JP6185135B2 - Integrated image processing system - Google Patents

Description

The present invention is directed to the vehicle periphery image camera module multiple is by combining a plurality of the cameras image captured on integrating image processing system outputs this to the display unit such as a display provided on the vehicle.

  Conventionally, a vehicle peripheral image is generated by combining camera images captured by a plurality of camera modules attached to the front, rear, left, and right of an automobile, and this is displayed on a display means such as a display installed in a vehicle interior to display the entire circumference of the vehicle. There is known a camera system in which the direction can be visually recognized on one screen. As such a camera system, there is also known one that displays a vehicle peripheral image as a bird's-eye view image by subjecting each camera image to viewpoint conversion processing (for example, Patent Document 1).

  In addition, a technique is known in which an object such as a walk or a bicycle is recognized from a vehicle peripheral image displayed on a display unit by object recognition processing, and a marker is attached to the recognized object (for example, Patent Document 2).

Japanese Patent No. 4742733 JP 2012-113605 A

  In a conventional camera system, a plurality of camera images captured by each camera module are input to a control device of a vehicle-side display (for example, a navigation system), and the control device combines the camera images, performs an overhead view process, Object recognition processing and marker addition processing to the recognition object are performed. Therefore, in order to construct such a camera system, a high-level processing capability is required for the display control device on the vehicle side, and the system becomes expensive or a system with low versatility.

  Therefore, each camera module is provided with a function for performing a bird's-eye view processing of the captured camera image, a function for detecting a recognition object from the camera image by object recognition processing, and a function for attaching a marker to the recognition object of the camera image. As a means, an inexpensive camera system that can use an inexpensive device having only an image display function is conceivable.

  An object of the present invention made in view of such circumstances is to provide an inexpensive camera system that facilitates display associated with object recognition even when an image is displayed on a display that does not have a high processing capability. There is.

To solve the above problems, integrated image processing system of the present invention, form if each camera image by the camera module of the multiple is captured, integrated image processing for displaying a vehicle periphery image on the display unit in the vehicle A plurality of object recognition units that are provided corresponding to the respective camera modules and that perform object recognition processing from camera images captured by the camera modules , and are connected to the corresponding object recognition units, A plurality of communication units that perform communication via a vehicle network provided in the vehicle, and each of the communication units includes coordinate information of a recognized object recognized by the object recognition unit to which the communication unit is connected. output to the vehicle network, at least one of the plurality of communication unit, collected coordinate information of the recognition object from the other of the communication unit via the vehicle network Characterized in that it.

According to the integrated image processing system of the present invention, even when an image is displayed on a display that does not have a high level of processing capability, an integrated image processing system that facilitates display associated with object recognition is provided at low cost. be able to.

1 is a diagram illustrating a schematic configuration of a vehicle including a camera system including an integrated image processing system according to an embodiment of the present invention. It is a block diagram which shows the structural example of the camera system containing the integrated image processing system which concerns on one Embodiment of this invention. It is a block diagram which shows the structural example of the front camera module and side camera module which are shown in FIG. It is a block diagram which shows the structural example of the rear camera module shown in FIG. It is a figure which shows an example of the vehicle periphery image displayed on a display means. It is a figure which shows an example of the state in which the recognition object in the vehicle periphery image displayed on the display means was located in the boundary of the mutually adjacent camera image. It is a figure which shows an example of the relationship between a main control part and a control part. It is a figure which shows the allocation area | region of the marker provision of each camera module in the mutually adjacent camera image. It is a flowchart figure of the marker display control in the camera system containing the integrated image processing system which concerns on one Embodiment of this invention.

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

As shown in FIGS. 1 and 2, a camera system 1 including an integrated image processing system according to an embodiment of the present invention is provided in a vehicle 2 such as an automobile. This camera system 1 includes four camera modules 3 to 6 attached to the front, rear, left and right of the vehicle 2 and display means 7 installed in the vehicle interior.

  As shown in FIG. 3, the front camera module 3 attached to the front end of the vehicle 2 includes a lens unit 3a, an imaging unit 3b, a viewpoint conversion image generation unit 3c, an output unit 3d, a power supply unit 3e, a control unit 3f, and a communication unit. 3g and a storage unit 3h are provided. Similarly, the side camera module 4 attached to the right side of the vehicle 2 and the side camera module 5 attached to the left side of the vehicle 2 are respectively a lens unit 4a, 5a, an imaging unit 4b, 5b, a viewpoint conversion image generation unit 4c, 5c, output units 4d and 5d, power supply units 4e and 5e, control units 4f and 5f, communication units 4g and 5g, and storage units 4h and 5h. Since the front camera module 3 and the side camera modules 4 and 5 have basically the same configuration, in FIG. 3, each part constituting the camera modules 3 to 5 corresponds to each camera module 3 to 5. The code | symbol is attached | subjected.

  On the other hand, as shown in FIG. 4, the rear camera module 6 attached to the rear end of the vehicle 2 includes a lens unit 6a, an imaging unit 6b, a viewpoint conversion image generation unit 6c, an output unit 6d, a power supply unit 6e, a control unit 6f, A communication unit 6g and a storage unit 6h are provided, and an image composition unit 6i is provided. Here, the image composition unit 6i is a software process that operates on the control unit 6f, and should be described in the control unit 6f like the object recognition units 3j to 6j. Will be described.

  By the way, in the drawing, the image composition unit is described only in the rear camera module 6 and is not described in the front camera module 3 and the both side camera modules 4 and 5. However, the image composition unit 6i may be provided equally in any camera module 3-6. That is, only the camera module recognized as the rear camera module 6 or the camera module recognized as the main control unit by the processing described later activates the software of the image composition unit, and the other camera modules are software of the image composition unit. If the camera module is configured to be inactivated, the configuration of each camera module can be made the same, leading to a reduction in the overall cost.

  The power supply units 3e to 6e provided in the camera modules 3 to 6 are connected to a vehicle-mounted battery and supply power to the imaging units 3b to 6b and the control units 3f to 6f of the camera modules 3 to 6, respectively. To do. Moreover, the memory | storage parts 3h-6h provided in each camera module 3-6 are the imaging parts 3b-6b, the viewpoint conversion image generation parts 3c-6c, the control parts 3f-6f, and the communication parts 3g-6g, respectively as needed. Data necessary for such control can be stored.

  Each camera module 3-6 images the periphery of the vehicle using lens units 3a-6a and imaging units 3b-6b, respectively. For example, the front camera module 3 images the periphery of the front side of the vehicle 2 at a predetermined angle of view using the lens unit 3a and the imaging unit 3b. Further, the right side camera module 4 uses the lens unit 4a and the imaging unit 4b to image the periphery of the right side of the vehicle 2 at a predetermined angle of view, and the left side camera module 5 includes the lens unit 5a and the imaging unit 5b. And the left side of the vehicle 2 is imaged at a predetermined angle of view. Further, the rear camera module 6 images the periphery of the vehicle 2 in the backward direction with a predetermined angle of view using the lens unit 6a and the imaging unit 6b.

  For example, an image sensor such as a CMOS or a CCD is used as the imaging units 3b to 6b. Further, as the lens portions 3a to 6a, for example, lenses having a wide angle of view such as fisheye lenses are used.

  As shown in FIGS. 3 and 4, the imaging units 3 b to 6 b of the camera modules 3 to 6 are connected to the viewpoint conversion image generation units 3 c to 6 c, respectively, and the images captured by the imaging units 3 b to 6 b are viewpoint conversion images. It is taken into the generating units 3c to 6c. The viewpoint-converted image generation units 3c to 6c adjust the distortion of the images captured from the imaging units 3b to 6b, and generate overhead images in which the viewpoints of the images are converted.

  As shown in FIG. 2, the front camera module 3 is connected to the rear camera module 6. The bird's-eye view image generated by the front camera module 3 is output from the output unit 3d as a camera image captured by the camera module 3, and input to the image composition unit 6i provided in the rear camera module 6. Similarly, both side camera modules 4 and 5 are connected to the rear camera module 6. The bird's-eye view images generated by the two side camera modules 4 and 5 are output from the output units 4d and 5d as camera images captured by the camera modules 4 and 5, respectively, and are output to the image composition unit 6i provided in the rear camera module 6. Entered.

  The image composition unit 6i includes an overhead image (camera image) generated by the viewpoint conversion image generation unit 6c of the rear camera module 6 and three overhead images (camera images) input from the other camera modules 3 to 5. The camera modules 3 to 6 are combined in an arrangement corresponding to the mounting position of the camera modules 3 to 6 on the vehicle 2 to generate a vehicle periphery image.

  As shown in FIG. 2, the rear camera module 6 is connected to the display means 7. The output unit 6 d of the rear camera module 6 converts the vehicle surrounding image synthesized by the image synthesis unit 6 i into a format (for example, NTSC) adapted to the display unit 7 and outputs the converted image to the display unit 7.

  As shown in FIG. 5, the display means 7 displays the vehicle periphery image input from the rear camera module 6 on the screen 7a. That is, the vehicle model 8 is displayed at the center of the screen 7a of the display means 7, and a camera image (overhead image) captured by the front camera module 3 is displayed in the upper divided range 7a1 around the vehicle model 8. . Further, camera images (overhead images) captured by the side camera modules 4 and 5 are displayed in the divided range 7a2 on the right side of the screen and the divided range 7a3 on the left side of the screen of the display unit 7, respectively. Further, a camera image (overhead image) captured by the rear camera module 6 is displayed in the divided range 7a4 on the lower side of the screen of the display means 7.

  As the display means 7, for example, a display device such as a liquid crystal display or an organic EL display can be used.

  As shown in FIG. 2, the camera modules 3 to 6 are connected to a vehicle network (in-vehicle LAN) 9 provided in the vehicle 2. The vehicle network 9 may be a mobile machine bus line using a protocol such as CAN (Controller Area Network), FlexRay, or LIN (Local Interconnect Network).

  Each camera module 3-6 can communicate via the vehicle network 9 by the communication parts 3g-6g, respectively. That is, the camera modules 3 to 6 can communicate with the other camera modules 3 to 6 via the vehicle network 9 by the communication units 3g to 6g, respectively. Further, as shown in FIG. 1, a processing unit 10 of the vehicle 2 such as an engine control unit (ECU) or a shift position sensor is connected to the vehicle network 9, and each camera module 3-6 communicates with each other. The units 3g to 6g can communicate with the processing unit 10 via the vehicle network 9.

  As these communication units 3g to 6g, for example, when the vehicle network 9 is a CAN, a CAN transceiver is used. For example, a communication device conforming to the protocol of the vehicle network 9 is used.

  The control units 3f to 6f provided in the camera modules 3 to 6 abstract the mechanism blocks of software that operates on a microcomputer provided with a CPU, respectively, and as shown in FIGS. One of the functions is a function as the object recognition units 3j to 6j. The object recognition units 3j to 6j perform object recognition processing from the images captured by the imaging units 3b to 6b of their own camera modules 3 to 6. That is, the object recognition units 3j to 6j can recognize predetermined objects such as pedestrians and bicycles existing in the images by performing object recognition processing on the images captured by the imaging units 3b to 6b. As the object recognition processing by the object recognition units 3j to 6j, for example, an object recognition technique such as pattern recognition can be used.

  When the control units 3f to 6f of the camera modules 3 to 6 recognize a predetermined object such as a pedestrian or a bicycle from images captured by the imaging units 3b to 6b of their own camera modules 3 to 6, respectively. Based on the coordinate information of the recognized object recognized by the object recognition units 3j to 6j, a marker for the recognized object is displayed on the image. For example, the markers can be added to the images by the control units 3f to 6f by a method of overlaying (superimposing) the marker images on the images captured by the imaging units 3b to 6b. As described above, when there is a predetermined object in the images captured by the image capturing units 3b to 6b, the camera modules 3 to 6 output the camera images with the markers attached to the output units 3d to 6d. Output from. By the way, the coordinate information is the coordinates of the drawn object recognized as a predetermined object, and each camera module 3 is considered in consideration of the distortion component that is taken through the fisheye lens (lens portions 3a to 6a). The object recognizing units 3j to 6j calculate the coordinates in the images photographed at .about.6.

  Therefore, as shown in FIG. 5, when the predetermined object 12 exists in the vehicle peripheral image displayed on the screen 7a of the display means 7, the vehicle peripheral image is composed of the camera images of the camera modules 3 to 6. Thus, the vehicle peripheral image with the marker 13 attached to the object 12 is displayed. By attaching the marker 13 for the object 12 to the vehicle periphery image, the predetermined object 12 displayed in the vehicle periphery image can be easily recognized by the driver or the like. In the illustrated case, the marker 13 is displayed as a rectangular dotted frame, but can be set to various shapes and sizes such as a rectangular shape and a circular shape according to the shape and size of the object 12, and a solid line or the like. It can also be represented by

  As shown in FIG. 6, when a predetermined object 12 is positioned on both of a pair of adjacent camera images displayed in the divided range 7 a 1 and the divided range 7 a 2, that is, the boundary A between the pair of camera images, The marker 13 for the object 12 is displayed across both of these camera images. In this case, a part of the marker 13 is displayed on one camera image displayed in the divided range 7a1, and the remaining part is displayed on the other camera image displayed on the divided range 7a2. Therefore, when displaying the marker 13 on each camera image, it is necessary to smoothly display the marker 13 displayed on each camera image and display it correctly as one marker 13.

Therefore, in the camera system 1 including the integrated image processing system of the present invention, each of the camera modules 3 to 6 transmits the coordinate information of the predetermined object 12 recognized from the images captured by the imaging units 3b to 6b to the communication units 3g to 3g. 6g is output to the vehicle network 9, and the coordinate information can be shared among the camera modules 3 to 6 via the vehicle network 9. Then, the camera modules 3 to 6 that output one of the adjacent camera images acquire the coordinate information of the recognized object from the camera modules 3 to 6 that output the other camera image via the vehicle network 9. Using this coordinate information, a marker 13 that smoothly connects to the marker 13 displayed in the other camera image can be generated.

  More specifically, as shown in FIG. 6, a predetermined boundary A is formed between the camera image of the front camera module 3 displayed in the divided range 7a1 and the camera image of the right side camera module 4 displayed in the divided range 7a2. When the object 12 is located, the marker 13 is displayed in the following procedure.

  When the front camera module 3 recognizes the predetermined object 12 from the image captured by the imaging unit 3b by the object recognition unit 3j, the coordinate information of the recognized object 12, that is, the recognized object 12 in the image is transmitted by the communication unit 3g. It is output to the vehicle network 9. Similarly, when the right side camera module 4 recognizes the predetermined object 12 by the object recognition unit 4j from the image captured by the imaging unit 4b, the coordinate information of the recognized object 12 in the image is transmitted by the communication unit 4g. It is output to the vehicle network 9. As the coordinate information of the recognition object 12 output by the communication units 3g and 4g, for example, the size information is added to the coordinate information of the center position of the recognition object 12, and the recognition object 12 is enclosed. Information that can generate the marker 13 may be included.

  Here, each camera module 3 to 6 is assigned identification information such as an address by a processing unit 10 such as an ECU connected to the vehicle network 9 in accordance with a negotiation protocol in the standard of the vehicle network 9. . Based on this identification information, the relationship between the master and the slave is set between the camera modules 3 to 6. For example, in the present embodiment, when the predetermined object 12 is located at the boundary A or the boundary B shown in FIG. 6, the front camera module 3 is set as the master as shown in FIG. , 5 are set as slaves. On the other hand, when the predetermined object 12 is located at the boundary C or the boundary D shown in FIG. 6, the rear camera module 6 is set as a master as shown in FIG. Each is set as a slave. Therefore, when the object 12 is located at the boundary A shown in FIG. 6, the front camera module 3 is set as a master, and the right side camera module 4 is set as a slave. Since the camera image of the front camera module 3 and the camera image of the rear camera module 6 do not have a boundary, a master / slave relationship is not set between these camera modules 3 and 6.

  When the predetermined object 12 is located at the boundary A between the camera image of the front camera module 3 and the camera image of the right side camera module 4, that is, the x1 region of the camera image of the front camera module 3 schematically shown in FIG. When the object 12 is located across the y1 region of the camera image of the side camera module 4, the side camera module 4 that is a slave transmits the camera image of the front camera module 3 to the front camera module 3 via the vehicle network 9. The coordinate information of the recognition object 12 is acquired. Then, the control unit 4f of the side camera module 4 uses the coordinate information of the recognized object 12 acquired from the front camera module 3, that is, the coordinate information of the recognized object 12 in its own camera image and the camera image of the front camera module 3. The marker 13 is generated by the front camera module 3 by changing at least one of the position, shape and size of the marker 13 generated by itself based on the coordinate information of the recognition object 12 in FIG. To connect smoothly. In this way, the control unit 3f of the side camera module 4 can detect the position of the marker 13 generated by itself according to the position, shape and size of the marker 13 generated by the front camera module 3 obtained via the vehicle network 9. Adjust the shape and size.

  In the above description, the case where the object 12 is positioned at the boundary A of the vehicle peripheral image has been described. However, when the object 12 is positioned at the boundary B of the vehicle peripheral image, the front camera module 3 serves as a master and the left side camera module 5 Becomes a slave and the same control is performed. When the object 12 is located at the boundary C of the vehicle peripheral image, the rear camera module 6 is the master and the right side camera module 4 is the slave, and the object 12 is located at the boundary D of the vehicle peripheral image. , The rear camera module 6 becomes the master and the left side camera module 5 becomes the slave, and the same control is performed.

  As described above, when the predetermined object 12 is located at the boundaries A to D of the camera images adjacent to each other, the control units 3 f to 6 f of the one camera module 3 to 6 perform the other camera via the vehicle network 9. The coordinate information of the recognition object 12 recognized from the images captured by the imaging units 3b to 6b of the other camera modules 3 to 6 is acquired from the modules 3 to 6, and the marker 13 is displayed on the own camera image using this. Is attached. Therefore, even when the predetermined object 12 is located at the boundaries A to D of the adjacent camera images, the smooth marker 13 across the adjacent camera images is accurately attached to the recognized object 12. Can do.

  On the other hand, when the predetermined object 12 is located in the x region of the camera image of the front camera module 3 schematically shown in FIG. 8, the marker 13 is attached to the object 12 by the control unit 3 f of the front camera module 3. . Similarly, when the predetermined object 12 is in the y region of the camera image of the side camera module 4, the marker 13 is attached to the object 12 by the control unit 4 f of the side camera module 4.

  Next, an example procedure of marker display control (camera control method) in the camera system 1 will be described with reference to FIG.

  First, the control units 3f to 6f of the camera modules 3 to 6 communicate with the processing unit 10 of the vehicle 2 via the vehicle network 9 to determine whether or not the gear of the vehicle 2 is reverse (step S1). ). When it is determined in step S1 that the gear of the vehicle 2 is reverse, the control units 3f to 6f acquire images captured from the imaging units 3b to 6b of their own camera modules 3 to 6 (step S2). Further, vehicle information such as vehicle speed and steering angle is acquired via the vehicle network 9 (step S3). In this case, the marker 13 is attached to the object 12 recognized in the vehicle peripheral image only when the vehicle 2 moves backward. However, the vehicle 2 is recognized in the vehicle peripheral image even when the vehicle 2 moves forward or stops. A marker 13 may be attached to the object 12.

  Next, the control units 3f to 6f determine whether or not the predetermined object 12 has been recognized by the object recognition process from the acquired image (step S4, recognition step). If it is determined in step S4 that the recognition object 12 is present in the image, each camera module 3-6 outputs the coordinate information of the recognition object 12 to the vehicle network 9 (step S5, output process). Can be shared by the camera modules 3 to 6.

  Next, the control units 3f to 6f determine whether or not the recognition object 12 is located at the boundaries A to D between a pair of adjacent camera images (step S6). In step S6, if it is determined that the recognition object 12 is positioned at the boundaries A to D of the adjacent camera images, the camera units 3f to 6f of the pair of camera modules 3 to 6 that capture these camera images capture each camera image. Is marked with a marker 13. At this time, the control units 3 f to 6 f of any one of the camera modules 3 to 6 acquire the coordinate information of the recognition object 12 in the camera images of the other camera modules 3 to 6 through the vehicle network 9. And the said control parts 3f-6f attach | subject the marker 13 smoothly connected with the marker 13 displayed on an adjacent camera image to an own camera image using this coordinate information (step S7, addition process).

  As described above, when the marker 13 is attached to the recognition object 12 in the image, the viewpoint is then converted by the viewpoint conversion image generation units 3c to 6c in each of the camera modules 3 to 6, and an overhead view image, that is, a camera image. (Step S8). Then, the bird's-eye view images of the camera modules 3 to 6 generated in step S8 are combined by the image combining unit 6i of the rear camera module 6 to be a vehicle peripheral image (step S9), and from the output unit 6d of the rear camera module 6 The image is output toward the display means 7 (step S10).

  On the other hand, if it is determined in step S4 that there is no recognized object 12, the process of steps S8 to S10 is performed without adding the marker 13 to the image.

  If it is determined in step S6 that the recognized object 12 is not located at the boundary between adjacent camera images, the camera modules 3 to 6 that capture the camera image where the object 12 is positioned recognizes the recognized image in the camera image. The marker 13 is attached to the object 12 (step S11), and the processing from step S8 to step S10 is performed.

  As described above, in the camera system 1, the object recognition units 3j to 6j that detect the recognition object 12 by the object recognition process from the images captured by the camera modules 3 to 6, and the marker for the recognition object 12 in the captured image. 13 is provided with control units 3f to 6f. As a result, it is possible to realize the camera system 1 that enables display associated with the object recognition process even when an inexpensive device that has only an image display function, for example, is used as the display unit 7. The cost of the entire system can be reduced. Further, by providing viewpoint conversion image generation units 3c to 6c for converting the camera image into a bird's-eye view image in each camera module 3 to 6, an inexpensive one having only an image display function is used as the display means 7. As a result, the cost of the camera system 1 can be reduced.

  Furthermore, in the camera system 1, the communication units 3g to 6g for outputting the coordinate information of the recognition object 12 recognized from the images captured by the imaging units 3b to 6b to the vehicle network 9 are provided to the camera modules 3 to 6, respectively. Since the recognition object 12 is located at the boundaries A to D of the adjacent camera images, the two camera modules 3 to 6 that attach the markers 13 to these camera images are connected via the vehicle network 9. The coordinate information of the recognized object 12 recognized in the images of the other camera modules 3 to 6 can be used, and the marker 13 is accurately placed on the recognized object 12 to the control units 3f to 6f of the camera modules 3 to 6. It can be attached.

  Although the present invention has been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these variations and modifications are included in the scope of the present invention.

  In the embodiment, the camera modules 3 to 6 as slaves acquire the coordinate information of the recognition object 12 from the master camera modules 3 to 6 via the vehicle network 9 and use the acquired coordinate information in the image. Although the marker 13 is attached, it is not limited to this. That is, at least one camera module 3 to 6 acquires the coordinate information of the recognition object 12 from the other camera modules 3 to 6 via the vehicle network 9, and uses the acquired coordinate information to place the marker 13 in the image. Any structure may be used. For example, both camera modules 3 to 6 that output adjacent camera images acquire the coordinate information of the recognition object 12 from the other camera modules 3 to 6 via the vehicle network 9 and use the acquired coordinate information. Thus, the marker 13 may be added to the image.

  Further, the control units 3f to 6f of any one of the camera modules 3 to 6 are set as the main control unit, and the control units 3f to 6f, which are set as the main control unit, all other camera modules via the vehicle network 9. It is also possible to obtain the coordinate information of the recognition object 12 from 3 to 6 and attach the marker 13 to all the camera images, that is, the vehicle peripheral images using the obtained coordinate information. In this case, identification information such as an address is assigned to each of the camera modules 3 to 6 in accordance with the negotiation protocol in the standard of the vehicle network 9 by the processing unit 10 such as an ECU connected to the vehicle network 9. The control units 3f to 6f of any one of the camera modules 3 to 6 can be configured to be set as the main control unit according to the identification information. The control units 3f to 6f serving as the main control unit acquire coordinate information of the recognition object 12 in other camera images from the control units 3f to 6f of the other camera modules 3 to 6 via the vehicle network 9. Then, marker addition processing is performed on all camera images (vehicle peripheral images) based on the coordinate information of the recognized object 12 in its own camera image and the coordinate information of the recognized object 12 in other camera images. As described above, the control units 3f to 6f, which are the main control units, can control the other control units 3f to 6f through the vehicle network 9 to attach the marker 13 to the vehicle peripheral image.

  In this case, only the control unit 6f set as the main control unit, that is, the control unit 6f of the rear camera module 6 activates software having a function of attaching the marker 13 in the image, and controls the other camera modules 3 to 5. The units 3f to 5f are configured such that software having a function of attaching the marker 13 in the image is inactivated. Further, when the control unit of another camera module is set as the main control unit, only the control unit activates software having a function of attaching the marker 13 in the image, and the function of the control unit of the other camera module is activated. The software is deactivated. Therefore, even if it is a case where it is a case where it is set as the structure which carries out the integrated control of the other control part via the vehicle network 9 by the control part set to the main control part, it shall use the thing of the same structure as each camera module 3-6. As a result, the overall cost of the camera system 1 can be reduced.

  Furthermore, in the said embodiment, although the four camera modules 3-6 are provided in the vehicle 2, if not only this but the periphery of the vehicle 2 can be imaged, the number of the camera modules 3-6 will be. It can be set arbitrarily.

  Furthermore, in the said embodiment, although the marker 13 is attached | subjected to the image imaged by the imaging parts 3b-6b, and the image to which the marker 13 was attached is output to the display means 7, it is not restricted to this, A camera A marker image generated separately from the image may be input to the display unit 7 via the vehicle network 9, and the marker image may be superimposed on the camera image on the display unit 7.

  Furthermore, in the said embodiment, the viewpoint conversion image generation parts 3c-6c are provided in each camera module 3-6, and the image imaged by the imaging parts 3b-6b is processed into a bird's-eye view image by the viewpoint conversion image generation parts 3c-6c. However, the present invention is not limited to this, and the images captured by the imaging units 3b to 6b may be output to the display unit 7 as they are. Alternatively, the viewpoint conversion image generation units 3c to 6c may be provided only in the rear camera module 6, and the images synthesized by the image synthesis unit 6i may be collectively processed into an overhead image. Alternatively, the viewpoint conversion image generation units 3c to 6c may be provided in the display unit 7 so that each camera image or vehicle peripheral image is processed into a bird's-eye view image in the display unit 7.

  Further, in the above embodiment, the front camera module 3 and the both side camera modules 4 and 5 are connected to the rear camera module 6, and each camera module 3 to 6 is included in the image composition unit 6 i provided in the rear camera module 6. Although the captured camera images are synthesized and output to the display unit 7, the present invention is not limited thereto, and all the camera modules 3 to 6 may be directly connected to the display unit 7. In this case, the rear camera module 6 has the same configuration as the other camera modules 3 to 5 shown in FIG. 3, and the versatility of the camera modules 3 to 6 can be enhanced.

  Furthermore, in the said embodiment, each camera module 3-6 outputs the coordinate information of the recognition object 12 recognized from the image which self imaging part 3b-6b image | photographed to the vehicle network 9 by the communication parts 3g-6g. However, in addition to the coordinate information of the recognition object 12, at least one of the movement vector information of the recognition object 12 and the coordinate information of the marker 13 generated by the control units 3f to 6f of the camera modules 3 to 6 is further transmitted to the vehicle. You may make it output to the network 9. FIG. In this case, in addition to the coordinate information of the recognition object 12 from the master camera modules 3 to 6 acquired via the vehicle network 9, the slave camera modules 3 to 6 move vector information of the recognition object 12 and each camera module. The own marker 13 can be generated using the coordinate information of the marker 13 generated by the 3 to 6 control units 3f to 6f. Thereby, the marker 13 can be more accurately displayed by the control units 3f to 6f. As the coordinate information of the marker 13, for example, when the marker 13 is rectangular, the coordinate information of the reference position of the marker 13 in the camera image and information such as the length and direction of each side, and when the marker 13 is circular The coordinate information that can generate the marker 13 can be included, such as coordinate information of the center position and information of the radial dimension thereof.

  Furthermore, in the said embodiment, although the case where a pair of adjacent camera images in a vehicle periphery image touches boundary A-D is shown, it displays so that the object 12 may be located in both adjacent camera images. If possible, there may be a gap between adjacent camera images.

DESCRIPTION OF SYMBOLS 1 Camera system 2 Vehicle 3 Front camera module 4,5 Side camera module 6 Rear camera module 3a-6a Lens part 3b-6b Imaging part 3c-6c Viewpoint conversion image generation part 3d-6d Output part 3e-6e Power supply part 3f-6f Control unit 3g to 6g Communication unit 3h to 6h Storage unit 6i Image composition unit 3j to 6j Object recognition unit 7 Display unit 7a Screen 7a1 to 7a4 Division range 8 Vehicle model 9 Vehicle network 10 Processing unit 11 Object recognition unit 12 Object 13 Marker A ~ D boundary x, x1 area y, y1 area

Claims (1)

Form if each camera image by the camera module of the multiple is captured, an integrated image processing system for displaying a vehicle periphery image on the display unit in the vehicle,
A plurality of object recognition units that are provided corresponding to the respective camera modules and that perform object recognition processing from camera images captured by the camera modules ;
A plurality of communication units that are connected to the corresponding object recognition units and perform communication via a vehicle network provided in the vehicle,
Each of the communication units outputs the coordinate information of the recognized object recognized by the object recognition unit to which the communication unit is connected to the vehicle network ,
At least one of the plurality of communication units acquires coordinate information of the recognized object from another communication unit via the vehicle network .

Images