JP2011057101A - Obstacle detector for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an obstacle detector for a vehicle which generates by itself a state view of an overlooked visual point that can use a plurality of images recorded by a camera to discriminate an obstacle from an area except the obstacle. <P>SOLUTION: The obstacle detector for the vehicle includes an imaging part which can pick-up an image of the periphery of the vehicle, an obstacle detecting part for detecting the obstacle included in the image of the periphery of the vehicle recorded by the imaging part, an overlooked image generating part for converting the image of the periphery of the vehicle recorded by the imaging part to an overlooked image viewed from a prescribed virtual visual point and outputting the overlooked image and a state view generating part for converting the overlooked image to an image in which the image of the obstacle included in the overlooked image can be discriminated from the image except the obstacle to generate the state view of the overlooked visual point showing the state of the periphery of the vehicle and a display part for displaying the state view. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a vehicle obstacle detection device that mounts a camera on a vehicle, creates a situation diagram including obstacles around the vehicle based on an image taken by the camera, and warns as necessary.

  Conventionally, a vehicle equipped with a plurality of cameras, taking images of the surroundings of the vehicle, displaying the taken images on a display provided on the vehicle, etc., so as to avoid the vehicle from colliding with or colliding with an obstacle There is a detection device. In addition, recently, there is an image obtained by combining images taken by a plurality of cameras to generate an overhead image and displaying it like a top view image. By displaying the bird's-eye view image, it becomes easier to grasp the host vehicle and the surrounding situation, and the driving of the driver can be supported. Note that the bird's-eye view image is an image viewed from a virtual viewpoint set above the host vehicle, and the appearance of the bird's-eye view image can be changed by changing the position of the virtual viewpoint.

  By the way, in the conventional obstacle detection device, when an obstacle is detected, a warning is issued to the driver so as to avoid a collision. However, obstacles such as garage walls, tools in the garage, and curbstones are Since the three-dimensional object is known to the driver, issuing a warning each time the three-dimensional object is recognized as an obstacle may cause an excessive warning to the driver. In addition, if it is possible to self-generate a situation diagram of an overhead view that can distinguish an obstacle from other areas, road information can be presented to the driver after passing the same road. However, there was no effective means of presentation.

  Patent Document 1 discloses an image processing apparatus that inputs images taken by a plurality of cameras, generates an overhead image viewed from a virtual viewpoint, and displays an illustration image showing the host vehicle.

  Patent Document 2 discloses an obstacle detection device that detects a three-dimensional obstacle or a moving object, and displays the detected obstacle on a display or warns the driver by voice or the like. I have to. However, in the example of Patent Document 2, a warning is also given for an environmental solid object already recognized by the driver, such as a curb of a parking lot or a garage wall, which may be a troublesome warning for the driver.

  In Patent Document 3, the presence of a road structure at the current position of the host vehicle is determined based on GPS and map information. If a road structure exists, the sensitivity of the sensor with respect to the direction is reduced. There is disclosed an obstacle detection device that prevents annoying warnings to a driver. However, in the example of Patent Document 3, it is necessary to input map information from the outside, and there is a problem that obstacles such as garages that are not included in the map information cannot be detected.

Japanese Patent No. 3300334 JP 2003-44996 A JP 2006-123569 A

  In the conventional obstacle detection device, when an obstacle is detected, a warning is issued to the driver. However, every time a three-dimensional object known to the driver is recognized as an obstacle, a warning is issued to the driver. It can be an excessive warning. Further, when determining the presence of a road structure using map information, it is necessary to input map information from the outside, and there is a problem that obstacles such as garages that are not included in the map information cannot be detected.

  In view of such circumstances, the present invention uses a plurality of images taken by a camera to self-generate an overhead view situation diagram that can identify an obstacle and other regions, and has already been recognized by the driver. It is an object of the present invention to provide an obstacle detection device for a vehicle in which an excessive warning is not detected when a three-dimensional environmental object is detected.

  The obstacle detection device for a vehicle according to claim 1, wherein an imaging unit capable of photographing the periphery of the vehicle, an obstacle detection unit that detects an obstacle included in an image around the vehicle photographed by the imaging unit, and the photographing A bird's-eye view image generation unit that converts and outputs an image of the vehicle surroundings taken by the unit into a bird's-eye view image viewed from a predetermined virtual viewpoint, and an obstacle image included in the bird's-eye view image and an image other than the obstacle can be identified A situation diagram generating unit that converts the image into an image and creates a situation diagram of a bird's eye view that represents the situation around the vehicle, and a display unit that displays the situation diagram are provided.

  The vehicle obstacle detection device according to claim 4 is an imaging unit capable of capturing an image of the periphery of the vehicle, and a first obstacle detection unit that detects an obstacle included in an image around the vehicle captured by the imaging unit. A bird's-eye view image generation unit that converts and outputs an image around the vehicle photographed by the photographing unit into a bird's-eye view image viewed from a predetermined virtual viewpoint, and an obstacle image included in the bird's-eye view image and an image other than the obstacle A situation diagram generating unit that converts an image into an identifiable image, creates a situation diagram of an overhead view that represents the situation around the vehicle, and updates the situation diagram according to the progress of the vehicle, and the situation diagram The storage unit that stores the situation diagram created by the generation unit, the situation diagram stored in the storage unit, and the situation diagram updated and created by the situation diagram generation unit are compared to detect the presence of a new obstacle When the second obstacle detection unit and the situation diagram are displayed Moni, characterized by comprising a display and a warning unit for issuing a warning when it detects the presence of at least the new obstacle.

  According to the vehicle image display device of the present invention, it is possible to accurately notify the driver of the situation around the vehicle and to assist safe driving. Further, it is possible to prevent the driver from being excessively warned.

The block diagram which shows the structure of the image display apparatus for vehicles which concerns on one Embodiment of this invention. The block diagram which shows the structure of the image process part in one Embodiment. Explanatory drawing explaining schematic operation | movement of this invention. Explanatory drawing which shows the example of creation of the situation figure in the case of stopping a vehicle in a parking lot. Explanatory drawing which showed the production | generation of the bird's-eye view image 31 concretely. Explanatory drawing which showed the production | generation of the situation figure 33 concretely. Explanatory drawing which shows the example of preparation of the situation figure at the time of a vehicle passing a narrow bend. Explanatory drawing which showed the production | generation of the bird's-eye view image 51 concretely. Explanatory drawing which showed the production | generation of the situation figure 53 concretely. Explanatory drawing explaining the driving assistance using the situation figure 53. FIG. The flowchart which shows creation operation of a situation diagram. The flowchart which shows the update operation of a situation figure. The block diagram which shows the structure of the obstacle detection apparatus for vehicles which concerns on the 2nd Embodiment of this invention. Explanatory drawing explaining operation | movement of 2nd Embodiment.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a vehicle obstacle detection device according to an embodiment of the present invention. The vehicle image display apparatus of FIG. 1 includes an imaging unit 11, an image processing unit 12, a vehicle information generation unit 13, a storage unit 14, and a display / warning unit 15.

  The imaging unit 11 includes a plurality of cameras 111, 112, 113... Attached to the vehicle 10. The plurality of cameras 111, 112, 113... Captures surrounding images such as the front, rear, left side, and right side of the vehicle, and a fisheye camera may be used.

  The image processing unit 12 uses a plurality of images obtained by photographing the periphery of the vehicle in time series so that an environmental solid object that always exists as a part of the environment can be distinguished from a plane that is not an obstacle such as a road surface or a white line. Self-generated situation diagram of viewpoint. The situation diagram of the bird's-eye view is generated by converting a previously captured image into a viewpoint from above, and rotating / parallel-moving multiple images excluding the obstacle area based on the position, direction, and angle of the vehicle at the time of shooting. The plurality of images are generated by weighted averaging. Details of the image processing unit 12 will be described later.

  The vehicle information generation unit 13 includes a vehicle speed meter 131, a gyro 132, and a GPS device 133. The speedometer 131 generates vehicle speed information of the host vehicle, the gyro 132 detects the tilt of the host vehicle, and the GPS device 133 has a recording medium storing map data, a search unit for searching a route to the destination, and the like. is doing. Information such as the current position, direction and angle of the host vehicle, the traveling direction (forward, back, right / left turn) and the like can be obtained from the vehicle information generation unit 13, and the information is transmitted to the image processing unit 12.

  The storage unit 14 stores the situation diagram generated by the image processing unit 12. The display / warning unit 15 displays an image output from the image processing unit 12 and is configured by a liquid crystal display or the like. It also issues warnings as necessary.

  FIG. 2 is a block diagram illustrating a configuration of the image processing unit 12. The image processing unit 12 includes an input unit 121, an obstacle detection unit 122, an overhead image generation unit 123, a rotation / movement unit 124, an image adjustment unit 125, and a situation diagram generation unit 126.

  The input unit 121 inputs an image around the vehicle photographed by the camera 11. The obstacle detection unit 122 detects obstacles such as a three-dimensional object and a moving object from the image input to the input unit 121. For example, an image captured by the camera 11 is processed, and an object or a moving body protruding from the road surface is determined.

  The overhead image generation unit 123 converts the output image from the obstacle detection unit 122 into, for example, an overhead image viewed from a virtual viewpoint set above the host vehicle. In order to convert to a viewpoint from above, for example, projective conversion is performed on the assumption that all pixels of the image are on the road surface. That is, a front image, a rear image, a left side image, a right side image, and the like of the vehicle photographed by the camera 11 are combined to generate a background image, and the image of the host vehicle is combined with the background image to An overhead image (top view image) viewed from the viewpoint is generated. The overhead image changes by changing the position of the virtual viewpoint. Note that details of the method for generating the overhead image are described in, for example, Patent Document 1.

  In addition, the overhead image generation unit 123 converts the pixel value of the obstacle area detected by the obstacle detection unit 122 into a specific value (for example, 0 value) for the converted overhead image, and the pixels in the other areas are converted. An image normalized so as to have a value different from the obstacle area (hereinafter referred to as a normalized image) is generated.

  The rotation / movement unit 124 creates an image obtained by rotating and translating the normalized image from the position, direction, and angle of the host vehicle based on the information from the vehicle information generation unit 13. In addition, when rotating and moving in parallel, the correlation value is calculated by block matching between the situation diagram and the rotated / translated normalized image while slightly changing the rotation amount and the parallel movement amount, and the highest correlation value is obtained. The rotation amount and the parallel movement amount are obtained, the measured values of the vehicle position / direction and angle are corrected, and the rotation and the parallel movement are performed using the corrected values.

  The image adjustment unit 125 performs contrast adjustment and blurring processing on the image converted by the rotation / movement unit 124. The situation diagram generation unit 126 creates a situation diagram by performing a weighted average process on the already created situation diagram and the newly created situation diagram for areas other than the obstacle region, and updates the situation diagram.

  The operation of the present invention will be described below. FIG. 3A shows an example in which the vehicle 10 is put in the garage while being backed. For example, when the vehicle 10 is put in the garage 1 at home, the obstacles 2 and 3 placed in the walls and the garage are known to the driver, and a warning is issued every time the obstacles 2 and 3 are detected. Is very annoying for the driver. Therefore, a warning is issued only when a new obstacle is detected or when it is likely to collide with known obstacles 2 and 3.

  FIG. 3B assumes a case where the vehicle 10 passes through a narrow corner surrounded by the wall 4. If you pass through a narrow corner, you can provide driving assistance such as steering wheel, accelerator, brake control, etc. if you can present a situation diagram that shows the width of the road, the curvature of the corner, and the distance between the vehicle 10 and the wall 4 and obstacles (electric poles etc.) It can be carried out.

  In the present invention, in consideration of such a situation, an obstacle detection device is provided that presents a situation diagram to notify the driver when there is an obstacle on a road or the like and prevents excessive warnings. Is.

  First, creation of a situation diagram will be specifically described. FIG. 4 shows an example of creating a situation diagram when stopping a vehicle in a parking lot. FIG. 4A shows a situation where the vehicle 10 is parked in the parking lot in the back. For example, it is assumed that a white line 21 is drawn in the parking lot, a car stop 22 is located behind, and a wall 23 and a pole 24 are present at the boundary with the adjacent vehicle. In such a situation, when a surrounding image is shot with the camera 11 installed in the vehicle 10 and the virtual viewpoint is set right above the vehicle, the overhead image generation unit 123 receives the image as shown in FIG. An overhead image 31 can be obtained. As the bird's-eye view image 31, a plurality of images are sequentially generated in the time axis direction t as the vehicle 10 moves.

  The overhead image generation unit 123 sets the pixel value of the obstacle area (car stop 22, wall 23, pole 24) detected by the obstacle detection unit 122 in the overhead image 31 to 0 (black). A normalized image 32 (FIG. 6) in which regions (road surface and white line 21) are normalized to 1 to 255 gradations is generated, and an average image of regions excluding obstacle regions is generated from these normalized images 32. A situation diagram 33 as shown in FIG.

  FIG. 5 is an explanatory diagram showing the generation of the overhead image 31 more specifically. FIGS. 5 (a1), (b1), and (c1) are diagrams showing the movement of the vehicle in time series, and FIGS. 5 (a2), (b2), and (c2) are generated in accordance with the movement of the vehicle. The bird's-eye view image 31 is shown.

  FIG. 6 is an explanatory diagram showing the generation of the situation diagram 33 more specifically. The overhead image 31 of FIG. 5 is processed to convert the pixel value of the obstacle region into a zero value (black), and the others A normalized image 32 is generated by normalizing the pixels in the area of 1 to 255 to a value of 1 to 255 gradations. FIGS. 6A, 6B, and 6C show the normalized images 32 generated with the movement of the vehicle in time series. FIG. 6D shows a plurality of normalized images 32 (a, b, c) on the time axis, and these normalized images 32 (a, b, c) are represented by the movement, direction and angle of the vehicle. The situation diagram 33 as shown in FIG. 6E is created by performing parallel movement and rotation in accordance with the above and combining them.

  Since the situation diagram 33 is self-generated based on the image taken by the camera 11, it can be created without inputting map information from the outside. In the situation diagram 33, the image of the obstacle is displayed in black, for example, so that it can be distinguished from an image other than the obstacle. Therefore, if the bird's-eye view image of the host vehicle is displayed in an overlapping manner, the distance between the host vehicle and the obstacle can be clearly shown.

  FIG. 7 shows an example of creating a situation diagram when the vehicle 10 passes through a narrow turn. FIG. 7A shows a situation where the vehicle 10 is about to pass a road surrounded by walls 42 and 43. For example, it is assumed that a white line 41 is drawn on the road, walls 42 and 43 are present on both sides of the road, and a utility pole 44 is present at the corner. In such a situation, when the surrounding video is shot by the camera 11 installed in the vehicle 10 and the virtual viewpoint is set right above the vehicle, the overhead image generation unit 123 has an overhead view as shown in FIG. An image 51 can be obtained. The bird's-eye view image 51 rotates and moves with the movement of the vehicle 10 and the rotation of the steering wheel, and a plurality of images are sequentially generated in the time axis direction t.

  The overhead image generation unit 123 sets the pixel values of the obstacle areas (walls 42 and 43, and the power pole 44) detected by the obstacle detection unit 122 in the overhead image 51 to 0 (black), and other areas. Generate a normalized image 52 in which (road surface and white line 41) is normalized to a value of 1 to 255 gradations, and generate an average image of regions excluding the obstacle region from these normalized images 52 (FIG. 8) Then, a situation diagram 53 as shown in FIG.

  FIG. 8 is an explanatory diagram showing the generation of the overhead image 51 more specifically, and FIGS. 8 (a1), (b1), and (c1) are diagrams showing the movement of the vehicle 10 in time series. FIGS. 8 (a2), (b2), and (c2) show a bird's-eye view image 51 that is generated in accordance with the movement, direction, and angle of the vehicle. At this time, since the virtual viewpoint is set right above the vehicle, the overhead image 51 rotates according to the direction and angle of the vehicle.

  FIG. 9 is an explanatory diagram showing the generation of the situation diagram 53 more specifically. FIGS. 9A, 9B, and 9C show the normalized image 52 generated as the vehicle moves. Shows in series. FIG. 9D shows a plurality of normalized images 52 (a, b, c) on the time axis, and these normalized images 52 (a, b, c) are represented by the movement, direction and angle of the vehicle. The situation diagram 53 as shown in FIG. 9E is created by performing parallel movement and rotation in accordance with the above and combining them.

  Since the situation diagram 53 is self-generated based on an image taken by the camera 11, it can be created without inputting map information from the outside. Since the image of the obstacle is displayed in, for example, black in the situation diagram 53, it can be distinguished from an image other than the obstacle. If the overhead image of the own vehicle is displayed in an overlapping manner, the distance between the own vehicle and the obstacle is displayed. Etc. can be clearly shown.

  FIG. 10 is a diagram for explaining driving assistance for the host vehicle 10 using the situation diagram 53. Since the boundary between the road surface and the environmental solid object can be understood by referring to the situation view 53 of the overhead view viewpoint, for example, if the situation figure 53 is created when the vehicle has passed the point before for a corner with a poor view, When passing the point, it can pass with reference to the situation figure 53. FIG.

  For example, when the host vehicle 10 passes through a narrow alley and turns around a corner, if the overhead image 54 of the host vehicle is superimposed on the situation diagram 53 shown in FIG. 10, the road width, corner curvature, host vehicle and obstacles are displayed. The distance from the objects (walls 42, 43, utility pole 44) can be recognized, and driving assistance such as control of the steering wheel, accelerator, and brake can be performed. The situation diagrams 33 and 53 are updated when the vehicle is parked at the same place or when the vehicle passes the same road.

  FIG. 11 is a flowchart showing a situation diagram creation operation, and FIG. 12 is a flowchart showing a situation diagram update operation. This flowchart mainly shows the operation of the situation diagram generation unit 126.

  In FIG. 11, step S0 is a start step for creating a situation diagram. In step S1, it is determined whether a situation diagram has already been created. If a situation diagram has already been created, the process proceeds to step S2 and proceeds to a situation diagram update process (FIG. 12). If no situation diagram has been created in step S1, all the situation diagrams are initialized as obstacle regions, that is, pixel values are set to 0 in step S3.

  After initialization, a plurality of current images that are the basis of the initialized situation diagram and the situation diagram generated by the image adjustment unit 126 are used, and a plurality of steps in steps S4 to S8 are performed in accordance with the progress of the vehicle. Repeat to create a situation diagram. That is, in step S4, it is determined whether each pixel in the situation diagram is an obstacle region, that is, whether the pixel value is 0. In step S5, the pixel determined to be an obstacle area in step S4 sets the value of the corresponding pixel in the situation diagram to the value of the corresponding pixel in the current image.

  In step S4, for pixels in the situation diagram that are determined not to be an obstacle region, it is determined in step S6 whether the corresponding pixel in the current image is an obstacle region. If it is determined that the region is an obstacle region, the process ends without updating the corresponding pixel in the situation diagram in step S7. If it is determined in step S6 that the pixel is other than the obstacle region, the weighted average of the corresponding pixel in the situation diagram and the corresponding pixel of the current image is averaged in step S8. Thus, the situation diagrams 33 and 53 shown in FIG. 6 or FIG. 9 are created based on the pixels in the obstacle region in step S5 and the pixels other than the obstacle region in step S8, and the creation of the situation diagram is finished in step S9. .

  FIG. 12 is a flowchart showing a situation diagram update operation. Step S10 in FIG. 12 is an update start step, and the situation diagram is updated in the procedure of steps S11 to S18 using the current image generated by the image adjustment unit 126 based on the newly taken image. Is called. Moreover, it returns to step S11 according to advancing of a vehicle, and a situation figure is updated sequentially.

  That is, when a new current image is generated, it is determined in step S11 whether or not the corresponding pixel in the situation diagram is an obstacle region. If it is determined that the corresponding pixel is an obstacle region, the process proceeds to step S12, and the past several times are confirmed. If all of the corresponding pixels are other than obstacles at the past several times in step S12 (YES), the process proceeds to step S13, and the average of the pixels at the past several times of passing is taken to update the updated situation diagram. Let the corresponding pixel. If it is determined in step S12 that the vehicle has been an obstacle during the past several passes (NO), the process proceeds to step S14, and the situation diagram is not updated.

  On the other hand, if it is determined in step S11 that the corresponding pixel in the situation diagram is not an obstacle area, the process proceeds to step S15, and it is determined whether or not the corresponding pixel of the normalized current pixel is an obstacle area. If it is determined in step S15 that the area is an obstacle area (YES), it is determined in step 16 whether or not all the corresponding pixels during the past several passes are obstacle areas. If all are obstacle regions (YES), the process proceeds to step S17, and the relevant pixel in the updated situation diagram is set as the obstacle pixel region. If it is determined in step 16 that an area other than an obstacle has been detected during the past several passes (NO), the process proceeds to step S14, and the situation diagram is not updated.

  If it is determined in step 15 that the corresponding pixel of the current image is a pixel other than the obstacle region (NO), the weighted average of the image normalized in step S18 is taken and the corresponding pixel in the updated situation diagram is obtained. If the situation is the same as when it passed in the past, the situation diagram is not updated. If a new obstacle occurs, it is added to the situation diagram and averaged. In addition, when an obstacle that has existed in the past has been removed, the obstacle is removed from the situation diagram by passing through a plurality of times. In step S19, the update process is terminated, and the environmental solid object is displayed on the situation diagram.

  As described above, in the embodiment of the present invention, it is possible to provide driving assistance by presenting a situation diagram when there is an obstacle on a road or the like. The situation diagrams 33 and 53 are updated when the vehicle is parked at the same place or when the vehicle passes the same road, and when a new obstacle exists, it can be recognized as an obstacle.

  Next, a second embodiment of the present invention will be described. In the second embodiment, a warning is given to the driver when a new obstacle is detected.

  FIG. 13 is a block diagram illustrating a configuration of the vehicle obstacle detection device according to the second embodiment. In FIG. 13, a second obstacle detection unit 127 is provided to detect an obstacle other than an environmental solid object, and a warning message is displayed on the display / warning unit 15 using the detection result of the second obstacle detection unit 127. It is characterized by displaying a warning sound.

  That is, the second obstacle detection unit 127 compares the bird's-eye view image (31, 51) generated by the bird's-eye view image generation unit 123 with the pre-generated situation diagram (33, 53) stored in the storage unit 14, A new obstacle area is detected.

  FIG. 14 is an explanatory diagram for explaining the operation of FIG. 13, and FIG. 14 (a) shows a situation where the vehicle 10 is about to park in the parking lot in the back. In this case, it is assumed that there are moving objects such as animals 25 and children in the parking lot. Assuming that the camera 11 installed in the vehicle 10 captures a surrounding image and sets the virtual viewpoint directly above the vehicle, the overhead image generation unit 123 generates an overhead image 31 as shown in FIG. The In the bird's-eye view image 31 in FIG. 14B, an animal image 25 is generated as a new obstacle in addition to an environmental solid object (shown by a dotted line) such as the car stop 22 and the pole 24.

  In addition, the overhead image generation unit 123 sets the pixel value of the obstacle area (the vehicle stop 22, the pole 24, and the animal 25) detected by the obstacle detection unit 122 in the overhead image 31 to 0 value (black). The region (road surface or white line 21) is normalized to 1 to 255 gradations to generate a normalized image 32 as shown in FIG. FIG. 14D shows a known situation diagram 33 stored in the storage unit 14.

  The second obstacle detection unit 127 compares the normalized image 32 (c) with the situation diagram 33 (d), and when there is an obstacle region 25 that does not exist in the situation diagram 33, the second obstacle detection unit 127 responds to the driver. Issue a warning. In other words, when passing again through the point where the situation diagram 33 has already been created, by comparing the already created situation diagram with the newly created image, an environmental solid object that always exists as a part of the environment, Newly appeared obstacles other than can be distinguished. Accordingly, a warning message such as “There is an obstacle. Please be careful” can be displayed on the display of the display / warning unit 15 or a voice can be generated from the speaker to notify the driver.

  In this way, when a new obstacle that does not exist in the situation diagram 33 is detected, the driver checks the rear based on the warning and can drive to avoid danger such as a collision. Also, if the new obstacle 25 is a moving object, the next time it is parked (or passed) at the same location, there is a high possibility that it will not exist, so it will disappear as the situation diagrams 33 and 53 are updated. .

  As a result, it is possible to prevent the driver from being excessively warned when there is a three-dimensional environmental object such as a wall or a curb of a parking lot that has already been recognized by the driver and is considered to have a low risk of traveling. In order to further increase safety, a warning may be issued when an environmental solid object is detected. However, the color of the warning sound and warning display for the driver is changed depending on whether an environmental solid object is detected or a new obstacle is detected. It is better to make the displayed characters stand out by blinking red.

  As described above, according to the embodiment of the present invention, an image captured by a camera is converted into a bird's-eye view image viewed from a virtual viewpoint, and a situation diagram of a place through which the host vehicle passes is created using a plurality of bird's-eye view images. By storing, a situation diagram can be created even in a place that does not exist in the map information.

  Therefore, when the host vehicle passes through the same place, the situation diagram can be read and displayed, and driving assistance can be performed. In addition, by updating the situation diagram, it is possible to distinguish between environmental solid objects that are always in the same place and known to the driver and obstacles that are not known to the driver. Can be prevented. Thus, it is possible to accurately inform the driver of the situation around the vehicle and assist safe driving.

  Note that several variations of the embodiment of the present invention are conceivable. For example, when there is an environmental solid object, a warning may be issued only when it is likely to collide. Further, the created situation diagram may be aligned based on the position information of the navigation device. In this case, you can display a situation diagram when you reach a point to drive carefully, so you can know the width of the road, the curvature of the corner, etc., provide information about the steering angle when turning, etc. and travel control Can provide driving assistance. Furthermore, various modifications can be made without departing from the scope of the claims.

DESCRIPTION OF SYMBOLS 10 ... Imaging part 11 ... Camera 12 ... Image processing part 121 ... Input part 122 ... Obstacle detection part 123 ... Overhead image generation part 124 ... Rotation / movement part 125 ... Image adjustment part 126 ... Situation diagram generation part 127 ... Second Obstacle detection unit 13 ... vehicle information generation unit 14 ... storage unit 15 ... display / warning unit

Claims (5)

An imaging unit capable of shooting around the vehicle;
An obstacle detection unit for detecting an obstacle included in an image around the vehicle photographed by the imaging unit;
An overhead image generation unit that converts and outputs an image around the vehicle captured by the imaging unit into an overhead image viewed from a predetermined virtual viewpoint;
A situation diagram generating unit that converts an obstacle image included in the overhead image and an image other than the obstacle into an identifiable image, and creates a situation diagram of an overhead viewpoint that represents the situation around the vehicle;
A display unit for displaying the situation diagram;
An obstacle detection device for a vehicle, comprising:   The vehicle obstacle detection device according to claim 1, wherein the situation diagram generation unit updates and creates the situation diagram of the overhead viewpoint when the vehicle is parked at the same place or passes through the same place. The overhead image generation unit generates a plurality of overhead images rotated and moved in accordance with the progress of the vehicle based on information from the vehicle information generation unit,
The vehicle obstacle detection device according to claim 1, wherein the situation diagram generation unit generates the situation diagram using the plurality of overhead images. An imaging unit capable of shooting around the vehicle;
A first obstacle detection unit for detecting an obstacle included in an image around the vehicle photographed by the imaging unit;
An overhead image generation unit that converts and outputs an image around the vehicle captured by the imaging unit into an overhead image viewed from a predetermined virtual viewpoint;
The obstacle image included in the overhead image and the image other than the obstacle are converted into an identifiable image, and a situation diagram of the overhead viewpoint representing the situation of the surroundings of the vehicle is created, and according to the progress of the vehicle A situation diagram generation unit for updating and creating the situation diagram;
A storage unit for storing the situation diagram created by the situation diagram generation unit;
A second obstacle detection unit that compares the situation diagram stored in the storage unit with the situation diagram updated and created by the situation diagram generation unit, and detects the presence of a new obstacle;
A display / warning unit that displays the situation diagram and issues a warning when at least the presence of the new obstacle is detected,
An obstacle detection device for a vehicle, comprising: The overhead image generation unit generates a plurality of overhead images rotated and moved in accordance with the progress of the vehicle based on information from the vehicle information generation unit,
The vehicle obstacle detection device according to claim 4, wherein the situation diagram generation unit generates the situation diagram using the plurality of overhead images.

Images