JP7247772B2 - Information processing device and driving support system - Google Patents

Description

The present disclosure relates to technology for recognizing objects around a vehicle.

2. Description of the Related Art Conventionally, there is known a technique of detecting an object around a vehicle based on an image captured by a camera mounted on the vehicle and performing driving assistance based on the detection result. For example, Patent Literature 1 below proposes a technique for outputting the accuracy of the position of a detected object based on how parallax information is generated from images captured by a stereo camera.

JP 2016-148962 A

As a result of detailed studies by the inventor, the following problems were found. Even if the object detection accuracy deteriorates due to raindrops adhering to the windshield in front of the photographing device, the object detection accuracy may be quickly recovered if the raindrops are removed by a wiper or the like. Therefore, if it can be determined with high accuracy that the reason why the object is no longer detected is due to raindrops, it is possible to perform appropriate control that is different from the case where the object does not actually exist.

One aspect of the present disclosure provides a new technique for determining raindrop adhesion.

One aspect of the present disclosure is an information processing device (13) used in a vehicle (3) including an imaging device (11) and a driving support device (12). The information processing device includes a detection section (31) and a raindrop determination section (37). The detection unit is configured to detect feature points indicating objects existing around the vehicle in surrounding images (52, 53), which are images acquired using the imaging device and which show the surroundings of the vehicle. The raindrop determination unit compares the number of feature points included in the predetermined regions (61, 62) in the peripheral image where the object exists with the number of feature points included in the predetermined region in the peripheral image acquired in the past. It is configured to determine a raindrop adhering state in which raindrops adhere to the front of the photographing device when the amount of raindrops has decreased beyond a predetermined criterion.

According to such a configuration, it can be determined that raindrops have adhered when the number of feature points decreases beyond a predetermined criterion.
Another aspect of the present disclosure may further include an estimating section (39) and an output section (40) in the above one aspect of the present disclosure. The estimating unit is configured to acquire estimation information indicating the estimated behavior of the object based on the feature points detected by the detecting unit. The output unit is configured to output the estimation information acquired by the estimation unit to the driving assistance device when a predetermined output condition is satisfied. The output condition includes as a requirement that the raindrop determining unit determines that the raindrop is attached.

According to such a configuration, when it is determined that raindrops are attached, that is, when the number of feature points is reduced and it becomes difficult to detect an object by image recognition, the estimated information is output to the driving assistance device. can do. Therefore, the driving support device can continue driving support using the estimated information, and can extend the duration of driving support.

Another aspect of the present disclosure is a driving assistance system (1) mounted on a vehicle (3) and used, comprising a photographing device (11), a driving assistance device (12), and information of another aspect described above. a processor (13). According to such a configuration, it is possible to realize the effect of extending the duration of driving assistance by providing the information processing device of another aspect of the present disclosure described above.

1 is a block diagram showing the configuration of a driving assistance system; FIG. 1 is a functional block diagram of an information processing device according to an embodiment; FIG. It is a flow chart of information output processing. 9 is a flowchart of first density calculation processing; It is a figure explaining the calculation method of the area of a detection frame. It is a figure explaining the calculation method of the number of parallax points in a detection frame. 9 is a flowchart of second density calculation processing; It is a figure explaining movement of preceding vehicles. It is a figure explaining the calculation method of the area of an estimation frame. It is a figure explaining the calculation method of the number of parallax points in an estimation frame. It is a functional block diagram of an information processing device of a modification. FIG. 10 is a diagram illustrating an example of processing when a plurality of detection frames are generated;

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[1. embodiment]
[1-1. composition]
A driving assistance system 1 shown in FIG. 1 is a system mounted on a vehicle 3 and used, and includes a stereo camera 11 , a driving assistance device 12 , and an information processing device 13 .

The stereo camera 11 includes a right camera 11a and a left camera 11b. The right camera 11a and the left camera 11b are equipped with image sensors such as CCD sensors and CMOS sensors that are synchronized with each other. The right camera 11a and the left camera 11b are mounted, for example, inside the windshield of the vehicle 3 at the same height from the road surface with a predetermined gap in the vehicle width direction. The right camera 11a and the left camera 11b repeatedly photograph the surroundings of the vehicle 3 at predetermined intervals (for example, 0.1 second intervals). The stereo camera 11 corresponds to a photographing device.

The driving support device 12 is a device provided with a microcomputer including a CPU, ROM, RAM and I/O (not shown). The driving assistance device 12 is configured to be capable of controlling so-called adaptive cruise control (hereinafter referred to as ACC) that accelerates and decelerates so as to maintain a distance to the preceding vehicle. The driving assistance device 12 executes ACC using object information or estimation information, which will be described later, output by the information processing device 13 .

The information processing device 13 includes a microcomputer having a CPU 21 and a semiconductor memory such as RAM or ROM (hereinafter referred to as memory 22). Each function of the information processing device 13 is realized by executing a program stored in the memory 22 by the CPU 21 .

The information processing device 13 includes a parallax detection unit 31 and a raindrop determination unit 37, as shown in FIG. The information processing device 13 also includes an object detection unit 32, an object tracking unit 33, a distance calculation unit 34, a speed calculation unit 35, a type determination unit 36, an operation determination unit 38, an estimation unit 39, and an output unit. 40 and .

The parallax detection unit 31 detects parallax points, which are feature points representing objects shown in parallax images acquired using the stereo camera 11 . Objects that appear in the parallax image are objects existing around the vehicle 3 . A parallax point can be obtained by a known technique. For example, the parallax detection unit 31 converts the analog data of the left and right images acquired each time the right camera 11a and the left camera 11b take pictures into digital data of a predetermined luminance gradation for each pixel. Then, the parallax detection unit 31 performs stereo matching of the left and right digital images to calculate parallax and generate a parallax image. An object to be processed in the present embodiment is a preceding vehicle running in front of the vehicle 3 in the same direction as the vehicle 3 . A parallax image corresponds to a peripheral image. Also, the parallax detection unit 31 corresponds to the detection unit. It should be noted that the detection of the object from the parallax image is executed by the object detection unit 32 using a known method.

The raindrop determining unit 37 determines whether or not the stereo camera 11 is in a raindrop adhering state in which raindrops are attached in front of the stereo camera 11 . The front of the stereo camera 11 is a transmissive member installed between the lens of the camera and the subject, and corresponds to the windshield in this embodiment.

The above-described raindrop determining unit 37 compares the number of parallax points included in a predetermined region in which an object exists in the parallax image with the number of parallax points included in the predetermined region in the parallax image captured in the past. It is determined that raindrops are attached when the amount decreases beyond the standard. Specifically, the raindrop determining unit 37 determines the number of parallax points in a predetermined region of the parallax image captured at the first timing, and the number of parallax points in the parallax image captured at the second timing after the first timing. When the number of parallax points in the predetermined area and the rate of change are equal to or greater than a predetermined threshold value, it is determined that raindrops are attached.

The operation determination unit 38 determines whether or not a wiper (not shown) mounted on the vehicle 3 is in operation. "While the wiper is in operation" includes the period when the wiper is stopped when the wiper is intermittently operated.

The estimation unit 39 acquires object information and estimation information. These pieces of information are information indicating the behavior of the preceding vehicle, and include information regarding the position or movement of objects included in the parallax images. In the following description, an object refers to an object detected in parallax images.

The object information includes object distance, longitudinal relative velocity, object lateral position, lateral relative velocity, object lateral width, object type, and the like. The object distance is the distance from the vehicle 3 to the detected object. The longitudinal relative velocity is the relative velocity of an object with respect to the vehicle 3 in the traveling direction of the vehicle 3 . The relative velocity can be calculated by using feature points related to a plurality of past parallax images. The object lateral position is the position of the object in the direction perpendicular to the traveling direction of the vehicle 3 with respect to the vehicle 3 . The lateral relative velocity is the relative velocity of an object with respect to the vehicle 3 in a direction perpendicular to the traveling direction of the vehicle 3 . The object width is the length of the object in the orthogonal directions mentioned above. The object type is the type of object estimated from the size, moving speed, shape, and the like of the object.

The estimation unit 39 also acquires object information based on the parallax points detected by the parallax detection unit 31 . The estimation unit 39 identifies the position of the object based on the position of the parallax point on the image and the distance indicated by the parallax point. Also, the velocity of the object is obtained using the parallax points of the preceding vehicle included in the current and past parallax images. The object information is information indicating the current behavior of the preceding vehicle, which is obtained using the parallax points of the preceding vehicle included in the latest parallax image.

The estimated information includes the same types of content as the object information, that is, object distance, vertical relative velocity, horizontal position of object, relative horizontal velocity, horizontal width of object, type of object, and the like. On the other hand, unlike the object information, the estimation information is information indicating the behavior estimated for the preceding vehicle. In other words, it is information that indicates the predicted behavior of the position and speed of the preceding vehicle at the timing after the timing at which a certain parallax image is acquired (hereinafter referred to as the reference timing). Each content of estimated information is obtained as follows. The object distance is calculated in consideration of the movement of the vehicle 3, assuming that the object has a constant velocity at the reference timing. As for the vertical relative velocity, the horizontal position of the object, the horizontal width of the object, and the type of the object, the values and contents at the reference timing are used as they are. The lateral relative velocity is set to "0". In other words, it is presumed that relative movement in the lateral direction has not occurred.

The output unit 40 outputs the object information and estimated information acquired by the estimation unit 39 to the driving support device 12 . The estimated information is output when a predetermined output condition is satisfied. The output condition is a condition that includes as a requirement that the raindrop determination unit 37 determines that the raindrop is attached. In the present embodiment, it is included as one of the requirements for satisfying the output condition that the operation determination unit 38 determines that the operation is being performed. If the output condition is not satisfied, the object information is output to the driving support device 12 as normal processing.

[1-2. process]
Next, information output processing executed by the information processing device 13 will be described using the flowchart of FIG. This processing is executed each time an image is acquired by the stereo camera 11 . Note that in the following processing, the detected object is the preceding vehicle.

First, in S1, the parallax detection unit 31 performs parallax calculation. Here, the parallax detection unit 31 generates a parallax image including parallax points based on the left and right captured images captured by the right camera 11a and the left camera 11b.

In S2, the object detection unit 32 detects an object from the parallax images. At this time, a detection frame may be set for each detected object. A detection frame is a frame that specifies a predetermined area where the object exists on the image. Also, the detection frame may be set in a process different from the process described in this embodiment. Details of the detection frame will be described later.

In S3, the object tracking unit 33 tracks the object. Here, the object detected in the parallax image (hereinafter also referred to as the previous image) one time before the parallax image acquired this time (hereinafter also referred to as the current image) is tracked and identified in the current image. . A specific method of object tracking is not particularly limited. For example, the likelihood of the object detected in the previous image and the object detected in the current image may be calculated, and the objects whose likelihood is equal to or greater than a predetermined threshold may be determined to be the same object. Also, if there are a plurality of objects in the current image, the one with the highest likelihood may be determined to be the same object. Note that the current image is an example of the parallax image captured at the first timing described above, and the previous image is an example of the parallax image captured at the second timing described above.

In S4, the distance calculator 34 and the speed calculator 35 calculate the position and speed of the object tracked in S4. The speed of the object can be calculated from, for example, the speed of the vehicle 3 and the relative speed between the vehicle 3 and the object.

In S5, the type determination unit 36 performs template matching and the like to determine the type of the object detected in S2.
In S6, the type determination unit 36 determines whether or not the object detected in S2 is the object being tracked. The object being tracked is the preceding vehicle. When the type determination unit 36 determines in S6 that the object is not being tracked, the information output processing in FIG. 3 is terminated. On the other hand, when the type determination unit 36 determines in S6 that the object is being tracked, the process proceeds to S7.

In S7, the raindrop determination unit 37 performs a first density calculation process for calculating the parallax point density in the previous image with respect to the previous detection frame.
Here, the first density calculation process executed by the information processing device 13 will be described using the flowchart of FIG.

First, in S21, the raindrop determination unit 37 calculates the area of the previous detection frame. The previous detection frame is a detection frame set for the object detected in the previous image. The previous detection frame may be set in the process of S2 in FIG. 3 based on the previous image. As shown in FIG. 5, area S is the product of height H and width W of detection frame 61 in image 51 . The detection frame 61 may be set as, for example, a rectangular area surrounding the preceding vehicle 101 that is the object detected by the object detection unit 32 . In addition, in FIG. 5 and FIGS. 9 and 12 to be described later, the display of parallax points is omitted in order to facilitate understanding of the drawings.

In S22, the raindrop determination unit 37 calculates the number of parallax points N within the previous detection frame for the previous image. As shown in FIG. 6, the parallax image 52 includes a large number of parallax points 71 around edge portions on the image. The number of parallax points N is the number of parallax points 71 located inside the detection frame 61 in the parallax image 52 . The distance information of the parallax points can be indicated by colors, for example, but the display of the distance information is omitted in FIG. 6 and FIG. 10 described later for easy understanding of the drawings.

In S23, the raindrop determination unit 37 calculates the parallax point density Dens _pre . The parallax point density Dens _pre is calculated as follows: Dens _pre = number of parallax points N/area S. After S23, the process proceeds to S8.

Returning to FIG.
In S8, the raindrop determination unit 37 performs a second density calculation process for calculating the parallax point density in the current image with respect to the estimation frame from the previous time.

Here, the second density calculation process executed by the information processing device 13 will be described using the flowchart of FIG. 9 and 10 are examples of images blurred by raindrops on the windshield.

First, in S31, the raindrop determination unit 37 sets an estimation frame to be used in the current image from the object information acquired in the previous image. The estimation frame is a detection frame estimated from the object information of the previous image, and is set at a position where the object is predicted to exist in the current image regardless of the object detected in the current image. As shown in FIG. 8, the preceding vehicle 101a at the previous image acquisition moves to the position of the preceding vehicle 101b at the current image acquisition. The position of the preceding vehicle 101b can be estimated from the position and speed of the preceding vehicle 101a at the time of previous image acquisition. Then, by transforming the estimated position of the preceding vehicle 101b into the image coordinate system, the position of the estimation frame in the current image is set.

At S32, the raindrop determination unit 37 calculates the area S of the estimated frame 62 estimated at S31 in the image 53, as shown in FIG.
In S33, the raindrop determination unit 37 calculates the number of parallax points N within the estimation frame 62 for the current image. The number of parallax points N here is the number of parallax points 72 positioned inside the estimation frame 62 in the parallax image 54 as shown in FIG.

In S34, the raindrop determination unit 37 calculates the parallax point density Dens _cur . The parallax point density Dens _cur is calculated as follows: Dens _cur = number of parallax points N/area S. After S34, the process proceeds to S9.

Returning to FIG.
In S9, the raindrop determining unit 37 calculates a parallax point density ratio R _dens . R _dens is determined as R _dens =Dens _cur /Dens _pre .

Subsequently, in S10, the raindrop determination unit 37 determines whether or not R _dens <R _th . That is, the raindrop determination unit 37 determines whether or not the raindrops are attached. Rth here is a predetermined reference value. In other words, the case where the above conditions are satisfied is the case where the number of parallax points in a predetermined area between the previous image and the current image sharply decreases exceeding a predetermined rate of change. When raindrops adhere to the front of the stereo camera 11, the parallax point density drops rapidly at that timing. On the other hand, when there is no significant change in the density of parallax points, or when the density of parallax points gradually decreases, the probability of being affected by raindrops is low. Therefore, when the above conditions are satisfied, it is determined that the raindrops are attached.

When the raindrop determination unit 37 determines that R _dens <R _th in S10, the process proceeds to S11.
In S11, the operation determination unit 38 determines whether the wipers are in operation. If the operation determination unit 38 determines in S11 that the wipers are not in operation, the information output processing of FIG. 3 is terminated. At this time, the output unit 40 outputs the object information to the driving support device 12 as normal processing.

On the other hand, when the actuation determination unit 38 determines in S11 that the wipers are in operation, the process proceeds to S12. That is, when it is determined that the output condition is satisfied through S10 and S11, the process proceeds to S12.

In S12, the estimation unit 39 generates estimation information, and the output unit 40 extrapolates. That is, the output unit 40 outputs the estimated information to the driving assistance device 12 . Normal processing by the output unit 40 is temporarily stopped. After that, this process is terminated.

When the raindrop determination unit 37 determines that R _dens <R _th in S10 described above, the process proceeds to S13.
In S13, the output unit 40 determines whether or not it is in the previous extrapolation state. The previous extrapolation state is a state in which extrapolation of estimated information is executed in S12 and normal processing is stopped. When the output unit 40 determines in S13 that it is in the previous extrapolation state, the output unit 40 proceeds to S14, returns to normal processing, outputs the object information to the driving support device 12, and then performs the information output processing of FIG. finish.

On the other hand, when the output unit 40 determines in S13 that the previous image is not in the previous extrapolation state, it ends the information output processing of FIG. At this time, the output unit 40 outputs the object information to the driving support device 12 as normal processing.

[1-3. effect]
According to the embodiment detailed above, the following effects are obtained.
(1a) According to the information processing device 13, raindrops can be detected based on changes in density of parallax points. In addition, the information processing device 13 can output estimated information to the driving assistance device 12 when it is determined that raindrops are attached, that is, when it becomes difficult to detect an object by image recognition. Therefore, the driving support device 12 can continue driving using the estimated information.

It should be noted that the raindrop adhesion state is a state in which the number of parallax points has rapidly decreased, in other words, a case in which parallax images based on the stereo camera 11 have rapidly deteriorated. A cause of such a situation is, for example, the adhesion of raindrops to the windshield. Normally, the raindrops are removed periodically by the wiper, so the situation in which the parallax image is degraded by the raindrops is resolved in a relatively short period of time. Therefore, by extrapolating the estimated information when it is determined that raindrops are attached, it is possible to suppress control stoppage of the driving assistance device 12 due to a short-term object loss, and extend the duration of driving assistance. .

(1b) The information processing device 13 includes, as a requirement, that the wiper is in operation as an output condition for outputting the estimated information. If the wipers are in operation, it is highly probable that it is raining or the windshield is dirty. Therefore, the information processing device 13 can accurately determine a scene in which the profit of outputting the estimated information is large.

(1c) According to the information processing device 13, by using the parallax points included in the parallax images acquired by the stereo camera 11, highly accurate object recognition is possible.
(1d) According to the information processing device 13 , whether or not raindrops are attached is determined based on the change rate of the density of parallax points in the previous detection frame 61 and the estimation frame 62 . Therefore, compared to the case where the number of parallax points itself is used as a determination criterion, it is possible to suitably determine that the parallax image has rapidly deteriorated, that is, that it is affected by raindrops or the like.

[2. Other embodiments]
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and various modifications can be made.

(2a) In the above embodiment, the parallax point was exemplified as the feature point included in the peripheral image, but the present invention is not limited to this. For example, when detecting edge points from a captured image and using the edge points, the edge points may be used as feature points. If points other than parallax points such as edge points are used as feature points, a camera other than a stereo camera may be used. Note that an edge point is a portion where the brightness of pixels adjacent in the horizontal direction or the vertical direction changes abruptly.

(2b) In the above embodiment, the content of the driving assistance executed by the driving assistance device 12 is ACC. However, the driving assistance executed by the driving assistance device 12 may be driving assistance other than ACC. For example, the driving assistance device 12 may be configured to perform a lane departure prevention function or an automatic driving function.

(2c) In the above embodiment, the rectangular detection frame and estimation frame were exemplified as the frame for specifying the predetermined area. However, the specific form of the predetermined area is not particularly limited as long as the area includes the object to be detected. For example, the predetermined area may be set so as to match the outer shape of the detected object in the image.

(2d) In the above embodiment, the change rate of the density of the parallax points in the detection frame and the estimation frame is calculated, and when the change rate is greater than a predetermined threshold value, it is determined that the raindrops are applied. . However, the specific method is not particularly limited as long as it can be determined that the number of feature points in the predetermined area of the current image is reduced by more than a predetermined criterion compared to the feature points in the predetermined area of the previous image. For example, when calculating the number of parallax points for the current image, the number of parallax points in the detection frame newly set for the current image may be used instead of using the estimation frame set based on the previous image. Alternatively, the rate of change may be obtained by comparing the number of parallax points themselves without using the density.

(2e) In the above-described embodiment, the configuration is illustrated in which the output of the estimation information is terminated when it is determined in S10 of FIG. 3 that the raindrops are attached. However, the timing of ending the output of the estimation information may be timing other than the above. For example, the output of the estimated information may be terminated after a certain period of time from when it is determined that the raindrops are attached. In addition, the output unit 40 terminates the output of the estimated information when the number of parallax points in the predetermined area becomes equal to or greater than a predetermined threshold after the raindrop determination unit 37 determines that the raindrops are attached. may be configured to According to such a configuration, when the number of parallax points in the detection frame recovers to the predetermined threshold value, the output of the estimation information is stopped and the object information is output. Therefore, it is possible to prevent the estimated information from being output even when the accuracy of the object information is high, and it is possible to prevent an increase in the processing load.

(2f) In the above embodiment, the output unit 40 outputs the object information to the driving support device 12 as normal processing, and in S12, the normal processing is temporarily stopped. However, the method of outputting the object information by the output unit 40 is not particularly limited, and for example, the object information may be output all the time.

(2g) In the above embodiment, the conditions for the output are raindrops and wiper operation. However, the output conditions are not limited to the above contents. For example, wiper operation need not be a requirement for output conditions. Further, like the information processing device 13A shown in FIG. One of the requirements of the output condition may be that the unit 41 determines that the vehicle is in a straight running state. The determination by the road determination unit 41 may be performed based on the image captured by the imaging device, or may be performed by comparing the current position of the vehicle 3 with map information including information indicating whether the road is a straight road. good too.

(2h) As shown in the image 55 of FIG. 12, the detection frame includes the preceding vehicles 101c and 101d existing in the range of the lane 81 on which the vehicle 3 is traveling and the lane 82 adjacent to the lane. It may be set as a target. Further, when a plurality of detection frames 64 and 65 exist in the image 55, determination by the raindrop determination unit 37, acquisition of estimated information by the estimation unit 39, and output by the output unit 40 are performed for each of the plurality of detection frames. and may be configured to be executed.

(2i) Information processors 13, 13A and techniques described in this disclosure are provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. may be implemented by a dedicated computer designed for Alternatively, the information processing apparatus and techniques described in this disclosure may be implemented by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the information processing apparatus and method thereof described in the present disclosure are a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. may be implemented by one or more dedicated computers configured by Computer programs may also be stored as computer-executable instructions on a computer-readable non-transitional tangible storage medium. The method for realizing the functions of the units included in the information processing apparatus does not necessarily include software, and all the functions may be realized using one or more pieces of hardware.

(2j) A plurality of functions possessed by one component in the above embodiment may be realized by a plurality of components, or a function possessed by one component may be realized by a plurality of components. . Also, a plurality of functions possessed by a plurality of components may be realized by a single component, or a function realized by a plurality of components may be realized by a single component. Also, part of the configuration of the above embodiment may be omitted. Moreover, at least part of the configuration of the above embodiment may be added or replaced with respect to the configuration of the other above embodiment.

(2k) In addition to the information processing device 13 described above, a system having the information processing device 13 as a component, a program for causing a computer to function as the information processing device 13, and a non-transitional device such as a semiconductor memory storing this program The present disclosure can also be implemented in various forms such as a physical recording medium and a raindrop determination method.

REFERENCE SIGNS LIST 1 driving support system 3 vehicle 11 stereo camera 12 driving support device 13, 13A information processing device 31 parallax detection unit 37 raindrop determination unit 39 estimation unit 40 output unit 52, 54... parallax images, 61, 64... detection frames, 62... estimation frames, 71, 72... parallax points

Claims (9)

An information processing device (13, 13A) used in a vehicle (3) including an imaging device (11) and a driving support device (12),
A detection unit ( 31) and
comparing the number of feature points included in a predetermined region (61, 62) in the peripheral image where the object exists, with the number of feature points included in the predetermined region in the peripheral image acquired in the past; a raindrop determination unit (37) configured to determine a raindrop adhesion state in which raindrops adhere to the front of the photographing device when the raindrops have decreased beyond a predetermined standard ,
The raindrop determination unit determines the number of feature points in the predetermined region of the peripheral image captured at a first timing, and the number of the peripheral images captured at a second timing after the first timing. The information processing apparatus configured to determine that the raindrops are attached when a rate of change in the number of feature points in the predetermined area is equal to or greater than a predetermined threshold value. The information processing device according to claim 1,
The imaging device is a stereo camera,
The peripheral image is a parallax image acquired using the stereo camera,
The information processing apparatus, wherein the feature point is a parallax point (71, 72) shown in the parallax image. The information processing device according to claim 1 or claim 2 ,
The information processing apparatus, wherein the predetermined area is set for the object existing in a range of a lane (81) on which the vehicle is traveling and a lane (82) adjacent to the lane. The information processing device according to any one of claims 1 to 3 ,
an estimating unit (39) configured to acquire estimated information indicating a behavior estimated for the object based on the feature points detected by the detecting unit;
The estimated information acquired by the estimating unit is output to the driving support device when an output condition including as a requirement that the raindrop determining unit determines that the raindrops are attached is satisfied. and an output unit (40) configured. The information processing device according to claim 4 ,
Furthermore, an operation determination unit (38) configured to determine whether the wiper mounted on the vehicle is in operation,
The information processing apparatus, wherein the output condition includes, as one of requirements, that the operation determination unit determines that the operation is being performed. The information processing device according to claim 4 or claim 5 ,
The output unit terminates the output of the estimated information when the number of feature points in the predetermined area becomes equal to or greater than a predetermined threshold after the raindrop determination unit determines that the raindrops are attached. An information processing device configured to: The information processing device according to any one of claims 4 to 6 ,
Furthermore, a road determination unit (41) configured to determine whether or not the vehicle is in a straight running state in which the vehicle is running on a straight road,
The information processing apparatus, wherein the output condition includes, as one of requirements, that the road determination unit determines that the vehicle is in a straight running state. The information processing device according to any one of claims 4 to 7 ,
when a plurality of the predetermined regions exist in the peripheral image, the determination by the raindrop determination unit, the acquisition of the estimated information by the estimation unit, and the output by the output unit for each of the plurality of predetermined regions and an information processing device on which . A driving support system (1) mounted on a vehicle (3) and used,
an imaging device (11);
a driving support device (12);
A driving support system comprising the information processing device (13, 13A) according to any one of claims 4 to 8 .

Images