JP2018101295A - Object detection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an object detection device capable of reliably determining the presence or absence of an object in a blind spot region both when the object exits the blind spot region and when the object stays in the blind spot region. When an object is located in a blind spot peripheral region AB surrounding a blind spot region DBL at time t-1, and the direction of the relative velocity of the object at time t-1 is the direction of the blind spot region DBL, the blind spot region DBL is set. It is a candidate for a blind spot area where an object invades at time t. Then, when the object does not exist in the blind spot peripheral areas ALB and AB surrounding the candidate blind spot area DBL at time t, it is determined that the object has invaded the blind spot area DBL. Further, when the object exists in the blind spot peripheral areas ALB and AB surrounding the blind spot area DBL where the object is determined to exist at time t + α, it is determined that the object has exited from the blind spot area DBL. [Selection diagram] Fig. 1

Description

  The present invention relates to an object detection device suitable for use in an automatic driving system.

  Patent Document 1 discloses a technique for avoiding contact between the host vehicle and the object by detecting an object existing around the host vehicle using a plurality of radars. According to this technique, when an object that is determined to be the same object as the object that has been recognized as the control target object by the previous time enters the radar non-detection area (dead angle area), The object to be controlled is extrapolated in the non-detection area based on the relative position and the relative speed of. When the extrapolated control target object exists in the non-detection region, the operation content having a higher contact avoidance effect than the operation content when the extrapolated control target object exists outside the non-detection region. Is set.

JP 2009-212557 A

  However, in the above-described prior art, even after the real object has left the blind spot area, the extrapolated object with low accuracy remains in the blind spot area for a long time, so there is a risk that it is erroneously determined that the real object remains in the blind spot area. is there. Further, since the extrapolated object with low accuracy has left the blind spot area even though the real object remains in the blind spot area, it may be erroneously determined that there is no real object in the blind spot area.

  The present invention has been made in view of the above-described problems, and reliably determines whether there is an object in the blind spot area both when the object leaves the blind spot area and when the object remains in the blind spot area. It is an object of the present invention to provide an object detection device that can do this.

In order to achieve the above object, an object detection device according to the present invention includes:
Object detection means for detecting an object existing around the host vehicle by a plurality of sensors;
Current time object information acquisition means for acquiring the relative position and relative speed of the object detected by the object detection means;
Previous time object information acquisition means for acquiring the relative position and relative speed of the object detected by the object detection means at the previous time;
Of the blind spot area outside the detection range of the object detection means, the object is located in the blind spot surrounding area surrounding the blind spot area at the previous time, and the direction of the relative speed of the previous time of the object is the direction of the blind spot area. Intruding candidate blind spot area judging means for outputting the object as a candidate for the blind spot area into which the object enters at the current time,
A blind spot area intrusion judging means for judging that the object has entered the blind spot area when there is no object at the current time in the blind spot surrounding area surrounding the blind spot area made a candidate by the intrusion candidate blind spot area judging means; ,
In the blind spot surrounding area surrounding the blind spot area where it is determined that the object exists at the previous time, if there is an object at the current time, a blind spot area exit determining means for determining that the object has exited from the blind spot area;
The blind spot area where the object exists at the previous time, the blind spot area where the object entered at the current time, and the blind spot area where the object exited at the current time are integrated, and the blind spot area where the object exists at the current time is integrated. A judgment integration means for judging;
It is characterized by providing.

  According to the object detection device of the present invention, by monitoring the detection state of the object in the blind spot surrounding area surrounding the blind spot area, both when the object leaves the blind spot area and when the object stays in the blind spot area, The presence or absence of an object in the blind spot area can be reliably determined.

It is a figure explaining the outline | summary of the object detection process which concerns on embodiment. 1 is a block diagram showing a configuration of an automatic driving system according to an embodiment. It is a flowchart which shows the flow of the object detection process which concerns on embodiment. It is a figure which shows the setting method of a blind spot periphery area | region.

  Embodiments of the present invention will be described below with reference to the drawings. However, in the embodiment shown below, when referring to the number of each element, quantity, quantity, range, etc., unless otherwise specified or clearly specified in principle, the reference However, the present invention is not limited to these numbers. Further, the structures described in the embodiments described below are not necessarily essential to the present invention unless otherwise specified or clearly specified in principle.

  First, the outline of the object detection process according to the embodiment will be described with reference to FIG. In the example shown in FIG. 1A, detection ranges SF, SB, SL, SR by a plurality of sensors are fanned out in the front-rear and left-right directions with the host vehicle as the center. In this example, between the front detection range SF and the left side detection range SL, between the front detection range SF and the right side detection range SR, between the rear detection range SB and the left side detection range SL, and behind Between the detection range SB and the right side detection range SR, blind spot areas DFL, DFR, DBL, DBR deviating from the detection range of the sensor are formed.

  This object detection process introduces the concept of a blind spot peripheral area. The blind spot peripheral area is an area surrounding the blind spot area, and is not a physically determined range such as a sensor detection range or a blind spot area but a range that can be arbitrarily set. A specific method for setting the blind spot peripheral area will be described later. In the example shown in FIG. 1A, for example, blind spot peripheral areas ALB and AB are set so as to surround the left rear blind spot area DBL, and blind spot peripheral areas ARB and AB are set so as to surround the right rear blind spot area DBR. Yes. Although the blind spot peripheral area is also set around the left front blind spot area DFL and the right front blind spot area DFR, illustration of these is omitted.

  Now, the object detection process is executed at a constant cycle. Objects existing around the host vehicle are detected every time by a plurality of sensors, and the detection results are stored. The object to be detected is a moving body, and includes vehicles such as automobiles, motorcycles, bicycles, and pedestrians. The detection result of the object includes a relative position and a relative speed of the detected object with respect to the own vehicle. Here, as shown in FIG. 1A, it is assumed that an object is detected in the blind spot peripheral area AB at the previous time t-1. At the current time t, the relative position and relative speed of the detected object at the previous time t−1 are acquired, and the blind spot area DBL that is in the direction in which the object is heading is specified. Then, as shown in FIG. 1B, when no object is detected in the blind spot surrounding area AB or ALB surrounding the blind spot area DBL at the current time t, it is determined that the object exists in the blind spot area DBL. . Thereafter, the determination that the object is present in the blind spot area DBL is continued until an object is detected again in the blind spot peripheral area ALB surrounding the blind spot area DBL at a certain time t + α.

  As can be seen from the above outline, this object detection process monitors the detection state of the object in the blind spot peripheral area, so that the object detection process is performed in the blind spot area both when the object leaves the blind spot and when the object stays in the blind spot. The presence or absence of an object can be determined with certainty. Hereinafter, the configuration of the automatic driving system to which the object detection process is applied and the details of the object detection process will be described.

  FIG. 2 is a diagram illustrating a configuration of the automatic driving system according to the embodiment. The automatic driving system performs automatic driving of a vehicle on which the automatic driving system is mounted. The automatic driving system 1 includes a control device 10 that functions as an object detection device, a lidar (LIDAR: Laser Imaging Detection and Ranging) 2 that is an autonomous recognition sensor, a radar 3, a stereo camera 4, a sonar 21, and an infrared sensor 22. . These autonomous recognition sensors are connected to the control device 10 directly or via a communication network such as a CAN (Controller Area Network) built in the vehicle. The control device 10 can grasp the situation of substantially the entire periphery of the vehicle from the object information obtained by the autonomous recognition sensor.

  The automatic driving system 1 further includes a GPS (Global Positioning System) receiver 5, a map database 6, a navigation system 7, an HMI (Human Machine Interface) 8, and an actuator 9. The GPS receiver 5 is means for acquiring position information indicating the position of the host vehicle based on a signal transmitted from a GPS satellite. The map database 6 is formed in a storage such as an HDD or an SSD mounted on the vehicle, for example. The map information included in the map database 6 includes road position information, road shape information, intersection and branch point position information, road lane information, and the like. The navigation system 7 calculates the route traveled by the host vehicle based on the position information of the host vehicle measured by the GPS receiver 5 and the map information in the map database 6, and the calculated route information is transmitted via the HMI 8. While transmitting to a driver | operator, it outputs to the control apparatus 10. The HMI 8 is an interface for outputting and inputting information between the occupant and the control device 10. The actuator 9 is a device that operates according to an operation signal from the control device 10 and changes the traveling state of the vehicle by the operation. The actuator 9 is provided in each of a drive system, a braking system, and a steering system, for example. In addition to these, the automatic driving system 1 is provided with an internal sensor for obtaining information on the traveling state of the host vehicle, such as a vehicle speed sensor and an acceleration sensor.

  The control device 10 is an ECU (Electronic Control Unit) having at least one processor, at least one ROM, and at least one RAM. The processor includes a register and a cache memory. The ROM stores various types of data including various programs and maps for automatic operation. Various functions are realized in the control device 10 by loading a program stored in the ROM into the RAM and executing the program by the processor. In addition, the control apparatus 10 may be comprised from several ECU.

  In FIG. 2, among functions for automatic driving that the control device 10 has, functions related to object detection are expressed in blocks. The illustration of other functions for automatic operation of the control device 10 is omitted.

  In the automatic driving system 1, an object existing around the host vehicle is detected by a rider 2, a radar 3, a stereo camera 4, a sonar 21 and an infrared sensor 22. However, there is a blind spot area outside the detection range of these sensors 2, 3, 4, 21, and 22 around the host vehicle. The control device 10 has a function of detecting not only an object within the detection range of the sensors 2, 3, 4, 21 and 22, but also an object within a blind spot area outside the detection range. This function includes an object detection unit 11, a current time target information acquisition unit 12, a previous time target information acquisition unit 13, an intrusion candidate blind spot area determination unit 14, a blind spot area intrusion determination unit 15, and a blind spot area exit. This is realized by the determination unit 16 and the determination integration unit 17. However, these units 11, 12, 13, 14, 15, 16, and 17 do not exist as hardware in the control device 10, but are software-like when the program stored in the ROM is executed by the processor. To be realized.

  The object detection unit 11 acquires measurement data of objects around the host vehicle from the sensors 2, 3, 4, 21, and 22. The measurement data includes information about the distance to the measurement point, information about the height of the measurement point, and information about the direction of the measurement point. In addition, depending on the sensor, information on the speed of the measurement point is included. The object detection unit 11 uses information obtained by the rider 2, uses information obtained by the radar 3, uses information obtained by the stereo camera 4, uses information obtained by the sonar 21, and an infrared sensor 22. Measurement data can be acquired by at least one method of using the information obtained in the above, and combining and using information of a plurality of types of sensors 2, 3, 4, 21, and 22 by sensor fusion. The object detection unit 11 clusters the acquired measurement data. Clustering is based on the position and height of each measurement point on the reference coordinate system with the X axis (vertical axis) in the front-rear direction of the host vehicle and the Y axis (horizontal axis) in the left-right direction centered on the host vehicle. Done. Further, for measurement data having speed information, clustering may be performed based on the speed. The specific method of clustering is not particularly limited, and a known clustering method can be used. The clustered measurement point group is surrounded by a rectangular frame in the reference coordinate system as one target. The position of the target in the reference coordinate system indicates the relative position of the detected object with respect to the host vehicle, and the range of the frame of the target indicates the range on the plane of the detected object. The object detection unit 11 corresponds to the object detection means described in claim 1.

  The current time target information acquisition unit 12 acquires the relative position and relative speed of the target at the current time on the reference coordinate system. The relative position and relative speed of the target on the reference coordinate system correspond to the relative position and relative speed of the object in real space. Therefore, the current time target information acquisition unit 12 corresponds to the current time object information acquisition means described in claim 1. The relative position of the target at the current time is obtained from the object detection unit 11. The relative speed of the target at the current time is calculated from the amount of change and the direction of change in the relative position of the target from the previous time to the current time. The current time target information acquisition unit 12 temporarily stores the acquired relative position and relative speed of the target at the current time in a temporary storage unit such as a cache memory or a register. In addition, the current time target information acquisition unit 12 inputs the relative position and relative speed of the acquired target at the current time to the blind spot area intrusion determination part 15 and the blind spot area exit determination part 16.

  The previous time target information acquisition unit 13 acquires a relative position and a relative speed of the target at the previous time on the reference coordinate system. Specifically, the relative position and the relative speed of the target stored in the temporary storage unit by the current time target information acquisition unit 12 at the previous time are read out. The previous time target information acquisition unit 13 inputs the relative position and relative speed of the acquired target at the previous time to the intrusion candidate blind spot area determination unit 14. The previous time target information acquisition unit 13 corresponds to the previous time object information acquisition means described in claim 1.

  Based on the relative position and relative speed of the target at the previous time input from the previous time target information acquisition unit 13, the intrusion candidate blind spot area determination unit 14 selects a blind spot area candidate for the target to invade at the current time. to decide. The intrusion candidate blind spot area determination unit 14 corresponds to the intrusion candidate blind spot area determination means described in claim 1. Specifically, the intrusion candidate blind spot area determination unit 14 performs the following processing.

  On the reference coordinate system, all blind spot areas formed around the host vehicle are set, and blind spot surrounding areas surrounding each blind spot area are set. The invasion candidate blind spot area determination unit 14 determines, for each blind spot area, whether or not the target is located in the surrounding area of the blind spot surrounding it. For the blind spot area where the target is located in the corresponding blind spot peripheral area at the previous time, it is determined whether the direction of the relative speed of the target at the previous time is the direction of the blind spot area. If the direction of the relative speed of the target at the previous time is the direction of the blind spot area, the blind spot area is determined as a candidate for the blind spot area where the target enters at the current time. The invasion candidate blind spot area determination unit 14 inputs the determined blind spot area candidate to the blind spot area intrusion determination unit 15.

  The blind spot area intrusion determining unit 15 specifies a blind spot surrounding area surrounding the blind spot area that is a candidate by the intrusion candidate blind spot area determining unit 14. Then, it is determined whether or not the relative position of the target at the current time input from the current time target information acquisition unit 12 is included in the identified blind spot peripheral area. The blind spot area intrusion determination unit 15 determines that the target has entered the candidate blind spot area when no target exists at the current time in the identified blind spot peripheral area. The blind spot area intrusion determining unit 15 inputs the blind spot area determined to have entered the target at the current time to the determination integrating unit 17. The blind spot area intrusion determination unit 15 corresponds to a blind spot area intrusion determination unit described in claim 1.

  The blind spot area exit determination unit 16 specifies a blind spot peripheral area surrounding the blind spot area in which the target is determined to exist at the previous time by the determination integration unit 17 described later. Then, it is determined whether or not the relative position of the target at the current time input from the current time target information acquisition unit 12 is included in the identified blind spot peripheral area. When there is a target at the current time in the identified blind spot peripheral area, the blind spot area exit determination unit 16 determines that the target has exited from the blind spot area where the target is determined to exist at the previous time. The blind spot area exit determination unit 16 inputs the blind spot area in which it is determined that the target has exited at the current time to the determination integration unit 17. The blind spot area exit determination unit 16 corresponds to the blind spot area exit determination means described in claim 1.

  The judgment integration unit 17 integrates the blind spot area where the target existed at the previous time, the blind spot area where the target entered at the current time, and the blind spot area where the target exited at the current time. The blind spot area where the target exists is output. The judgment integration unit 17 corresponds to a judgment integration unit described in claim 1. The information regarding the blind spot area where the target exists at the current time output from the judgment integration unit 17 is used, for example, in lane change control in automatic driving.

  The control device 10 functions as an object detection device by the functions of the above-described units 11, 12, 13, 14, 15, 16, and 17. FIG. 3 is a flowchart showing the flow of object detection processing according to the present embodiment. The control device 10 as the object detection process repeatedly performs the process shown in this flowchart at a predetermined cycle. In this flowchart, the current time is represented by t and the previous time is represented by t-1.

  Step S1 is a process of acquiring target information at the previous time t-1. The control device 10 acquires the relative positions and relative velocities of all targets on the reference coordinate system at the previous time t-1. The acquired information is used in step S3. Specifically, the result of the process in step S5 at the previous time t-1 held in the temporary storage means is read in this process.

  Step S2 is a process of setting a blind spot peripheral region. The control device 10 sets a blind spot peripheral area around each blind spot area on the reference coordinate system. The set blind spot peripheral area is used in steps S3, S6, and S7. The area around the blind spot can be set in various ways. FIG. 4 is a diagram illustrating an example of a method for setting a blind spot peripheral region. In the example shown in FIG. 4A, the blind spot peripheral areas ALB and AB are set with reference to the range of the possibility that the target moves to the blind spot area DBL from time t-1 to time t. In this case, the range of the blind spot peripheral areas ALB and AB is changed according to the magnitude and direction of the relative speed of the target. In the example shown in FIG. 4B, the blind spot peripheral areas ALB and ABL are set with reference to a range of a fixed distance from the edge of the blind spot area DBL. In the example shown in FIG. 4C, the blind spot peripheral areas ALB and ABL are set based on a range of a fixed angle from the edge of the blind spot area DBL.

  Step S3 is a process of determining the invasion candidate blind spot area. The control device 10 selects a blind spot area where the target is likely to enter at the current time t as an intrusion candidate blind spot area. The selected invasion candidate blind spot area is used in step S6. Specifically, the intrusion candidate blind spot area determination process in step S3 includes the following steps S3-1, S3-2, and S3-3.

  In step S3-1, it is determined based on the information acquired in step S1 whether the relative position at the current time t is included in the blind spot surrounding area set in step S2 for all the targets on the reference coordinate system. In step S3-1, in order to consider the error of the relative position, the determination may be made using the past relative position history from the previous time t-1. When it is determined that the target is included in the blind spot peripheral area, next, Step S3-2 is executed.

  In step S <b> 3-2, it is determined whether a blind spot area exists in the traveling direction of a target determined to be included in the blind spot peripheral area (hereinafter, a determination target target). When the traveling direction of the determination target is specified, the relative speed of the determination target is referred to. The relative speed to be referenced may be a relative longitudinal speed, a relative lateral speed, or a combined speed of the relative longitudinal speed and the relative lateral speed. In step S <b> 3-2, the traveling direction of the determination target may be specified using a history of relative speeds past from the previous time t−1 in order to take into account errors in relative speeds. Further, if the absolute value of the angle formed by the vector from the relative position of the determination target to the center of gravity of the blind spot area and the velocity vector of the determination target is equal to or less than the threshold value, the blind spot in the traveling direction of the determination target It may be determined that the region exists. And when it determines with a blind spot area | region existing in the advancing direction of a determination target, next, step S3-3 is performed.

  In step S3-3, the blind spot area determined to exist in the traveling direction of the determination target is selected as one of the target intrusion candidate blind spot areas. Note that the size of the determination target or the identification result may be used for the determination in step S3. For example, when the determination target is a large bus or truck, a part of the determination target protrudes from the blind area depending on the relationship with the size of the blind area. In such a case, even if a blind spot area exists in the traveling direction of the determination target, it may be determined that the determination target does not enter the blind spot area. Furthermore, in order to consider an error in the size of the determination target and an error in the identification result, the determination may be made using the size of the determination target in the past from the previous time t-1 or the history of the identification result.

  Step S4 is a process of acquiring a determination at the previous time t-1 regarding the blind spot area where the target exists. The control apparatus 10 acquires information regarding the blind spot area where the target exists at the previous time t-1 among the blind spot areas set in step S2. The acquired information is used in step S7. Specifically, the result of the process in step S8 at the previous time t-1 held in the temporary storage means is read in this process.

  Step S5 is a process of acquiring target information at the current time t. The control device 10 acquires the relative positions and relative speeds of all targets on the reference coordinate system at the current time t. Specifically, measurement of objects around the vehicle obtained from information of autonomous recognition sensors (including sonar and infrared sensors) such as rider 2, radar 3, stereo camera 4 or a combination of such information. The data is converted into a target, and the relative position and relative speed of the target on the reference coordinate system centered on the host vehicle are calculated.

  Step S6 is processing for determining whether the target has entered the blind spot area. The control device 10 determines the blind spot area where the target actually entered between the previous time t-1 and the current time t for the intrusion candidate blind spot area selected in step S3. The result of the intrusion determination is used in step S8. Specifically, the blind spot area intrusion determination process in step S6 includes the following steps S6-1, S6-2, and S6-3.

  In step S6-1, a blind spot peripheral area around the intrusion candidate blind spot area selected in step S3 is selected from the blind spot peripheral areas set in step S2.

  In step S6-2, based on the information acquired in step S5, it is determined whether the relative position at the current time t is included in the blind spot peripheral area selected in step S6-1 for all targets on the reference coordinate system. To do.

  In step S6-3, if there is a blind spot peripheral area that does not include the target at the current time t in the blind spot peripheral area selected in step S6-1, an intrusion candidate blind spot area corresponding to the blind spot peripheral area is determined. It is determined that this is a blind spot area where the target has intruded between the previous time t-1 and the current time t.

  Step S7 is a process for determining whether the target has left the blind spot area. Among the blind spots where it is determined that the target exists at the previous time t−1, the blind spot area where the target has exited at the current time t is determined. The result of the exit determination is used in step S8. Specifically, the blind spot exit determination process in step S7 includes the following steps S7-1, S7-2, and S7-3.

  In step S7-1, a blind spot peripheral area around the target blind spot area at the previous time t-1 acquired in step S4 is selected from the blind spot peripheral areas set in step S2.

  In step S7-2, based on the information acquired in step S5, it is determined whether the relative position at the current time t is included in the blind spot peripheral region selected in step S7-1 for all targets on the reference coordinate system. To do. In step S7-2, in order to prevent an error in determination, among the targets on the reference coordinate system at the current time t, those that have been detected in the past outside the blind spot peripheral region and selected in step S7-1. It may be determined by excluding those that have entered the blind spot surrounding area at the current time t. Moreover, in order to consider the error of the relative position, the determination may be made using the history of the relative position in the past from the current time t. Furthermore, the size of the target or the identification result may be used for the determination in step S7-2. For example, when the target is a large bus or truck, since the target is not hidden in the blind spot area, the target may be excluded from the target exited from the blind spot area. Furthermore, in order to consider an error in the size of the target and an error in the identification result, the determination may be made using the target size or the history of the identification result in the past from the current time t.

  In step S7-3, if there is a blind spot surrounding area including the target at the current time t in the blind spot surrounding area selected in step S7-1, the blind spot area corresponding to the blind spot surrounding area is set to the previous time t. It is determined that it is a blind spot area where the target has left between -1 and the current time t.

  Step S8 is a judgment integration process. The control device 10 determines at the previous time t-1 regarding the blind spot area where the target exists, determines regarding the blind spot area where the target has entered between the previous time t-1 and the current time t, and the previous time t. The determination regarding the blind spot area where the target has left between -1 and the current time t is integrated, and the blind spot area where the target exists at the current time t is output. Specifically, the blind spot area where the target has entered at the current time t obtained in step S6 is added to the blind spot area where the target existed at the previous time t-1 acquired in step S4, and obtained in step S7. The dead area where the target has left at the current time t is subtracted. The blind spot area thus obtained is output as a blind spot area where the target exists at the current time t. In step S8, in order to cope with a case where a plurality of targets enter or leave the same blind spot area, the number of judgments in steps S4, S6, and S7 is added or subtracted for each blind spot area set in step S2. You may judge the blind spot area | region where the result of addition / subtraction is one or more as a blind spot area | region where the target exists.

  According to the object detection process performed in the above procedure, the object exits from the blind spot area in real space by monitoring the detection state of the target in the blind spot surrounding area surrounding the blind spot area set on the reference coordinate system. In both the case and the case where the object stays in the blind spot area, it is possible to reliably determine the presence or absence of the object in the blind spot area.

1 Autonomous Driving System 2 Rider 3 Radar 4 Stereo Camera 5 GPS Receiver 6 Map Database 7 Navigation System 8 HMI
9 Actuator 10 Control device (object detection device)
11 Object detection unit (object detection means)
12 Current time target information acquisition unit (current time object information acquisition means)
13 Previous time target information acquisition unit (Previous time object information acquisition means)
14 Invasion candidate blind spot area determination unit (invasion candidate blind spot area determination means)
15 Blind spot area intrusion determination unit (Blind spot area intrusion determination means)
16 Blind Spot Area Exit Determination Unit (Bad Spot Area Exit Determination Unit)
17 Judgment Integration Unit (Judgment Integration Unit)
21 Sonar 22 Infrared sensor

Claims (1)

Object detection means for detecting an object existing around the host vehicle by a plurality of sensors;
Current time object information acquisition means for acquiring the relative position and relative speed of the object detected by the object detection means;
Previous time object information acquisition means for acquiring the relative position and relative speed of the object detected by the object detection means at the previous time;
Of the blind spot area outside the detection range of the object detection means, the object is located in the blind spot surrounding area surrounding the blind spot area at the previous time, and the direction of the relative speed of the previous time of the object is the direction of the blind spot area. Intruding candidate blind spot area judging means for outputting the object as a candidate for the blind spot area into which the object enters at the current time,
A blind spot area intrusion judging means for judging that the object has entered the blind spot area when there is no object at the current time in the blind spot surrounding area surrounding the blind spot area made a candidate by the intrusion candidate blind spot area judging means; ,
In the blind spot surrounding area surrounding the blind spot area where it is determined that the object exists at the previous time, if there is an object at the current time, a blind spot area exit determining means for determining that the object has exited from the blind spot area;
The blind spot area where the object exists at the previous time, the blind spot area where the object entered at the current time, and the blind spot area where the object exited at the current time are integrated, and the blind spot area where the object exists at the current time is integrated. A judgment integration means for judging;
An object detection apparatus comprising:

Images