JP2019197270A - Object detection device and object detection method - Google Patents

Abstract

To provide an object detection device and an object detection method capable of detecting objects including an object moving in a range to be a dead angle of a camera.SOLUTION: An object detection device includes a camera device and a radar device. The camera device includes an imaging unit and a detection unit. The imaging unit images a predetermined range in the periphery of a vehicle. The detection unit detects a moving object existing in the predetermined range on the basis of an image imaged by the imaging unit. The radar device includes a determination unit. When the radar device detects a moving object in a period when the camera device does not detect a moving object, the determination unit determines that there is a possibility of existence of a moving object in a dead angle range to be a dead angle of the camera device within the predetermined range.SELECTED DRAWING: Figure 1

Description

  Embodiments disclosed herein relate to an object detection apparatus and an object detection method.

  2. Description of the Related Art Conventionally, an object detection device called a fusion sensor that improves object detection accuracy by detecting an object based on captured image information around a vehicle imaged by an in-vehicle camera and object detection information by a millimeter wave radar (For example, refer to Patent Document 1).

JP 2017-47706 A

  However, in the conventional object detection device, an object that moves within a blind spot of the camera is not a detection target. One aspect of the embodiments has been made in view of the above, and provides an object detection device and an object detection method capable of detecting an object including an object that moves in a blind spot of a camera. Objective.

  An object detection device according to an aspect of an embodiment includes a camera device and a radar device. The camera device includes an imaging unit and a detection unit. The imaging unit images a predetermined range around the vehicle. The detection unit detects a moving object existing in the predetermined range based on an image captured by the imaging unit. The radar apparatus includes a determination unit. When the radar apparatus detects a moving object during a period in which the moving apparatus is not detected by the camera device, there is a possibility that the moving object exists in a blind spot range that is a blind spot of the camera device in the predetermined range. Judge that there is.

  The object detection device and the object detection method according to an aspect of the embodiment can detect objects including an object that moves in a range that is a blind spot of the camera.

FIG. 1 is an explanatory diagram of an object detection method according to the embodiment. FIG. 2 is an explanatory diagram illustrating an example of the configuration of the object detection device according to the embodiment. FIG. 3A is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 3B is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 4A is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 4B is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 5A is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 5B is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 6A is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 6B is an explanatory diagram illustrating a specific example of the operation of the alerting unit according to the embodiment. FIG. 7 is a flowchart illustrating an example of processing executed by the image processing ECU according to the embodiment. FIG. 8 is a flowchart illustrating an example of processing executed by the radar ECU according to the embodiment. FIG. 9 is a flowchart illustrating an example of processing executed by the alerting unit according to the embodiment. FIG. 10A is an explanatory diagram of a switching mode of a stationary object only flag according to the first modification of the embodiment. Drawing 10B is an explanatory view of the change mode of the stationary object only flag concerning modification 2 of an embodiment.

  Hereinafter, embodiments of an object detection device and an object detection method will be described in detail with reference to the accompanying drawings. In addition, this invention is not limited by embodiment shown below. FIG. 1 illustrates an outline of an object detection method according to an embodiment.

  FIG. 1 is an explanatory diagram of an object detection method according to the embodiment. As shown in FIG. 1, the object detection device 1 according to the embodiment is a device that is mounted on a vehicle 10 and detects an object that exists in front of the vehicle 10, for example.

  In addition, the object detection apparatus 1 can also detect an object existing behind or on the side of the vehicle 10 by being provided behind or on the side of the vehicle 10. Moreover, the mobile body in which the object detection apparatus 1 is mounted is not limited to the vehicle 10, and may be any mobile body such as a ship, an aircraft, a train, and the like.

  The object detection device 1 includes a camera device (hereinafter referred to as “camera 2”) and a radar device (hereinafter referred to as “radar 3”), the detection result of the object by the camera 2, and the object by the radar 3. This is a fusion sensor that detects an object based on the detection result.

  The camera 2 sequentially captures images of a predetermined range in front of the vehicle (hereinafter referred to as “detection range 2A”) in a predetermined cycle, and the image captured this time (current frame image) and the previous image (previous frame image). ) To detect a moving object.

  For example, the radar 3 receives a radio wave reflected by an object by emitting a radio wave in the millimeter wave band toward a detection range 3A that is substantially the same as the detection range 2A of the camera 2, and the object is detected based on the received radio wave. Detect.

  The object detection device 1 detects the object by comprehensively determining the presence of the object from the detection results of the two sensors, the camera 2 and the radar 3. Thereby, the object detection apparatus 1 can detect an object more reliably than the other object detection apparatus which detects an object based on the detection result by one sensor.

  However, in the object detection device 1, a range that becomes a blind spot of the camera 2 may occur depending on the surrounding environment of the vehicle 10. For example, as shown in FIG. 1, when the buildings 100 and 101 on both sides of the vehicle 10 are stopped in front of the T-junction R <b> 1, the point where the T-junction R <b> 1 turns right and left is the blind spot of the camera 2. .

  In this case, the camera 2 is within the detection range 2A if there is no building 101, but walks toward the T-junction R1 before turning right at the T-junction R1 that becomes a blind spot because of the building 101. Even if there is a person T1, the pedestrian T1 cannot be detected.

  On the other hand, the radar 3 reflects the radiated radio wave B to the front wall 102 of the vehicle 10 and hits the pedestrian T1, and is reflected by the pedestrian T1 and receives the radio wave reflected by the wall 102 again. . As a result, the radar 3 cannot identify the position of the pedestrian T1, but can detect that there is a possibility that a moving object exists in the range that becomes the blind spot of the camera 2.

  Therefore, when the vehicle 10 is stopped and the pedestrian that is a moving object is not detected by the camera 2, the object detection device 1 detects an object whose relative speed with respect to the radar 3 is not zero, It is determined that there is a possibility that a moving object exists in a certain range. Thereby, the object detection apparatus 1 can detect an object including an object that moves in a range that is a blind spot of the camera 2.

  Next, an example of the configuration of the object detection device 1 according to the embodiment will be described with reference to FIG. FIG. 2 is an explanatory diagram illustrating an example of the configuration of the object detection device 1 according to the embodiment. As shown in FIG. 2, the object detection device 1 is connected to, for example, a car navigation device 5 and a vehicle control device 6.

  The car navigation device 5 includes a navigation unit 51, a storage unit 52, and a display unit 53. The storage unit 52 stores map information 54. When the destination of the vehicle 10 is set by the user of the vehicle 10, the navigation unit 51 superimposes the travel route of the vehicle 10 from the current location to the destination on the map information and causes the display unit 53 to display the route guidance to the user. I do.

  The vehicle control device 6 is a control device that performs overall control of the entire vehicle. For example, when information indicating that there is an object that obstructs the traveling of the vehicle 10 from the object detection device 1 is input, the vehicle control device 6 automatically activates a brake or avoids the object. To automatically steer and so on.

  The object detection device 1 includes a camera 2, a radar 3, and an alerting unit 4. The camera 2 includes an imaging unit 21 and an image processing ECU (Electronic Control Unit) 22. The imaging unit 21 includes, for example, a solid-state imaging device such as a CMOS (Complementary Metal Oxide Semiconductor), captures an image of the detection range 2A in front of the vehicle 10 at a predetermined period, and sequentially outputs the captured images to the image processing ECU 22. .

  The image processing ECU 22 includes a microcomputer having a central processing unit (CPU), a read only memory (ROM), a random access memory (RAM), and various circuits.

  The image processing ECU 22 includes a detection unit 23 that functions when the CPU executes a program stored in the ROM using the RAM as a work range. The detection unit 23 may be configured by hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  The detection unit 23 detects a moving object existing in the detection range 2 </ b> A based on the inter-frame difference between images sequentially input from the imaging unit 21. For example, when there is only a stationary object in the detection range 2A of the camera 2 and no moving object during the period when the vehicle 10 is stopped, if the camera 2 captures the image twice, The image is the same as the previously captured image.

  For this reason, if the detection unit 23 takes a difference (interframe difference) between the luminance value of each pixel of the image captured this time and the luminance value of each pixel of the image captured last time, the interframe difference is theoretically calculated. Therefore, it is determined that only a stationary object exists in the detection range 2A.

  However, in practice, the difference between frames does not become zero completely due to disturbances such as wind. For this reason, the detection unit 23 determines that only a stationary object exists in the detection range 2A when the interframe difference is equal to or less than the predetermined threshold, and detects the detection range 2A when the interframe difference exceeds the predetermined threshold. It is determined that there is a moving object inside.

  When an image is input from the imaging unit 21, the detection unit 23 first turns off a stationary object only flag indicating that only a stationary object exists in the detection range 2A. And the detection part 23 turns OFF a stationary object only flag, when the difference between frames of the image input this time and the image input last time is below a predetermined threshold value.

  Then, the image processing ECU 22 transmits an ON or OFF stationary object only flag to the radar 3 for each image. Accordingly, the radar 3 refers to the state of the stationary object only flag input from the detection unit 23 to determine whether the moving object is detected by the camera 2 or the period in which the moving object is not detected. can do.

  In addition, the detection unit 23 outputs the image input from the imaging unit 21 to the alerting unit 4 and the car navigation device 5, and causes the display unit 53 to display the route guidance by the car navigation device 5. . In addition, the detection part 23 can also output the image input from the imaging part 21 to a display apparatus, and can always display it, when the display apparatus is separately mounted in the vehicle 10. FIG.

  The radar 3 includes an antenna unit 31 and a radar ECU 32. The antenna unit 31 radiates (transmits) millimeter-wave radio waves whose frequency continuously increases or decreases toward the detection range 3A of the radar 3, and receives the radio waves reflected by the object. Then, the antenna unit 31 mixes the transmitted radio wave and the received radio wave to generate a beat signal, and outputs the generated beat signal to the radar ECU 32.

  The radar ECU 32 includes a microcomputer having a CPU, ROM, RAM, and various circuits. The radar ECU 32 includes a signal processing unit 33 and a recognition processing unit 34 that function when the CPU executes a program stored in the ROM using the RAM as a work range. The signal processing unit 33 and the recognition processing unit 34 may be configured by hardware such as an ASIC or FPGA.

  The signal processing unit 33 performs a two-dimensional FFT (Fast Fourier transform) on the beat signal input from the antenna unit 31 and extracts a peak from the FFT result. Subsequently, the signal processing unit 33 calculates the angle (azimuth) of the object with respect to the radar 3 to be detected, the distance to the object, and the relative speed with the object based on the extracted peak and the phase change of the peak.

  Then, the signal processing unit 33 outputs the calculated angle of the object, the distance to the object, and the relative speed with the object to the recognition processing unit 34. The recognition processing unit 34 detects a moving object and a stationary object existing in the detection range A3 of the radar 3 based on the angle of the object input from the signal processing unit 33, the distance to the object, and the relative speed with the object. .

  And the recognition process part 34 outputs the information which shows that to the vehicle control apparatus 6, when the object which prevents driving | running | working of the vehicle 10 ahead is detected. The vehicle control device 6 automatically activates a brake or avoids an object when information indicating that there is an object that obstructs the traveling of the vehicle 10 ahead is input from the recognition processing unit 34. Control such as steering.

  The recognition processing unit 34 includes a determination unit 35 that determines whether or not there is a possibility that a moving object exists in the blind spot range that is the blind spot of the camera 2. When the determination unit 35 detects an object whose relative speed with respect to the radar 3 is not zero during a period in which the vehicle 10 is stopped and no moving object is detected by the camera 2, a blind spot in the detection range 2 </ b> A of the camera 2 is detected. It is determined that there is a possibility that a moving object exists in the blind spot range.

  The determination unit 35 receives a stationary object only flag from the detection unit 23 of the camera 2. As described above, the detection unit 23 of the camera 2 turns from OFF to ON when only a stationary object exists in the detection range 2A of the camera 2.

  For this reason, the determination unit 35 can determine that the period in which the stationary object only flag received from the detection unit 23 of the camera 2 is ON is a period in which no moving object is detected by the camera 2. As a result, the determination unit 35 detects the blind spot of the camera 2 when a non-zero relative speed is input from the signal processing unit 33 while the vehicle 10 is stopped and the stationary object only flag received from the detection unit 23 is ON. It can be determined that there is a possibility that a moving object exists in the blind spot range.

  If the determination unit 35 determines that there is a possibility that there is a moving object in the blind spot range that is the blind spot of the camera 2, the determination unit 35 outputs information indicating that fact to the alerting unit 4. The alerting unit 4 displays alerting information when information indicating that there is a possibility that a moving object is present in the blind spot range that is the blind spot of the camera 2 is input from the determination unit 35.

  For example, the alerting unit 4 outputs alerting information to the display unit 53 of the car navigation device 5 and causes the alerting information to be superimposed on an image captured by the camera 2 displayed on the display unit 53. Thereby, the object detection apparatus 1 can urge the driver of the vehicle 10 to pay attention to a moving object that exists at a position that cannot be visually confirmed.

  At this time, for example, the alerting unit 4 estimates the blind spot range of the camera 2 based on the road shape determined from the map information stored in the car navigation device 5, and the blind spot in the image captured by the camera 2. The alert information is displayed at the position corresponding to the range.

  Thereby, the object detection apparatus 1 can make a driver | operator recognize intuitively the position of the blind spot range of the camera 2 in the image displayed on the display part 53. FIG. Therefore, the object detection device 1 can cause the driver to quickly find a moving object that appears from the blind spot range.

  In addition, the alerting unit 4 estimates, for example, a blind spot range caused by another vehicle in the image displayed on the display unit 53, and alerts a position corresponding to the blind spot range in the image captured by the camera 2. Display information.

  Thereby, the object detection apparatus 1 can make a driver | operator reconfirm the position of the blind spot range which becomes a blind spot by the other vehicle which a driver is distracted by other vehicles and tends to neglect attention, for example. . Therefore, the object detection device 1 can prompt the driver to find a moving object that appears from the blind spot range.

  Next, with reference to FIG. 3A-FIG. 6B, the specific example of operation | movement of the above-mentioned alerting part 4 is demonstrated. 3A to 6B are explanatory diagrams illustrating specific examples of the operation of the alerting unit 4 according to the embodiment.

  As shown in FIG. 3A, for example, when the vehicle 10 starts traveling from before the intersection R2 of the crossroads, the alerting unit 4 is based on the map information 54 on the road where the vehicle 10 intersects the stopped road. The hatched area is estimated to be the blind spot area.

  At this time, when it is determined by the determination unit 35 that there is a possibility that there is a moving object in the blind spot range, for example, a bicycle T2 trying to enter the intersection R2 or a pedestrian T3 heading to the intersection R2 is in the blind spot range. May exist.

  Therefore, as shown in FIG. 3B, the alerting unit 4 superimposes alerting information C on the left and right ends of the displayed image P1. Thereby, even if the bicycle T2 or the pedestrian T3 enters the intersection R2 from the blind spot range, the object detection device 1 can cause the driver to quickly find the bicycle T2 or the pedestrian T3.

  As shown in FIG. 4A, the alerting unit 4 has a hatched range on the side road based on the map information 54, for example, when the vehicle 10 starts traveling from the side road of the T-shaped road R3. Estimated to be in the blind spot range.

  At this time, when it is determined by the determination unit 35 that there is a possibility that a moving object exists in the blind spot range, for example, there is a possibility that the small animal T4 trying to cross the road on which the vehicle 10 travels exists in the blind spot range. There is.

  Therefore, as shown in FIG. 4B, the alerting unit 4 superimposes alerting information C on the left end of the image P2 being displayed, for example. Thereby, even if the small animal T4 jumps out of the side road, the object detection apparatus 1 can make the driver find the small animal T4 promptly.

  5A, for example, when the vehicle 10 starts to travel from the front of the right curve R4 where the vehicle 10 is poorly visible and passes through the mountainous area, the alerting unit 4 determines the right curve R4 based on the map information 54. Estimated that the hatched area near the exit is the blind spot area.

  At this time, if the determination unit 35 determines that there is a possibility that a moving object exists in the blind spot range, for example, there may be a wild animal T5 moving near the exit of the right curve R4 in the blind spot range. is there.

  Therefore, as illustrated in FIG. 5B, the alerting unit 4 superimposes alerting information C near the rightmost slope in the displayed image P3, for example. Thereby, even if the wild animal T5 appears on the road if the vehicle 10 approaches the exit of the right curve R4, the object detection apparatus 1 can make the driver find the wild animal T5 promptly.

  Further, as shown in FIG. 6A, the alerting unit 4 may, for example, form a pattern from an image captured by the camera 2 when the vehicle 10 starts a right turn from an intersection R5 where another vehicle T6 is stopped in the oncoming lane. The other vehicle T6 is detected by recognition.

  In such a case, the alerting unit 4 estimates that the hatched range behind the other vehicle T6 is the blind spot range. At this time, when it is determined by the determination unit 35 that there is a possibility that there is a moving object in the blind spot range, for example, the mini bike T7 may be waiting for a signal in the blind spot range.

  Therefore, as shown in FIG. 6B, the alerting unit 4 superimposes alerting information C on the rear position of the other vehicle T6 in the displayed image P4. As a result, the object detection device 1 can promptly find the minibike T7 even if the minibike T7 appears from the rear of the other vehicle T6 after passing the other vehicle T6.

  Next, an example of processing executed by the image processing ECU 22 included in the camera 2 will be described with reference to FIG. FIG. 7 is a flowchart illustrating an example of processing executed by the image processing ECU 22 according to the embodiment.

  The image processing ECU 22 executes the process shown in FIG. 7 every time an image captured in a predetermined cycle is input from the imaging unit 21. As shown in FIG. 7, when an image is input from the imaging unit 21, the image processing ECU 22 first turns off the stationary object only flag for the image (step S101).

  Subsequently, the image processing ECU 22 determines whether or not the host vehicle speed is 0 (step S102). If the image processing ECU 22 determines that the host vehicle speed is not zero (step S102, No), the process proceeds to step S107.

  Further, when it is determined that the vehicle speed is 0 (step S102, Yes), the image processing ECU 22 determines whether or not the moving object can be detected by the image (step S103). At this time, the image processing ECU 22 determines that the moving object cannot be detected by the image when it is difficult to capture the subject at night or in stormy weather.

  If the image processing ECU 22 determines that the moving object cannot be detected by the image (No at Step S103), the process proceeds to Step S107. Further, when it is determined that the moving object can be detected by the image (step S103, Yes), the image processing ECU 22 calculates an inter-frame difference between the image input from the imaging unit 21 this time and the previous frame. (Step S104).

  Subsequently, the image processing ECU 22 determines whether or not the calculated interframe difference is equal to or less than the threshold (step S105). If it is determined that the interframe difference is not equal to or less than the threshold (step S105, No), the processing is performed. Move to S107.

  If the image processing ECU 22 determines that the interframe difference is equal to or less than the threshold (Yes in step S105), the image processing ECU 22 turns on the stationary object only flag that was turned off in step S101 (step S106).

  Subsequently, the image processing ECU 22 transmits a stationary object only flag to the radar 3 (step S107), outputs an image to the display unit 53 and the alerting unit 4 (step S108), and ends the process. Then, when the next image is input from the imaging unit 21, the image processing ECU 22 starts the process again from step S101.

  Note that the image processing ECU 22 completes the processes of steps S101 to S107 described above until the processes of steps S201 to S204 executed by the radar ECU 32 described below with reference to FIG. 8 are completed.

  Next, an example of processing executed by the radar ECU 32 included in the radar 3 will be described with reference to FIG. FIG. 8 is a flowchart illustrating an example of processing executed by the radar ECU 32 according to the embodiment.

  Here, a process that is executed when the radar ECU 32 detects an object that moves in a range of the blind spot of the camera 2 will be described. Note that the process in which the radar ECU 32 detects an object that exists outside the range of the blind spot of the camera 2 is the same as that of a general radar, and thus the description thereof is omitted here.

  The radar ECU 32 executes the process shown in FIG. 8 every time a beat signal is input from the antenna unit 31 at a predetermined cycle. Further, the radar ECU 32 executes the processing of step S201 to step S210 shown in FIG. 8 in parallel during the period in which the image processing ECU 22 performs the processing of step S101 to step S208 shown in FIG.

  As shown in FIG. 8, when a beat signal is input from the antenna unit 31, the radar ECU 32 first performs FFT on the beat signal (step S201). Here, the radar ECU 32 performs a two-dimensional FFT.

  Subsequently, the radar ECU 32 performs peak extraction from the result of the two-dimensional FFT (step S202), and calculates an angle at which the object exists based on the phase change of the extracted peak (step S203).

  Thereafter, the radar ECU 32 calculates an object candidate speed based on the extracted peak (step S204). At this time, the radar ECU 32 calculates a plurality of (for example, five) candidate velocities in consideration of phase folding.

  Thereafter, the radar ECU 32 determines whether or not the own vehicle speed is zero (step S205). When the radar ECU 32 determines that the host vehicle speed is not 0 (No in step S205), the radar ECU 32 ends the process.

  If the radar ECU 32 determines that the host vehicle speed is 0 (step S205, Yes), the radar ECU 32 determines whether the stationary object only flag received from the camera 2 during the current process is ON (step S206). . When the radar ECU 32 determines that the stationary object only flag is not ON (step S206, No), the process is terminated.

  When the radar ECU 32 determines that the stationary object only flag is ON (step S206, Yes), the radar ECU 32 determines whether the candidate speed calculated in step S204 includes 0 [m / s] (step S207). ).

  When the radar ECU 32 determines that 0 [m / s] is included (step S207, Yes), the radar ECU 32 sets the relative speed to 0 for all peaks existing in the detection range 2A of the camera 2 (step S208). The process is terminated.

  Further, when it is determined that the candidate speed does not include 0 [m / s] (No at Step S207), the radar ECU 32 adds an out-of-range moving object flag to the information on the moving object (Step S209). Subsequently, the radar ECU 32 outputs information on the moving object to the vehicle control device 6 and the alerting unit 4 (step S210), and ends the process.

  Next, an example of processing executed by the alerting unit 4 according to the embodiment will be described with reference to FIG. FIG. 9 is a flowchart illustrating an example of processing executed by the alerting unit 4 according to the embodiment.

  When the information on the moving object is input from the radar ECU 32, the alerting unit 4 executes the process shown in FIG. Specifically, as illustrated in FIG. 9, when information on a moving object is input from the radar ECU 32, the alerting unit 4 determines whether or not an out-of-range moving object flag is added to the information on the moving object. (Step S301).

  And when the alerting part 4 determines with the out-of-range moving object flag not being provided (step S301, No), it complete | finishes a process. Further, when it is determined that the out-of-range moving object flag is assigned (Yes in step S301), the alerting unit 4 determines whether there is another vehicle in the image, for example, by pattern recognition ( Step S302).

  When the alerting unit 4 determines that there is another vehicle (step S302, Yes), the blind spot range of the camera 2 caused by the other vehicle is estimated (step S303), and the process proceeds to step S305.

  Further, when it is determined that there is no other vehicle in the image (No at Step S302), the alerting unit 4 estimates the blind spot range of the camera 2 based on the road shape determined from the map information 54 (Step S304). Then, the process proceeds to step S305. In step S305, the alerting unit 4 displays the alerting information C at a position corresponding to the estimated blind spot range of the camera 2 in the displayed image, and ends the process.

  In the above-described embodiment, the case where the determination unit 35 determines whether there is a possibility that a moving object exists in the blind spot range of the camera 2 when the vehicle 10 is stopped has been described. This is an example.

  Even if the traveling speed of the vehicle 10 is not completely zero even if the traveling speed is such that the detection of an object is not substantially hindered, the determination unit 35 can detect the blind spot range of the camera 2. It is possible to determine whether or not there is a possibility of moving objects.

  For example, in the determination unit 35, the radar 3 detects an object that moves at a sufficiently high relative speed with respect to the vehicle 10 than the traveling speed of the vehicle 10 estimated from the inter-frame difference between images captured by the camera 2. In this case, it can be determined that there is a possibility that a moving object exists in the blind spot range of the camera 2.

  However, when the traveling speed of the vehicle 10 is too high, the radar 3 becomes difficult to detect a speed difference between a stationary object such as a building and a moving object. Therefore, the determination unit 35 determines whether there is a possibility that a moving object exists in the blind spot range of the camera 2 when the traveling speed of the vehicle 10 is a predetermined speed (for example, 5 km / h) or less.

  Thereby, the object detection device 1 determines whether or not there is a moving object in the blind spot range of the camera 2 even when the vehicle 10 is traveling if the traveling speed of the vehicle 10 is equal to or less than a predetermined speed. Can be accurately determined.

  In the above-described embodiment, the case has been described in which the camera 2 covers the entire detection range 2A, and the stationary object only flag is set to ON only when there is only a stationary object in the detection range 2A. It is.

  Next, with reference to FIG. 10A and FIG. 10B, the modification 1 and the modification 2 of the switching mode of the stationary object only flag by the camera 2 are demonstrated. FIG. 10A is an explanatory diagram of a switching mode of a stationary object only flag according to the first modification of the embodiment. FIG. 10B is an explanatory diagram of a switching mode of the stationary object only flag according to the second modification of the embodiment.

  As shown in FIG. 10A, the camera 2 divides, for example, an image P to be captured into four regions P11, P12, P13, and P14, and each of the regions P11, P12, P13, and P14 has a stationary object only flag. May be configured to set.

  For example, when the camera 2 configured as described above detects a moving object in each of the three areas P11, P12, and P13 and does not detect a moving object only in the rightmost area P14, only the stationary object only flag in the rightmost area P14 is set to ON. The other stationary object only flag is turned OFF and transmitted to the radar 3.

  As a result, the radar 3 determines whether or not there is a possibility that a moving object exists in the blind spot range only for the detection range corresponding to the rightmost region P14, so that the processing load can be reduced.

  As shown in FIG. 10B, for example, the camera 2 sets the optical axis direction of the camera lens to 0 °, sets the clockwise direction from the optical axis direction around the camera lens as a positive angle, and counterclockwise angle. As an alternative, the range in which no moving object is detected may be notified by radar.

  The camera 2 having such a configuration detects, for example, a pedestrian T8 that crosses the front of the camera 2 in a range of −15 ° to 0 ° without detecting a moving object in a range of −30 ° to −15 °, and 0 ° There is a case where the oncoming vehicle T9 approaching in the range of ˜ + 30 ° is detected.

  In such a case, the camera 2 associates the ON stationary object only flag with the angle range of −30 ° to −15 ° and associates the OFF stationary object only flag with the angle range of −15 to + 30 ° to the radar 3. Send. As a result, the radar 3 determines whether or not there is a possibility that a moving object exists in the blind spot range only for the angle range of −30 ° to −15 °, so that the processing load can be reduced.

  Further effects and modifications can be easily derived by those skilled in the art. Thus, the broader aspects of the present invention are not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications can be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

DESCRIPTION OF SYMBOLS 1 Object detection apparatus 2 Camera 21 Imaging part 22 Image processing ECU
23 Detection unit 3 Radar 31 Antenna unit 32 Radar ECU
33 Signal Processing Unit 34 Recognition Processing Unit 35 Judgment Unit 4 Alerting Unit

Claims (6)

A camera device and a radar device;
The camera device is
An imaging unit for imaging a predetermined range around the vehicle;
A detection unit that detects a moving object existing in the predetermined range based on an image captured by the imaging unit;
The radar device is
When the radar apparatus detects a moving object during a period in which the moving object is not detected by the camera device, it is determined that there is a possibility that the moving object exists in a blind spot range that is a blind spot of the camera device in the predetermined range. An object detection apparatus comprising: a determination unit that performs When it is determined by the determination unit that there is a possibility that a moving object is present in the blind spot range, the display unit is provided with a warning unit that superimposes warning information on the image being displayed. The object detection apparatus according to claim 1. The alerting part is
The object detection device according to claim 2, wherein the blind spot range is estimated based on a road shape determined from map information, and the alert information is displayed at a position corresponding to the blind spot range in the image. The alerting part is
The said alerting | reporting information is displayed on the position corresponding to the said blind spot range in the said image by estimating the said blind spot range resulting from the other vehicle in the said image. Object detection device. The determination unit
It is determined whether there is a possibility that a moving object exists in the blind spot range when the traveling speed of the vehicle is equal to or lower than a predetermined speed. Object detection device. An imaging step of imaging a predetermined range around the vehicle with a camera device;
A detection step of detecting an object present in the predetermined range based on an image captured by the imaging step;
When the radar device detects a moving object during a period in which no moving object is detected by the camera device, the determination is made that there is a possibility that the moving object may exist in a blind spot range that is a blind spot of the camera device in the predetermined range. An object detection method comprising the steps of:

Images