontal scale corrected image on the reference line extending in the horizontal direction through the vanishing point in the image are not moved in the vertical direction, and the image above and below the reference line are respectively set to the reference line. This is a process of compressing toward the line (fourth invention).

  That is, the reference line extending in the horizontal direction through the vanishing point in the horizontal scale corrected image is a line corresponding to the horizon, and basically in the captured image before the horizontal scale correction processing is performed in the horizontal direction. A straight line extending to An arbitrary straight line (straight line passing through the vanishing point) extending in the same direction as the front-rear direction of the vehicle in the captured image is the predetermined range in the horizontal scale corrected image after the horizontal scale correcting process is performed. Of these, the upper part and the lower part of the reference line bend in opposite directions.

  Therefore, in the vertical scale correction processing, the pixels of the horizontal scale correction image on the reference line extending in the horizontal direction through the vanishing point in the horizontal scale correction image are not moved in the vertical direction, and In addition, the upper image and the lower image are respectively compressed toward the reference line, so that the vertical scale correction process appropriately corrects the curve of the traveling area division line in the horizontal scale corrected image. Can do.

The block diagram which shows the structure of the driving assistance device in one Embodiment of this invention. The figure which shows the structure of the vehicle in embodiment. The figure which shows the imaging area etc. of the camera in embodiment. The flowchart which shows the process of the display image generation part 7 shown in FIG. 5A and 5B are diagrams for explaining the processing of STEP 1 in FIG. 6A and 6B are diagrams for explaining the processing of STEP2 in FIG. 7A to 7C are views for explaining the processing of STEP 3 in FIG. 8A and 8B are diagrams for explaining the processing of STEP 3 in FIG.

  An embodiment of the present invention will be described below with reference to FIGS.

  With reference to FIG. 1, the driving assistance apparatus 1 of this embodiment is mounted on, for example, a right-hand drive vehicle 2 (see FIG. 2). The driving support device 1 includes a camera 3 that captures the rear side of the left side of the vehicle 2, a navigation device 4, a display 5 that displays captured images and navigation information acquired by the camera 3, and a captured image. And an arithmetic processing unit 6 for performing control processing such as display processing of the display 5 and the like.

  The camera 3 is constituted by a CCD camera, a CMOS camera, or the like, and is mounted on a left front side of the vehicle 2, for example, a door mirror 9L as a left side mirror as shown in FIG. The camera 3 captures the rear side of the left side of the vehicle 2 and outputs the captured image data to the arithmetic processing unit 6.

  In addition, although the left and right side mirrors of the vehicle 2 in the present embodiment are door mirrors, they may be attached to the left and right side portions (for example, the body portion on the side of the bonnet) of the front portion of the vehicle body.

  The camera 3 is a camera having a wide angle of view in the lateral direction (direction parallel to the road surface), and the driver of the vehicle 2 can visually recognize the rear side of the left side of the vehicle 2 through the left door mirror 9L. It is possible to image an area wider than the area (mainly a wide area in the lateral direction).

  Specifically, the imaging region (imageable region) of the camera 3 is a region between the straight lines L1 and L6 when viewed in a direction perpendicular to the road surface, for example, as shown in FIG. . This imaging area includes a door mirror area (area between straight lines L3 and L4 in the figure) that is visible to the driver of the vehicle 2 through the left door mirror 9L, and in particular, the door mirror area. This is a region having a wide angle of view on the outside (left side of the straight line L3).

  Therefore, the imaging area of the camera 3 is an area including a door mirror area and a blind spot area outside the door mirror area (an area that the driver cannot visually recognize on the door mirror 9L (an area between the straight lines L2 and L3 in the drawing)). ing. For this reason, as shown in FIG. 3, an object such as another vehicle existing in the blind spot area outside the door mirror area can be imaged by the camera 3.

  The horizontal field angle of the imaging region of the camera 3 is, for example, about 100 deg., And the horizontal field angle of the door mirror region is, for example, about 18-25 deg.

  The navigation device 4 has a function of sequentially detecting the current position of the vehicle 2 (own vehicle 2) by GPS or inertial navigation, a function of collating the detected position of the own vehicle 2 with map data, and the own vehicle 2 to the destination. It is a publicly known device having a navigation function such as a function for setting the guide route.

  The navigation device 4 calculates navigation information including the detected current position of the host vehicle 2, map data around the host vehicle 2 (map image data), a guide route of the host vehicle 2 based on the map data, and the like. Output to the processing unit 6.

  The display 5 is a monitor constituted by a liquid crystal display or the like, and as shown in FIG. 2, the display 5 is mounted on a location in front of the driver's seat in the interior of the vehicle 2 such as a dashboard so that the driver can see it. Has been.

  The arithmetic processing unit 6 is an electronic circuit unit composed of a CPU, a RAM, a ROM, an interface circuit, and the like, and is disposed at an appropriate position of the vehicle 2. In the present embodiment, the arithmetic processing unit 6 generates a display image to be displayed on the display 5 from a captured image of the camera 3 as a function realized by an installed program or the like (the captured image is displayed as a display image). Display image generating unit 7 and display control unit 8 for controlling the display content of display 5.

  The display image generation unit 7 corresponds to the display image generation means in the present invention, and detailed processing thereof will be described later.

  In more detail, the display control unit 8 appropriately displays an image to be displayed on the display unit 5 as a display image (hereinafter referred to as a display captured image) generated by the display image generation unit 7 and the navigation device 4. The display content of the display 5 is controlled, for example, by switching from one to the other navigation image (a map image, an image including a mark of the host vehicle 2, a guidance route mark, etc.) representing the navigation information input from.

  In this case, the display control unit 8 usually displays a navigation image on the display device 5. Then, when a display changeover switch (not shown) is operated by the driver, or when the blinker on the left side of the vehicle 2 is turned on (or the path to the left side of the vehicle 2 based on the measured value of the steering angle of the vehicle 2 or the like) When a request for displaying the display captured image is generated as in the case where a change is detected, the display captured image is displayed on the display 5.

  Note that the arithmetic processing unit 6 may be built in the camera 3. Further, the display image generation unit 7 and the display control unit 8 may be provided in different arithmetic processing units. In this case, for example, the display image generation unit 7 may be provided in an arithmetic processing device built in the camera 3, and the display control unit 8 may be provided in an arithmetic processing device of a navigation device or an audio device.

  Next, detailed processing of the display image generation unit 7 of the arithmetic processing unit 6 will be described below.

  A captured image of the camera 3 is given to the display image generation unit 7 of the arithmetic processing unit 6 at a predetermined control processing cycle. In this case, in this embodiment, the captured image given to the display image generation unit 7 is more specifically an image after correcting the distortion of the image due to the lens distortion of the camera 3 by a known correction processing method. is there.

  When the lens distortion is sufficiently small, the captured image provided to the display image generation unit 7 may be the captured image itself output from the camera 3.

  Then, the display image generating unit 7 generates a display captured image by performing predetermined processing (image processing) on the captured image of the given camera 3.

  The generation process is executed as shown in the flowchart of FIG.

  In the following, the display image generation unit 7 first executes the processing of STEP1. In STEP 1, the display image generation unit 7 acquires a display target captured image that is a partial image from a given captured image.

  The display target captured image means a captured image of a region to be displayed on the display 5 (hereinafter referred to as a display target region) out of the entire captured image of the camera 3.

  As shown in FIG. 3 or FIG. 5A, the display target area is included in the imaging area of the camera 3, and the horizontal angle of view is slightly smaller than the angle of view of the imaging area of the camera 3. It is. In this case, the horizontal angle of view of the display target area is, for example, about 80 deg. And this display object area | region includes the door mirror area | region in the location near the side surface of the own vehicle 2 of this area | region. And the whole area | region adjacent to the outer side of the door mirror area | region of the display object area | region is included as a blind spot area | region which cannot be visually recognized on the door mirror 9L among this display object area | region.

  Note that the horizontal angle of view of the display target area may match the angle of view of the imaging area of the camera 3. Further, of the horizontal boundaries of the display target area, the boundary close to the host vehicle 2 may coincide with the boundary on the inner side (right end side) of the door mirror area. In addition, the angle of view in the vertical direction of the display target area may be, for example, the same as or twice the angle of view of the area in the vertical direction (vertical direction) that can be visually recognized by the driver with the door mirror 9L.

  In STEP 1, the display image generation unit 7 displays the entire image of the display target area in the captured image of the camera 3 with a predetermined size (display on the display 5) as shown in FIG. (Screen size) is acquired as a display target captured image.

  Note that the size of the entire image of the display target area is corrected by scaling (enlarging or compressing) the entire image of the display target area in both the horizontal direction and the vertical direction at a predetermined magnification. .

  Next, the display image generation unit 7 executes the processing of STEP2. In STEP 2, the display image generation unit 7 generates a horizontal scale correction image by executing a horizontal scale correction process that is a process of correcting the scale of the display target captured image in the horizontal direction.

  This horizontal scale correction processing is performed by partially expanding and compressing the display target captured image in the horizontal direction while maintaining the entire horizontal angle of view of the display target captured image. Is a process for generating

  Here, since the display target captured image acquired in STEP 1 is an image of a display target region having a wide angle of view (horizontal angle of view) including the blind spot region outside the door mirror region, the display target captured image is displayed. Is displayed on the display 5 having a normal screen size as it is, the image size of an object such as another vehicle existing in the door mirror area in the display image becomes the size of the object visually recognized on the door mirror 9L. Compared to this, it becomes excessively small and difficult to see.

  Further, when the display target captured image is displayed as it is on the display device 5, the object and the own vehicle that are visually recognized from the image size of an object such as another vehicle existing in the door mirror region in the image. 2 (distance in the front-rear direction of the host vehicle 2) is likely to deviate from the distance recognized by the driver from the size of the object visually recognized on the door mirror 9L.

  Therefore, in the horizontal scale correction process of STEP 2, among the display target captured images, the captured image of the door mirror region is mainly enlarged in the horizontal direction, while the horizontal angle of view of the horizontal scale corrected image is the same as the display target captured image. In order to maintain (in other words, to make the horizontal range of the area projected on the horizontal scale corrected image the same as the horizontal range of the area projected on the display target captured image (display target area)), The captured image of the blind spot area that is not reflected on the door mirror 9L is compressed in the horizontal direction.

  In this case, in the present embodiment, the entire area (display area) of the display screen of the display 5 that displays the display target captured image is a predetermined line up in order in the horizontal direction as shown in FIG. It is divided into a plurality (N) of local areas A (1) to A (N) having a width. As shown in the graph of FIG. 6B, the horizontal compression rate CRh representing the horizontal expansion / contraction magnification of the image of each local region A (i) (i = 1, 2,..., N). Further, it is set in advance so as to change according to the position of each local area A (i) in the horizontal direction. In the following description, the value of the horizontal compression rate corresponding to the position of the i-th local area A (i) may be expressed as CRh (i).

  Here, the horizontal compression rate CRh (i) corresponding to the position of each local area A (i) is determined based on whether the image in the corresponding local area A (i) is the original image before the horizontal scale correction processing (input) This is an enlargement / reduction ratio which means that the image is a scale-corrected image whose width is 1 / CRh (i) times. Therefore, the horizontal compression rate CRh (i) is a value for compressing the original image in the corresponding local area A (i) in the horizontal direction when the value is larger than “1”. When the value is smaller than “1”, the magnification is used to enlarge the original image in the corresponding local area A (i) in the horizontal direction.

  The horizontal compression ratio CRh shown in the graph of FIG. 6B in the present embodiment is the local area A (i) in the horizontal direction range B1 in the figure in the entire area of the display screen. ) Is set to “1” so that the width of the original image does not change. The horizontal range of B1 is a range of images closest to the host vehicle 2 in the display target captured image, and is a region having an image of the side surface of the host vehicle 2.

  Further, in the lateral range of B2 adjacent to the outside of the lateral range of B1 (the side farther from the host vehicle 2 than B1), the image of each local region A (i) within that range is the original image. In order to obtain an image with an enlarged horizontal width, the value of the horizontal compression rate CRh is set to a value smaller than “1”. The range of B2 is a range including an image of the door mirror area (excluding an image in the lateral range of B1) and an image of a partial area near the door mirror area in the blind spot area outside the door mirror area. .

  In this case, the lateral compression rate CRh in the lateral direction range of B2 is set to be kept at a predetermined constant value (for example, 0.8) in the most part of the door mirror region (the range of C1a in the figure). Has been. Hereinafter, the range of C1a is referred to as a constant lateral compression rate range C1a.

  And, in the range closer to the host vehicle 2 than the constant lateral compression rate range C1a in the lateral range of B2, the lateral compression rate CRh becomes the above-mentioned constant value as it approaches the host vehicle 2 in the lateral direction. Is set to gradually approach “1” (value in the horizontal range of B1). Similarly, in the range in the lateral direction of B2, which is farther from the own vehicle 2 than the constant lateral compression rate range C1a, the lateral compression rate CRh becomes constant as the distance from the own vehicle 2 increases in the lateral direction. It is set so as to gradually approach “1” from the value.

  In addition, within the lateral range of B3 (the lateral range other than B1 and B2 in the entire area of the display screen) adjacent to the outside of the lateral range of B2 (the side farther from the host vehicle 2 than B2), In order for the image of each local area A (i) within the range to be an image obtained by compressing the width of the original image, the value of the horizontal compression rate CRh is set to a value of “1” or more. Yes. The horizontal range of B3 is a range including an image of an area other than the area within the horizontal range of B2 in the blind spot area. In this case, the value of the lateral compression rate CRh in the lateral range of B3 is set to gradually increase as the distance from the host vehicle 2 increases in the lateral direction.

  The horizontal compression rate CRh is an average value of CRh (i) corresponding to each local region A (i) (i = 1, 2,..., N) (= (CRh (1) + CRh (2) +... ... + CRh (N)) / N) is set to "1".

  In the horizontal scale correction process of STEP 2, the display image generation unit 7 performs image processing for each local area A (i) (i = 1, 2,..., N) according to the horizontal compression rate CRh set in advance as described above. A horizontal scale corrected image is generated by correcting the scale of the horizontal width of the original image. In this case, pixel interpolation associated with enlargement or compression of the image is required. As the pixel interpolation method, a known method such as bilinear interpolation or bicubic interpolation is used.

  Thereby, a horizontal scale corrected image as shown in FIG. 6A is obtained. The overall size (horizontal width and vertical width) of the horizontal scale corrected image is the same as the display target captured image. The horizontal scale corrected image is an image obtained by projecting a region (display target region) within the same horizontal angle of view as the display target captured image, as shown in the figure, while the image in the door mirror region is the display target. The captured image is enlarged in the horizontal direction.

  Thereby, in the horizontal scale corrected image, the size of an object such as another vehicle existing in the door mirror region can be made substantially the same size as the size on the door mirror 9L.

  In the present embodiment, the change in the horizontal compression rate CRh between the horizontal range of B3 and the range close to B3 of the horizontal range of B2 is an object such as another vehicle in the display target area. However, when the vehicle approaches the host vehicle 2 at a constant speed, the lateral movement speed of the object in the image of the door mirror area of the horizontal scale correction image and the image of the blind spot area The moving speed of the object in the lateral direction is set to be substantially constant.

  Next, the display image generation unit 7 executes the processing of STEP3. In STEP 3, the display image generation unit 7 finally displays the image on the display 5 by executing a vertical scale correction process that is a process of correcting the vertical scale of the horizontal scale corrected image generated in STEP 2. A vertical scale correction image as a display captured image to be generated is generated.

  This vertical scale correction processing is performed by partially compressing the horizontal scale corrected image in the vertical direction of the horizontal scale corrected image (compressing the horizontal scale corrected image in the vertical direction in a predetermined horizontal range), This is processing for generating a vertical scale corrected image.

  Here, in the horizontal scale correction processing in STEP 2, in particular, the lateral compression rate CRh in the blind spot area is increased as the distance from the host vehicle 2 increases. The image of the travel area dividing line such as a white line that should extend in a straight line is bent in the horizontal scale corrected image.

  For example, as shown in FIG. 6 (a), among the white lines as the travel area parting lines of the road on which the host vehicle 2 is traveling, the driving area parting line 21 existing in the blind spot area (in FIG. Among the roads on which No. 2 is traveling, the image of the white line near the roadside zone) is curved. Note that the image of the white line 22 existing in the door mirror region (in FIG. 6A, the dotted white line between the traveling lane of the host vehicle 2 and the traveling lane on the left side thereof) It extends linearly in substantially the same direction.

  Therefore, in the vertical scale correction processing of STEP 3, the image of the blind spot area and the area close to the blind spot area of the door mirror area (the change rate of the horizontal compression ratio CRh in the horizontal direction is relatively small). By compressing the image corresponding to the large area) in the vertical direction, the image of the running area dividing line bent in the blind spot area in the horizontal scale corrected image is a straight line substantially in the same direction as the front-rear direction of the host vehicle 2. To correct the image.

  In this case, in this embodiment, as in the case of the processing in STEP 2, the entire area of the display screen of the display 5 that displays the vertical scale corrected image is sequentially in the horizontal direction as shown in FIG. It is divided into N local areas A (1) to A (N) having a predetermined width arranged adjacent to each other. Then, the vertical compression rate CRv representing the vertical expansion / contraction ratio of the image of each local region A (i) (i = 1, 2,..., N) is shown in the graph of FIG. Further, it is set so as to change according to the position of each local area A (i) in the horizontal direction of the display screen. In the following description, the value of the vertical compression rate corresponding to the position of the i-th local area A (i) may be expressed as CRv (i).

  Here, the vertical compression rate CRv (i) corresponding to the position of each local area A (i) is the vertical width of the image in the corresponding local area A (i) in the horizontal scale corrected image 1 / CRv (i This is a scaling factor that means that the scale is corrected to a double size.

  The vertical compression ratio CRv shown in the graph of FIG. 7C in the present embodiment is the local area A within the horizontal direction range C1 in the figure of the entire area of the display screen. It is set to “1” so as not to change the vertical width of the image of (i). The lateral range of C1 is the position of the boundary (the value of CRh is close to the blind spot region) in the boundary of the lateral compression rate constant range C1a where the value of the lateral compression rate CRh is kept constant in the door mirror region. , A position closer to the host vehicle 2 than the position where the vehicle starts to increase as the distance from the host vehicle 2 increases.

  In addition, in the horizontal range of C2 on the blind spot area side of C1 (the horizontal range other than C1 in the entire area of the display screen), the vertical width of the image of each local area A (i) within that range Is compressed to a value larger than “1”. The horizontal range of C2 is a range including an image of the blind spot area and an image of a part of the door mirror area near the blind spot area.

  In this case, the longitudinal compression rate CRv at the position of each local region A (i) within the lateral range of C2 is set to gradually increase as the distance from the host vehicle 2 increases.

  In the present embodiment, the value of the vertical compression rate CRv (i) at the position of each local area A (i) (more specifically, in the horizontal scale correction image, a white line parallel to the front-rear direction of the host vehicle 2 or the like) The longitudinal compression ratio CRv (i) in the range farther from the own vehicle 2 than the vanishing point P, which is the intersecting point (the range of C3 in FIG. 7C), is the position of the vanishing point P ( The horizontal compression ratio CRh in the range from the position in the horizontal direction) to the position of each local area A (i) (the position in the horizontal direction) and the value of the horizontal compression ratio CRh in the position of the vanishing point P Accordingly, the following equation (1) is set.

  Here, referring to FIG. 7C, CRv (x (n)) in the above equation (1) is the vertical compression corresponding to the local region A (i) whose lateral position is x (n). CRh (x (j)) (j = 1, 2,..., N) is a value of the horizontal compression rate CRh at the horizontal position of x (j). Note that x (1), x (2),..., X (n) indicate the lateral position x (0) of the vanishing point P (local region A including the vanishing point P) in the direction away from the side surface of the host vehicle 2. (i) position) are arranged in order at regular intervals. Further, in the present embodiment, the lateral position x (0) of the vanishing point P is a position within the lateral compression rate constant range C1a in the door mirror region.

  Further, CRh_STD in the equation (1) is a value of the lateral compression rate CRh at the lateral position x (0) of the vanishing point P. Hereinafter, this CRh_STD is referred to as a reference lateral compression rate. In this embodiment, the reference lateral compression rate CRh_STD matches the value of the lateral compression rate CRh in the constant lateral compression rate range C1a in the door mirror region.

  Therefore, in the present embodiment, the vertical compression rate CRv (x (n)) corresponding to the local region A (i) whose horizontal position is x (n) is equal to the horizontal position x (n) and disappearance. It is set to coincide with the average value of the ratio of the lateral compression rate CRh to the reference lateral compression rate CRh_STD in the range between the point P and the lateral position x (0).

  In this case, in the constant horizontal compression rate range C1a in which the horizontal compression rate CRh is kept constant at the value of the reference horizontal compression rate CRh_STD, the vertical compression rate CRv ( x (n)) is “1”. In addition, the value of the longitudinal compression rate CRv (i) corresponding to the local area A (i) existing in the range closer to the host vehicle 2 than the vanishing point P is “1” regardless of the above formula (1). Is set.

  Supplementally, when the value of the longitudinal compression ratio CRv (i) is set by the above equation (1), it is necessary to specify the position of the vanishing point P. For example, the specification can be performed as follows. it can. That is, for example, a travel area parting line (for example, white line 22 in FIG. 7A) and a horizon line on the side close to the host vehicle 2 are extracted from the horizontal scale corrected image, and an auxiliary line (straight line) drawn on the traveling area parting line is drawn. And the position of the intersection of the auxiliary line (straight line) drawn on the horizon is specified as the position of the vanishing point.

  In addition, as a traveling area dividing line to be referred to in order to specify the position of the vanishing point, a traveling area dividing line (for example, a white line 21 in FIG. 7A) that bends within the range C2 of the horizontal scale corrected image is used. It is also possible. In this case, the position of the intersection point of the auxiliary line (straight line) having the same inclination as the travel area dividing line and the auxiliary line (straight line) drawn on the horizon in the constant lateral compression rate range C1a is the position of the vanishing point P. It may be specified as

  It is also possible to specify a straight line parallel to the front-rear direction of the host vehicle 2 by Hough transformation or the like, and to specify the position of the intersection of these straight lines as the position of the vanishing point P.

  In the vertical scale correction processing in STEP 3, the display image generation unit 7 performs vertical image correction for each local area A (i) (i = 1, 2,..., N) according to the vertical compression rate CRh set as described above. A vertical scale corrected image is generated by scaling the width size.

  In this case, among the horizontal scale corrected images, the images of the local areas A (i) included in the C2 horizontal range in which the vertical compression ratio CRv is set to a value larger than “1” are compressed in the vertical direction. Is done. Then, in the compression of the image of each local area A (i) in the lateral range of C2, as shown in FIG. 8A, a compressed reference line Lcrv extending in the lateral direction as a line corresponding to the horizon ( The image of the local area A (i) is the local area A (i) so that the pixels on the reference line extending in the lateral direction through the vanishing point P are maintained at positions on the compressed reference line Lcrv. ) In the vertical direction at a magnification of the vertical compression rate CRv corresponding to the position of.

  Therefore, the image above the compression reference line Lcrv (the image in the vertical range of D1 in FIG. 8A) is compressed downward toward the compression reference line Lcrv, and is an image below the compression reference line Lcrv. (Image in the vertical range of D2 in FIG. 8A) is compressed upward toward the compression reference line Lcrv.

  In this case, among the horizontal scale corrected images, images in the horizontal range of C2 are compressed in the vertical direction, while images other than the horizontal range of C2 are not compressed in the vertical direction. The vertical width of the inner image after compression is shorter than that of the image outside the horizontal range of C2. Therefore, on the upper side and the lower side of the compressed image within the lateral range of C2, images adjacent to the display target region on the upper side and lower side of the display target region of the captured image of the camera 3, respectively, The vertical compression is replenished.

  By the above vertical scale correction processing, a vertical scale corrected image as shown in FIG. 8B is obtained. The overall size (horizontal width and vertical width) of this vertical scale corrected image (specifically, the vertical scale corrected image after image replenishment on the upper and lower sides of the compressed image within the horizontal range of C2) is The horizontal scale correction image and the display target captured image are the same. In this case, by setting the vertical compression ratio CRv within the horizontal range of C2 as described above, the white line 21 as the running area dividing line that has been bent in the blind spot area of the horizontal scale corrected image is It is corrected to a line that extends linearly in the same direction as the front-rear direction.

  In this case, a line extending in the same direction as the front-rear direction of the host vehicle 2 in the real space, such as a guard rail, a curb, or a building edge, is not limited to the travel area dividing line, and a vertical scale corrected image (imaging for display) The image is corrected to a line extending in a straight line.

  In the present embodiment, the vertical scale correction image generated as described above is a display captured image to be displayed on the display 5. Then, when a request for displaying the captured image of the camera 3 on the display 5 is generated, the display control unit 8 displays the captured image for display (vertical scale corrected image) generated by the display image generation unit 7 as described above. ) Is displayed on the display 5.

  Supplementally, the vertical scale correction process is a process of correcting the scale of the horizontal scale correction image so that the line that should extend in parallel with the front-rear direction of the host vehicle 2 is linear in the captured image for display. In a real space in the blind spot area, a straight line (excluding a straight line passing through the vanishing point) that extends in the lateral direction of the host vehicle 2 (the vehicle width direction of the host vehicle 2) is a captured image for display (vertical scale correction). In the image).

  However, the straight curve extending in the lateral direction of the host vehicle 2 is relatively noticeable in the area near the side of the host vehicle 2 in the blind spot area, and there is an object such as another vehicle in the spot. In this case, in order to avoid contact between the object and the host vehicle 2, it is important that the driver can recognize that the object is present at that location. Therefore, the bending of the horizontal line of the object in the captured image for display has no problem in practice.

  According to the present embodiment described above, the display captured image displayed as the captured image of the camera 3 on the display 5 is, as described above, the B2 including most of the door mirror area of the display target captured image. The display target captured image is expanded in the horizontal direction in the horizontal range, and the display target captured image is compressed in the horizontal direction in the B3 horizontal region included in the blind spot region.

  For this reason, the size of the image of an object such as another vehicle existing in the door mirror region is maintained on the door mirror 9L while the horizontal angle of view of the captured image for display is kept in a wide range in the horizontal direction. It can be as large as the size. For this reason, the distance (the distance in the front-rear direction) between the object visually recognized from the size of the image of the object in the display captured image and the distance of the host vehicle 2 on the door mirror 9L. It can be matched to the distance visually recognized from the size.

  Further, in any of the door mirror area and the blind spot area in the captured image for display, the outline of the object relative to the host vehicle 2 is determined based on the lateral movement speed of the object such as another vehicle in the image. It is possible to visually recognize a typical moving speed (moving speed in the front-rear direction of the host vehicle 2) in the same manner.

  Furthermore, a travel line, such as a white line, that should extend linearly in substantially the same direction as the front-rear direction of the host vehicle 2 does not bend significantly in the captured image for display, and is substantially the same as the front-rear direction of the host vehicle 2. It will extend linearly in the same direction.

  As a result, the driver of the host vehicle 2 visually recognizes the captured image for display, so that the situation behind the host vehicle 2 including the blind spot area, for example, the other vehicle and the host vehicle 2 existing in the blind spot area, The relative positional relationship, the motion state of the other vehicle with respect to the host vehicle 2 and the like can be appropriately confirmed without a sense of incongruity.

  In the above-described embodiment, the case where the display image generated from the left rear side captured image of the host vehicle 2 is displayed on the display 5 is described as an example. In the case of a vehicle, a captured image of the rear side of the right side of the host vehicle 2 is acquired by a camera, and a display image (scaled in the horizontal and vertical directions) generated from the captured image in the same manner as in the above embodiment. The image that has been subjected to the above may be displayed on the display 5.

  Further, both the right and left rear captured images of the host vehicle 2 are acquired by the camera, and the display image generated from the right rear lateral captured image and the left rear side captured image are obtained. The display image generated from the captured image may be selectively displayed on the display 5.

  In the above-described embodiment, the compression ratio is used to represent the horizontal expansion / contraction magnification and the vertical expansion / contraction ratio, but an inverse value (that is, an expansion ratio) may be used instead of the compression ratio.

  In the embodiment, an example in which a display image is generated from a captured image (or a captured image of the camera 3 in which the lens distortion is sufficiently small) obtained by correcting the distortion of the image due to the lens distortion of the camera 3 is taken as an example. As described above, a display image may be generated by performing the horizontal scale correction process and the vertical scale process on a captured image that has not been corrected for distortion caused by lens distortion.

  In this case, the reference line for compressing the image in the vertical direction in the vertical scale processing (reference line corresponding to the compression reference line Lcrv) in the captured image before the horizontal scale correction processing is used. Of the straight line group that extends in the same direction as the front-rear direction, it is desirable to use a line that has as high a linearity as possible in the horizontal scale corrected image and that extends in the horizontal direction on the screen. In this case, a horizontal line drawn at a position where the direction of the straight curve on the image is switched may be used as the reference line.

  The vertical compression rate CRv is preferably set based on an arithmetic expression or a data table set in consideration of image distortion caused by lens distortion instead of setting based on the expression (1).

  In the embodiment, the horizontal and vertical angles of view of the display target captured image among the captured images of the camera 3 are always constant, but are changed according to the driving situation of the vehicle 2 and the like. Also good. For example, when the host vehicle 2 is moved backward and parked in a predetermined parking area, the other vehicle parked on the side of the host vehicle 2 or the wall existing on the side of the host vehicle 2 and the host vehicle 2 In order to make it easy to confirm the positional relationship in the horizontal direction, display target imaging is performed when the vehicle speed of the host vehicle 2 is equal to or lower than a predetermined value or when the shift position of the host vehicle 2 is set to a reverse position. You may make it make the angle of view of the horizontal direction of an image narrower than the angle of view at the time of the normal driving | running | working of the own vehicle 2 (the same angle of view as the said embodiment).

  In this case, when a display image is generated from a display target captured image with a narrowed angle of view in the horizontal direction, the horizontal position of the horizontal scaling factor (for example, the horizontal compression ratio CRh) in the horizontal scale correction processing is set. The corresponding change width is made smaller than in the case of generating a display image from a display target captured image with a wide horizontal angle of view, and the magnification for horizontal expansion and compression is made closer to “1”. You may make it set to a value.

  DESCRIPTION OF SYMBOLS 1 ... Driving assistance device, 2 ... Vehicle, 3 ... Camera, 5 ... Display, 7 ... Display image generation part (display image generation means).

Claims (4)

A camera that captures the rear side of the vehicle, a display image generating unit that generates a display image to be visually recognized by the driver from a captured image of the camera, and a display image generated by the display image generating unit A vehicle driving support device comprising a display device,
The display image generating means performs horizontal scale correction for correcting the scale of each part of the captured image in the horizontal direction at a horizontal scaling factor set so that the value changes according to the position in the horizontal direction of the captured image of the camera. And a travel area dividing line that extends linearly in the front-rear direction of the vehicle in real space, and is curved and displayed in the horizontal scale correction image generated from the captured image by the horizontal scale correction processing. The horizontal scale corrected image is partially vertically aligned with the vertical scaling factor set so that the image of the travel area dividing line is an image extending linearly in the longitudinal direction of the vehicle in the horizontal scale corrected image. A vehicle driving support device, characterized in that the display image is generated by executing a vertical scale correction process for correcting the scale. The vehicle driving support device according to claim 1,
The horizontal enlargement / reduction ratio in the horizontal scale correction processing is set so as to become a predetermined constant magnification value for enlargement in a predetermined range in the horizontal direction in an area visually recognized by the side mirror of the vehicle in the captured image of the camera. In addition, in a range outside the predetermined range, it is set to change from the constant magnification value to a compression magnification value according to the position in the lateral direction of the captured image,
In the vertical scale correction processing, the image of the travel area dividing line in the display image passes through the vanishing point in the horizontal scale correction image, and the travel area in the predetermined range of the horizontal scale correction image. Of the horizontal scale corrected images, the vertical scale magnification is set according to the horizontal position of the horizontal scale corrected image so that the image extends linearly in the same direction as the image of the dividing line. A driving support apparatus for a vehicle, characterized in that it is a process of compressing a portion of a range outside a predetermined range in a vertical direction. The vehicle driving support device according to claim 2,
The vertical scaling factor is a horizontal scaling factor in a portion outside the predetermined range of the horizontal scale correction image when the horizontal scaling factor at the vanishing point is used as a reference magnification. At the position, the value of the vertical scaling factor is set to be the magnification of the average value of the ratio of the horizontal scaling factor to the reference magnification in the range between the position and the vanishing point. A vehicle driving support device characterized by the above. The vehicle driving support device according to claim 2 or 3,
By the vertical scale correction process, the process of compressing the portion of the range outside the predetermined range in the horizontal scale correction image in the vertical direction extends in the horizontal direction through the vanishing point in the horizontal scale correction image. A process of compressing an image above and below the reference line toward the reference line while preventing the pixels of the horizontal scale corrected image on the reference line from moving vertically. A vehicle driving support device.