Vehicle recognition device and vehicle recognition method

Abstract

To provide a technique capable of suppressing deterioration of accuracy of a recognition of a vehicle when opening a shutter in the case where the vehicle is recognized on the basis of a camera image containing the vehicle that goes out from a closed space.SOLUTION: A vehicle recognition device comprises: a processor that performs recognition processing of a vehicle contained in a camera image acquired by a camera installed in a closed space and opening / closing processing of a shutter provided to an outlet and an inlet of the vehicle in the closed space. The processor performs a processing for transmitting an automatic open command when a distance from the shutter to the vehicle becomes an initial setting distance or smaller to the shutter in the opening / closing processing. The processor further determines whether or not an outside irradiation degree is higher than an inner irradiation degree on the basis of information on the outside irradiation degree and information on the inner irradiation degree. Also, the processor further performs a whiteout countermeasure processing in order to suppress the generation of a whiteout at a position of the vehicle in the camera image in the case where it is determined that the outside irradiation degree is higher than the inner irradiation degree.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to a technology for recognizing vehicles. [Background technology]

[0002] Patent Document 1 discloses an imaging device that improves the visibility of a subject that a user wants to focus on. In this conventional technology, the face of a person who is a subject of interest is detected based on an image that combines low-illuminance and high-illuminance images captured by a camera, and the luminance distribution of the face area is calculated. Then, based on the luminance distribution of the face area of ​​the person in the low-illuminance or high-illuminance area, the exposure amount of the face area is controlled so as to approach the appropriate exposure range. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2011-254340 A Summary of the Invention [Problem to be solved by the invention]

[0004] Now, consider a scene in which a vehicle leaving a closed space having a certain extent, such as a garage, a parking lot, a factory, etc., is recognized by a camera installed in the closed space. In this case, when a shutter installed at the vehicle entrance to the closed space is opened, the amount of light received by the light receiving sensor installed in the camera may become saturated and whiteout may occur in the camera image depending on the weather and time of day around the closed space. If a vehicle is included in the whiteout part of the camera image, the camera may not be able to recognize the vehicle.

[0005] One objective of the present disclosure is to provide a technology that can suppress a decrease in accuracy of vehicle recognition when a shutter opens, when recognizing a vehicle based on a camera image that includes the vehicle leaving a closed space. [Means for solving the problem]

[0006] A first aspect of the present disclosure relates to a vehicle recognition device that recognizes a vehicle present in a closed space. The vehicle recognition device includes a storage device that stores a camera image acquired by a camera installed in the closed space, information on inner illuminance indicating the illuminance inside the closed space and in the shooting space of the camera, and information on outer illuminance indicating the illuminance outside the closed space, and a processor that performs recognition processing of the vehicle included in the camera image and opening and closing processing of a shutter provided at the entrance and exit of the vehicle in the closed space. In the opening and closing processing, the processor performs processing to send an automatic opening command to the shutter when the distance from the shutter to the vehicle becomes equal to or less than an initial setting distance. The processor further determines whether the outer illuminance is higher than the inner illuminance based on the information on the outer illuminance and the information on the inner illuminance. Furthermore, when it is determined that the outer illuminance is higher than the inner illuminance, the processor further performs whiteout countermeasure processing to suppress whiteout occurrence at the position of the vehicle in the camera image.

[0007] A second aspect of the present disclosure has the following features in addition to the first aspect: The storage device further includes information on an estimated time until the vehicle reaches the shutter. In the overexposure countermeasure processing, if the estimated time is less than a threshold, the processor further extends the initial set distance based on a distance set according to the estimated time.

[0008] A third aspect of the present disclosure has the following features in addition to the first aspect: The storage device further includes information on an estimated time until the vehicle reaches the shutter. In the overexposure countermeasure processing, if the estimated time is less than a threshold, the processor further generates a command to decelerate the vehicle based on a deceleration amount of the vehicle set according to the estimated time.

[0009] A fourth aspect of the present disclosure has the following features in addition to the first aspect. The storage device further includes shutter open / close information. The processor further determines whether the shutter is open or closed based on the open / close information. The processor further performs a brightness correction process to correct the brightness of the camera image when it is determined that the shutter is open and the outer illuminance is higher than the inner illuminance.

[0010] A fifth aspect of the present disclosure relates to a vehicle recognition method for recognizing a vehicle present in a closed space. The vehicle recognition method includes: performing a process of recognizing a vehicle included in a camera image acquired by a camera installed in the closed space; performing an opening and closing process of a shutter installed at an entrance and exit of the vehicle in the closed space; transmitting an automatic opening command to the shutter when the distance from the shutter to the vehicle becomes equal to or less than an initial setting distance in the opening and closing process; determining whether the external illuminance is higher than the internal illuminance based on information on external illuminance indicating the illuminance outside the closed space and information on internal illuminance indicating the illuminance inside the closed space and in the shooting space of the camera; and performing an anti-whiteout process to suppress the occurrence of whiteout at the position of the vehicle in the camera image when it is determined that the external illuminance is higher than the internal illuminance. Effect of the Invention

[0011] According to the present disclosure, when the external illuminance is higher than the internal illuminance, that is, when a situation is present in which whiteout in the camera image may occur, a whiteout countermeasure process is performed to prevent whiteout from occurring at the position of the vehicle in the camera image. This makes it possible to prevent the accuracy of recognition of a vehicle based on a camera image that includes the vehicle from decreasing when the shutter opens. [Brief description of the drawings]

[0012] [Figure 1] 1 is a diagram for explaining an overview of a vehicle recognition device according to an embodiment; [Diagram 2] 1 is a diagram for explaining a specific example of a vehicle recognition device according to an embodiment; [Diagram 3]1 is a diagram for explaining a specific example of a vehicle recognition device according to an embodiment; [Figure 4] 1 is a block diagram showing an example of a configuration of a vehicle recognition device according to an embodiment; [Diagram 5] 4 is a flowchart illustrating an example of processing performed by the vehicle recognition device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] A vehicle recognition device and a vehicle recognition method according to an embodiment of the present disclosure will be described with reference to the accompanying drawings. In addition, common elements in the drawings will be denoted by the same reference numerals and duplicated explanations will be omitted.

[0014] 1. Overview FIG. 1 is a diagram for explaining an overview of a vehicle recognition device 1 according to an embodiment. The vehicle recognition device 1 is a device that recognizes a vehicle 4 in a closed space 2 by using a camera 5 installed in the closed space 2. As shown in FIG. 1(A), the vehicle recognition device 1 includes a camera 5, a first illuminance sensor 6, a second illuminance sensor 7, and an information processing device 10. The camera 5, the first illuminance sensor 6, and the information processing device 10 are all provided inside the closed space 2. The second illuminance sensor 7 is provided outside the closed space 2. The information processing device 10 may be provided outside the closed space 2.

[0015] The closed space 2 is a space with a certain amount of expanse in which vehicle movement is expected, such as a garage, parking lot, factory, etc. As shown in Fig. 1(A), the closed space 2 is provided with at least one shutter 3 connecting the inside of the closed space 2 to the outside of the closed space 2 for the entrance and exit of a vehicle 4. The shutter 3 has a structure that allows it to be opened and closed in the vertical direction.

[0016] Camera 5 captures the surrounding space in which camera 5 is installed. Space 8 captured by camera 5 (hereinafter also referred to as "camera 5 capture space") includes roads laid on the floor and walls that constitute closed space 2. Camera 5 may also have a zoom function so that, when vehicle 4 is recognized on the road in closed space 2, the space around vehicle 4 can be enlarged and captured.

[0017] The first illuminance sensor 6 is a sensor that acquires an inner illuminance indicating the illuminance in the shooting space of the camera 5. In the example shown in Fig. 1(A), the first illuminance sensor 6 is fixedly attached to an inner wall surface that constitutes the closed space 2 and is included in the shooting space of the camera 5. The first illuminance sensor 6 may be attached to the vehicle 4. In this case, the illuminance acquired from the first illuminance sensor 6 while the vehicle 4 is recognized by the camera 5 is treated as the inner illuminance.

[0018] The second illuminance sensor 7 is a sensor that acquires an external illuminance indicating the illuminance outside the closed space 2. In the example shown in FIG. 1(A), the second illuminance sensor 7 is fixedly attached to an outer wall surface that constitutes the closed space 2 and on which the shutter 3 is installed. The second illuminance sensor 7 may be attached to a vehicle that exists outside the closed space 2. In this case, for example, a vehicle that exists near the closed space 2 is identified based on the position information of the vehicle that exists outside the closed space 2. Then, the illuminance acquired by the second illuminance sensor 7 of the identified vehicle during the same time period as the time period during which the inside illuminance is acquired by the first illuminance sensor 6 is treated as the outside illuminance.

[0019] The information processing device 10 is a device that performs recognition processing of the vehicle 4 included in the camera image captured by the camera 5. The information processing device 10 is connected to each of the devices, namely, the shutter 3, the camera 5, the first illuminance sensor 6, and the second illuminance sensor 7. The information processing device 10 is connected to each of the devices by, for example, a cable. Note that, when each device is provided with a communication device, the information processing device 10 may acquire various information via each of the communication devices provided in each device.

[0020] Here, as shown in Fig. 1(B), consider a case where the shutter 3 is opened to move the vehicle 4 from inside the closed space 2 to outside the closed space 2. In this case, light 9 is incident toward the inside of the closed space 2 from an entrance connecting the inside of the closed space 2 to the outside of the closed space 2. If the reflected light of the light 9 incident inside the closed space 2 is captured by the camera 5 in a predetermined amount or more, there is a possibility that the camera image captured by the camera 5 may be overexposed.

[0021] According to the vehicle recognition device 1 according to the embodiment, when a vehicle 4 moving from inside the closed space 2 toward the outside of the closed space 2 approaches the shutter 3, the information processing device 10 issues an automatic opening command to the shutter 3. Furthermore, when the outside illuminance is higher than the inside illuminance, the information processing device 10 performs a whiteout countermeasure process for suppressing whiteout at the position of the vehicle 4 in the camera image. Furthermore, when the outside illuminance is higher than the inside illuminance and it is determined that the shutter 3 is open, the information processing device 10 performs a brightness correction process for correcting the brightness of the camera image. The determination of whether the shutter 3 is open is performed, for example, based on the opening / closing information of the shutter 3. The opening / closing information of the shutter 3 includes, for example, information on the opening / closing button (open button or close button) of the shutter 3. Details of each process will be described later.

[0022] 2. Specific examples 2-1. Example of automatic opening command FIG. 2 is an explanatory diagram showing a specific example of the vehicle recognition device 1 according to the embodiment. Specifically, FIG. 2 shows an example in which an automatic opening command is issued to the shutter 3 when the vehicle 4 approaches the shutter 3. In the example shown in FIG. 2(A), an example is shown in which the distance from the shutter 3 provided at the entrance of the vehicle 4 in the closed space 2 to the vehicle 4 is equal to or greater than the initial setting distance. In this case, since the distance from the vehicle 4 to the shutter 3 is equal to or greater than a certain distance, an automatic opening command is not issued to the shutter 3. The initial setting distance means a set distance from the vehicle 4 to the shutter 3 used as a condition for opening the shutter 3. In addition, the distance from the vehicle 4 to the shutter 3 is estimated based on, for example, the position of the front part of the vehicle 4 recognized by the camera 5. In another example, the distance is estimated based on sensor acquisition information acquired by a recognition sensor (such as an on-board camera, radar, LiDAR, etc.) mounted on the vehicle 4.

[0023] 2(B) shows an example in which the distance from the vehicle 4 to the shutter 3 is less than the initially set distance. In this case, since the distance from the vehicle 4 to the shutter 3 is not greater than a certain distance, the information processing device 10 issues an automatic opening command to the shutter 3 so that the vehicle 4 can move from inside the closed space 2 to outside the closed space 2.

[0024] Here, let us consider the cause of whiteout occurring in the camera image. Whiteout occurring in the camera image occurs when the external illuminance is higher than the internal illuminance. Therefore, the information processing device 10 judges whether the external illuminance is higher than the internal illuminance. If the judgment of whether the external illuminance is higher than the internal illuminance is performed in a state where the shutter 3 is open, the light 9 incident toward the inside of the closed space 2 and the vehicle 4 overlap, since the vehicle 4 is located close to the shutter 3, and whiteout may occur at the position of the vehicle 4 in the camera image. Therefore, the judgment is performed before the shutter 3 is opened. The judgment may be performed, for example, before an automatic opening command is issued to the shutter 3, or may be performed after an automatic opening command is issued to the shutter 3 and before the shutter 3 is opened.

[0025] Even if it is determined that the outer illuminance is higher than the inner illuminance, if the difference between the outer illuminance and the inner illuminance is less than a predetermined value, it can be said that there is a low possibility that whiteout will occur in the camera image. Therefore, the information processing device 10 may perform whiteout countermeasure processing at the position of the vehicle 4 in the camera image when the outer illuminance is higher than the inner illuminance and the difference between the outer illuminance and the inner illuminance is equal to or greater than a predetermined value.

[0026] 2-2. Example of overexposure prevention processing 3 shows an example of a process for preventing whiteout of a camera image executed by the information processing device 10. The whiteout prevention process includes one of a process for extending an initial setting distance used as a condition for issuing an automatic opening command to the shutter 3, and a process for sending a deceleration command to the vehicle 4. Each process will be described in detail below.

[0027] The process of expanding the initial setting distance will be described. This process is performed based on the estimated arrival time until the vehicle 4 reaches the shutter 3. Specifically, as shown in FIG. 3(A), when the estimated arrival time until the vehicle 4 reaches the shutter 3 is equal to or greater than a threshold, that is, when the estimated arrival time is long, an automatic opening command is issued to the shutter 3 when the vehicle 4 passes the initial setting distance. In this case, when the shutter 3 opens, there is a low possibility that the light 9 incident toward the inside of the closed space 2 and the vehicle 4 will overlap. Therefore, when the estimated arrival time until the vehicle 4 reaches the shutter 3 is equal to or greater than a threshold, it is not necessary to change the initial setting distance.

[0028] On the other hand, as shown in FIG. 3(B), a case is considered in which the estimated arrival time until the vehicle 4 reaches the shutter 3 is less than the threshold, that is, the estimated arrival time is short. In this case, when the vehicle 4 passes the initial setting distance, an automatic opening command is issued to the shutter 3, and the shutter 3 opens, the light 9 incident toward the inside of the closed space 2 and the vehicle 4 are likely to overlap. Therefore, there is a risk of the recognition accuracy of the vehicle 4 being reduced. For this reason, when the estimated arrival time until the vehicle 4 reaches the shutter 3 is less than the threshold, a change is made to extend the initial setting distance as shown in FIG. 3(B). As a result, the distance from the vehicle 4 to the shutter 3 becomes longer than the initial setting distance, and it is possible to prevent the vehicle 4 from being covered by the light 9 when the shutter 3 opens. Therefore, it is possible to suppress the occurrence of whiteout at the position of the vehicle 4 in the camera image. Note that, when the estimated arrival time is less than the threshold, the information processing device 10 may extend the initial setting distance based on the distance set according to the estimated arrival time.

[0029] Next, a process of transmitting a deceleration command to the vehicle 4 will be described. This process is performed based on the estimated arrival time until the vehicle 4 reaches the shutter 3, similar to the process of expanding the initial setting distance described above. Specifically, as shown in FIG. 3(C), when the estimated arrival time until the vehicle 4 reaches the shutter 3 is less than a threshold, that is, when the estimated arrival time is short, the initial setting distance is not changed and a deceleration command to the vehicle 4 is transmitted by the information processing device 10 before the vehicle 4 passes the initial setting distance. This lengthens the estimated arrival time until the vehicle 4 reaches the shutter 3, and the vehicle 4 can be prevented from being covered by the light 9 when the shutter 3 opens. Therefore, it is possible to suppress the occurrence of whiteout at the position of the vehicle 4 in the camera image. In addition, when the estimated arrival time is less than a threshold, the information processing device 10 may generate a command to decelerate the vehicle 4 based on the deceleration amount of the vehicle 4 set according to the estimated arrival time.

[0030] 2-3. Example of brightness correction processing The brightness correction process includes at least one of parameter adjustment of the camera 5 and correction of the camera image. The parameter adjustment of the camera 5 is performed based on setting information of the camera 5. The setting information of the camera 5 includes information on the shutter speed of the camera 5. In other words, the parameter adjustment of the camera 5 means adjusting the shutter speed of the camera 5. For example, in a situation where whiteout may occur in the camera image, the setting value of the shutter speed of the camera 5 is adjusted so as to increase the shutter speed of the camera 5. This makes it possible to suppress the amount of reflected light of the light 9 captured by the camera 5. Therefore, the situation where whiteout may occur in the camera image is improved.

[0031] Next, the correction of the camera image will be described. The correction of the camera image is performed based on the brightness calculated from the camera image. In other words, when the brightness calculated from the camera image is equal to or greater than a reference value, it can be said that the camera image is in a situation where blown-out highlights may occur. Therefore, by correcting the brightness of the camera image so that the brightness is less than the reference value, the situation where blown-out highlights may occur in the camera image is improved.

[0032] 3. Configuration example 4 is a block diagram showing an example of the configuration of the information processing device 10 in the vehicle recognition device 1 according to the embodiment. The information processing device 10 includes a processor 20 and a storage device 30. The processor 20 executes various processes such as a process of recognizing a vehicle 4 included in a camera image and a process of opening and closing a shutter 3 provided at an entrance and exit of the vehicle 4 in the closed space 2. The storage device 30 stores various information required for the processes by the processor 20.

[0033] The various information stored in the storage device 30 includes camera image information, opening / closing information of the shutter 3, information on inner illuminance, information on outer illuminance, and setting information of the camera 5. The opening / closing information of the shutter 3 includes information on the opening / closing button of the shutter 3. Furthermore, the setting information of the camera 5 includes information on the shutter speed of the camera 5.

[0034] The storage device 30 further includes a vehicle recognition program (not shown). The vehicle recognition program is a computer program executed by the processor 20. When the processor 20 executes the vehicle recognition program, the functions of the vehicle recognition device 1 are realized.

[0035] The connection between the various sensors including the camera 5, the first illuminance sensor 6, and the second illuminance sensor 7 and the information processing device 10 may be a cable (first connection) or communication via a communication device provided in the various sensors (second connection). When the connection between the various sensors and the information processing device 10 is the second connection, the information processing device 10 further includes a communication device capable of communicating with each sensor.

[0036] 4. Processing example FIG. 5 is a flowchart showing an example of processing by the information processing device 10 in the vehicle recognition device 1 according to the embodiment.

[0037] In step S100, the information processing device 10 acquires various information. After that, the process proceeds to step S110. The various information includes camera image information, shutter 3 open / close information, inner illuminance information, outer illuminance information, and camera 5 setting information.

[0038] In step S110, the information processing device 10 determines whether the outer illuminance is higher than the inner illuminance based on the information on the outer illuminance and the information on the inner illuminance. If it is determined that the outer illuminance is higher than the inner illuminance (step S110; Yes), the process proceeds to step S120. Otherwise (step S110; No), the process ends.

[0039] In step S120, the information processing device 10 executes a whiteout countermeasure process. After that, the process proceeds to step S130. As described above, the whiteout countermeasure process includes one of the process of extending the initial setting distance and the process of sending a deceleration command to the vehicle 4.

[0040] In step S130, the information processing device 10 judges whether or not the distance from the shutter 3 to the vehicle 4 is equal to or shorter than the initial setting distance. If it is judged that the distance from the shutter 3 to the vehicle 4 is equal to or shorter than the initial setting distance (step S130; Yes), the process proceeds to step S140. Otherwise (step S130; No), the process returns to step S130. Note that the initial setting distance used in the judgment in this step is based on the expanded set distance when the process of expanding the initial setting distance is executed in the above-mentioned step S120.

[0041] In step S140, the information processing device 10 transmits an automatic opening command to the shutter 3. After that, the process proceeds to step S150.

[0042] In step S150, the information processing device 10 determines whether the shutter 3 is open or not based on the opening / closing information of the shutter 3. If it is determined that the shutter 3 is open (step S150; Yes), the process proceeds to step S160. Otherwise (step S150; No), the process returns to step S150.

[0043] In step S160, the information processing device 10 executes a brightness correction process for the camera image. As described above, the brightness correction process includes at least one of a process for adjusting the shutter speed of the camera 5 and a process for correcting the brightness calculated from the camera image.

[0044] 5.Effects In this way, in the vehicle recognition device 1 according to the embodiment, when the external illuminance is higher than the internal illuminance, that is, when a situation is present in which whiteout may occur in the camera image, a whiteout countermeasure process is executed to prevent whiteout from occurring at the position of the vehicle 4 in the camera image. In the whiteout countermeasure process, one of a process of extending the initial setting distance used as a condition for issuing an automatic opening command to the shutter 3 and a process of sending a deceleration command to the vehicle 4 is executed. This makes it possible to prevent a decrease in the accuracy of recognition of the vehicle 4 based on the camera image including the vehicle 4 when the shutter 3 opens. [Explanation of symbols]

[0045] REFERENCE SIGNS LIST 1...vehicle recognition device, 2...closed space, 3...shutter, 4...vehicle, 5...camera, 6...first illuminance sensor, 7...second illuminance sensor, 8...shooting space, 9...light, 10...information processing device, 20...processor, 30...storage device

Claims

1. A vehicle recognition device that recognizes vehicles present in a closed space, A storage device that stores a camera image acquired by a camera installed in the closed space, information on inside illuminance indicating the illuminance inside the closed space and in the space photographed by the camera, and information on outside illuminance indicating the illuminance outside the closed space; A processor that performs a process of recognizing a vehicle included in the camera image and a process of opening and closing a shutter provided at an entrance and exit of the vehicle in the closed space; Equipped with The processor performs a process of transmitting an automatic opening command to the shutter when a distance from the shutter to the vehicle becomes equal to or shorter than an initial setting distance in the opening / closing process; The processor further comprises: determining whether the outer illuminance is higher than the inner illuminance based on the information on the outer illuminance and the information on the inner illuminance; When it is determined that the outer illuminance is higher than the inner illuminance, a whiteout countermeasure process is performed to suppress whiteout at the position of the vehicle in the camera image. A vehicle recognition device comprising:

2. The vehicle recognition device according to claim 1, The storage device further includes information on an estimated arrival time for the vehicle to arrive at the shutter, The processor further includes, in the whiteout countermeasure processing, When the estimated time of arrival is less than a threshold, the initially set distance is extended based on a distance set according to the estimated time of arrival. A vehicle recognition device comprising:

3. The vehicle recognition device according to claim 1, The storage device further includes information on an estimated arrival time for the vehicle to arrive at the shutter, The processor further includes, in the whiteout countermeasure processing, and generating a command to decelerate the vehicle based on a deceleration amount of the vehicle set according to the estimated arrival time when the estimated arrival time is less than a threshold value. A vehicle recognition device comprising:

4. The vehicle recognition device according to claim 1, The storage device further includes information on opening and closing of the shutter, The processor further comprises: determining whether the shutter is open or closed based on the opening / closing information; When it is determined that the shutter is open and the outer illuminance is higher than the inner illuminance, a brightness correction process is performed to correct the brightness of the camera image. A vehicle recognition device comprising:

5. A vehicle recognition method for recognizing a vehicle present in an enclosed space, comprising: Performing a process of recognizing a vehicle included in a camera image acquired by a camera installed in the closed space, and a process of opening and closing a shutter provided at an entrance and exit of the vehicle in the closed space; In the opening and closing process, when a distance from the shutter to the vehicle becomes equal to or shorter than an initial setting distance, an automatic opening command is sent to the shutter; Determining whether the outside illuminance is higher than the inside illuminance based on information on outside illuminance indicating the illuminance outside the closed space and information on inside illuminance indicating the illuminance inside the closed space and in a shooting space of the camera; When it is determined that the outer illuminance is higher than the inner illuminance, performing a whiteout countermeasure process to suppress whiteout at the position of the vehicle in the camera image. Includes A vehicle recognition method comprising:

Images