Information processing device and information processing program

Abstract

To provide an information processing device and an information processing program capable of improving accuracy in determining an object present in a dead angle area.SOLUTION: When entry of a detection object into a detection area set in a video photographed by a camera is detected by image recognition, an information processing device imparts a first ID to the detection object which has entered the detection area. When a movement of the detection object from the detection area to a dead angle area set on a lower side of the detection area is detected by image recognition, the information processing device imparts a second ID to the detection object to which the first ID has been imparted while keeping the first ID. When a re-entry of the detection object to which the second ID has been imparted from the dead angle area to the detection area is detected by image recognition, the information processing device deletes the second ID while keeping the first ID of the detection object to which the second ID has been imparted.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The disclosed embodiments relate to an information processing device and an information processing program. [Background technology]

[0002] A monitoring device that monitors the periphery of a vehicle outputs an alarm when a sensor, such as a camera, detects that a target object, such as a pedestrian or another vehicle, within the detection range of the sensor approaches the vehicle. The sensor has a blind spot outside its detection range where it cannot detect the target object.

[0003] It is desirable for the monitoring device to also output an alarm if there is an object in such a blind spot that requires attention. For this reason, there is a technology that determines that an object is present in the blind spot when the object crosses the detection range of the sensor and moves into the blind spot outside the detection range (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-15557 Summary of the Invention [Problem to be solved by the invention]

[0005] However, even if conventional monitoring devices can detect that an object has moved from the detection area into a blind spot area, they do not take into account the object's subsequent behavior, and therefore cannot accurately determine whether an object that has moved into the blind spot area continues to be present in the blind spot area.

[0006] One aspect of the embodiment has been made in consideration of the above, and aims to provide an information processing device and an information processing program that can improve the accuracy of determining objects present in blind spot areas. [Means for solving the problem]

[0007] An information processing device according to one aspect of the embodiment, when detecting by image recognition that a detection object has entered a detection area set for an image captured by a camera, assigns a first ID to the detection object that has entered the detection area. When detecting by image recognition that the detection object has moved from the detection area into a blind spot area set below the detection area, the information processing device assigns a second ID to the detection object that has been assigned the first ID while retaining the first ID. When detecting by image recognition that the detection object that has been assigned the second ID has re-entered the detection area from the blind spot area, the information processing device deletes the second ID from the detection object that has been assigned the second ID while retaining the first ID. [Effects of the Invention]

[0008] According to an embodiment, an information processing device and an information processing program manage the entry and exit of a detection object into a blind spot area by an ID assigned to the detection object, and determine the presence or absence of a detection object in the blind spot area based on whether or not a detection object assigned a second ID is present. This allows the information processing device and the information processing program to improve the accuracy of determining an object present in the blind spot area. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is an explanatory diagram of a method for monitoring a detection target according to an embodiment. [Figure 2] FIG. 2 is an explanatory diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 3] FIG. 3 is an explanatory diagram illustrating an example of detection target information according to the embodiment. [Figure 4] FIG. 4 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 5] FIG. 5 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 6]FIG. 6 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 7] FIG. 7 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 8] FIG. 8 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 9] FIG. 9 is an explanatory diagram illustrating an example of the operation of the information processing device according to the embodiment. [Figure 10] FIG. 10 is a flowchart illustrating an example of processing executed by the controller according to the embodiment. [Figure 11] FIG. 11 is a flowchart illustrating an example of processing executed by the controller according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an information processing device and an information processing program according to an embodiment will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to the following embodiments.

[0011] [1. Overview of the method for monitoring detection targets] An overview of a method for monitoring a detection target performed by an information processing device according to an embodiment will be described with reference to Fig. 1. Fig. 1 is an explanatory diagram of a method for monitoring a detection target according to an embodiment.

[0012] The information processing device according to the embodiment is a device that detects and monitors objects in front of the vehicle that require attention by performing image recognition on an image captured by a camera that captures the view ahead of the vehicle through the windshield of the vehicle. The detected objects include, for example, other vehicles, people, animals, and fallen objects.

[0013] 1, the information processing device according to the embodiment sets a detection area A1 in which a detection target P, such as a person, can be detected by image recognition in a captured image A of a camera. Furthermore, the information processing device sets a blind spot A2 located below the detection area A1 in the captured image A, in which the detection target P cannot be detected by image recognition.

[0014] The blind spot area A2 is an area in which, for example, the hood of the vehicle is captured. In other words, the blind spot area A2 is an area in front of the vehicle that is blocked by the hood and where the detection object P cannot be captured by the camera installed in the windshield.

[0015] The information processing device according to the embodiment detects a detection target P entering a detection area A1 by image recognition, and acquires feature information indicating the features of the detection target P. Furthermore, the information processing device assigns a unique first ID to the detection target P and stores the first ID in association with the feature information.

[0016] Thereafter, when the information processing device detects through image recognition that the object to be detected P has moved from the detection area A1 to the blind spot area A2, it assigns to the object to be detected P a second ID indicating that the object to be detected P is located within the blind spot area A2 in addition to the first ID.

[0017] Thereafter, when the information processing device detects by image recognition that the detection object P has re-entered the detection area A1, it deletes the second ID while retaining the first ID that was assigned to the detection object P. In this way, the information processing device assigns a first ID that allows each of the detection objects P to be detected, which allows each to be individually identified, and further assigns a second ID to a detection object P that is present in the blind spot area A2, thereby monitoring the movement of the detection object P.

[0018] As a result, if a detection object P assigned a second ID is present, the information processing device can determine that the detection object P is present in the blind spot area A2. Furthermore, since the information processing device stores feature information of the detected detection object P in association with the second ID, it can accurately recognize which detection object P is currently present in the blind spot area A2. Therefore, the information processing device according to the embodiment can improve the accuracy of determining an object present in the blind spot area A2.

[0019] 2. Configuration of information processing device 1 Next, the configuration of the information processing device 1 according to the embodiment will be described with reference to Fig. 2 and Fig. 3. Fig. 2 is an explanatory diagram showing an example of the configuration of the information processing device 1 according to the embodiment. Fig. 3 is an explanatory diagram showing an example of detection target object information 21 according to the embodiment.

[0020] As shown in Fig. 2, the information processing device 1 is connected to a camera 11, an alarm device 12, and a microphone 13. The camera 11 is an in-vehicle camera attached to the windshield or rearview mirror of the vehicle to capture images of the area ahead of the vehicle. The camera 11 may also be a camera 11 built into a drive recorder that captures images of the area around the vehicle. The camera 11 transmits a captured image A to the information processing device 1.

[0021] The alarm device 12 is a device that outputs an alarm indicating that a detection target P has been detected by the information processing device 1. The alarm device 12 includes a speaker and a display device. The information processing device 1 outputs the alarm by at least one of outputting an alarm sound, turning on and off an alarm lamp indicating the alarm, and displaying text indicating the object of the alarm.

[0022] The microphone 13 is installed, for example, on the front grille or front bumper of the vehicle. The microphone 13 is a directional microphone whose sound collection direction is directed toward the blind spot area A2. The microphone 13 collects sounds generated from a detection target P present in the blind spot area A2 and transmits the collected sounds to the information processing device 1.

[0023] The information processing device 1 includes a storage unit 2 and a controller 3. The storage unit 2 is, for example, an information storage device such as a data flash, and stores detection target object information 21 and user information 22. The detection target object information 21 is, for example, information in which a first ID, a second ID, and feature information are associated with each detection target object P detected by the information processing device 1, as shown in FIG.

[0024] For example, if the detection target P is an adult, the feature information includes information indicating that the detection target P is an adult, information about the person's face, information about the person's clothing, etc. Furthermore, if the detection target P is a child, the feature information includes information indicating that the detection target P is a child, information about the child's face, information about the child's clothing, etc.

[0025] If the detected object P is another vehicle, the feature information includes information such as the shape and color of the other vehicle. If the detected object P is an animal or a fallen object, the feature information includes information such as the shape and color of the animal or the fallen object. The user information 22 includes feature information of the user of the vehicle. The user feature information includes information indicating the user's facial features, etc.

[0026] The controller 3 includes a microcomputer having a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc., and various circuits. The controller 3 performs control such as detection of the detection target P, monitoring, and output of an alarm by the CPU executing an information processing program stored in the ROM using the RAM as a working area. The information processing program may be stored in a storage device from the outside via a communication line, etc.

[0027] The controller 3 may be partially or entirely configured with hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0028] The controller 3 includes an image recognition AI (Artificial Intelligence) that has been trained to perform image recognition of the detection target P from the captured video A received from the camera 11. The controller 3 detects and monitors the detection target P based on the captured video A received from the camera 11.

[0029] Furthermore, the controller 3 can also detect and monitor the detection target P based on the sound received from the microphone 13. When the controller 3 detects the detection target P, it causes the alarm device 12 to output an alarm in a manner that varies depending on the position of the detection target P relative to the vehicle. Specifically, the controller 3 realizes or executes the information processing actions described below.

[0030] 3. Example of operation of information processing device 1 The operation of the information processing device 1 according to the embodiment will be described with reference to Fig. 4 to Fig. 9. Fig. 4 to Fig. 9 are explanatory diagrams showing an example of the operation of the information processing device 1 according to the embodiment.

[0031] As shown in Figure 4, when the controller 3 detects through image recognition that a detection object P (here, an adult) has entered the detection area A1 from outside the detection area A1, it assigns a first ID "101" to the detection object P that has entered the detection area A1.

[0032] At this time, the controller 3 acquires characteristic information of the detection object P, associates it with the first ID, and stores it in the memory unit 2 as detection object information 21. Also, at this point, the detection object P has not moved into the blind spot area A2. Therefore, the controller 3 does not assign a second ID to the detection object P, that is, sets the second ID to "none."

[0033] As a result, while the detection target P is present in the detection area A1, the controller 3 can monitor the detection target P using the first ID of the detection target P and the feature information of the detection target P. In other words, the controller 3 can track the movement of the detection target P in the detection area A1 by continuing to recognize the image of the detection target P within the detection area A1.

[0034] When the controller 3 detects the entry of a detection target P into the detection area A1, it causes the alarm device 12 to output an alarm in a first manner. Specifically, the controller 3 causes the alarm to be output in the first manner when a detection target P that has been assigned a first ID but has not been assigned a second ID is present. In the first manner, for example, the alarm is output by displaying text or by flashing an alarm lamp without using sound.

[0035] Furthermore, when the controller 3 detects a detection object P in the detection area A1, the controller 3 derives the distance from the detection object P to the host vehicle by image recognition. Then, the controller 3 changes the content of the warning according to the distance from the host vehicle to the detection object P.

[0036] For example, the controller 3 shortens the blinking interval of the warning lamp as the distance between the detection object P and the vehicle decreases. Then, when the detection object P approaches immediately before the blind spot area A2, the controller 3 switches the light emission of the warning lamp from blinking to lighting. This allows the controller 3 to intuitively notify the user of the distance from the vehicle to the detection object P by the light of the warning lamp.

[0037] Thereafter, when the detected object P moves outside the detection area A1 from either the top end, right end, or left end of the detection area A1, the controller 3 cancels the alarm in the first mode because the detected object P no longer interferes with the vehicle's travel.

[0038] Then, the controller 3 deletes the first ID that has been assigned to the detection target P. That is, the controller 3 deletes the first ID when the detection target P leaves the detection area A1 without moving into the blind spot area A2.

[0039] For example, as shown in Fig. 5, when the distance between the detection object P and the vehicle increases and the detection object P moves from the upper end of the detection area A1 to the outside of the detection area A1, the controller 3 deletes the first ID "101" of the detection object P. As a result, both the first ID and the second ID of the detection object P become "none." In this case, when the detection object P re-enters the detection area A1, the controller 3 recognizes the detection object P by image recognition and assigns a new first ID to the detection object P.

[0040] Furthermore, there is a case where the detection object P in the detection area A1 approaches the vehicle and enters the blind spot area A2. In this case, when the controller 3 detects by image recognition that the detection object P has moved from the detection area A1 to the blind spot area A2, the controller 3 assigns a second ID to the detection object P that has been assigned the first ID while keeping the first ID.

[0041] 6, when a detection target P assigned with a first ID "101" moves into the blind spot area A2, the controller 3 assigns a second ID "201" to the detection target P while retaining the first ID "101." This allows the controller 3 to accurately determine that the detection target P is present in the blind spot area A2 if the detection target P assigned with the second ID is present.

[0042] In this case, when the controller 3 detects the advance of the host vehicle while the detection object P is present in the blind spot area A2, it outputs an alarm in a second manner different from the first manner. Specifically, when the controller 3 detects by image recognition that the detection object P assigned the first ID and the second ID is present and that the host vehicle has moved in the direction of the detection object P assigned the first ID and the second ID, it outputs an alarm in the second manner.

[0043] Alternatively, when the controller 3 detects that a detection object P assigned the first ID and the second ID is present and that the gear position of the vehicle has been switched to D range, it outputs an alarm in the second manner.

[0044] For example, in the second mode of warning, the controller 3 causes the warning device 12 to output a warning sound such as a warning tone or an alarm tone, or a warning message such as "There is an obstacle in the blind spot ahead. Please stop the vehicle."

[0045] As a result, the controller 3 can alert the driver in the first manner when a detected object P exists in the detection area A1, thereby drawing the driver's attention. Furthermore, when a detected object P exists in the blind spot area A2 and the vehicle moves in that direction, the controller 3 alerts the driver in the second manner, thereby drawing the driver's attention in truly dangerous situations.

[0046] Furthermore, the detection object P in the blind spot area A2 may re-enter the detection area A1 from the blind spot area A2. In this case, when the controller 3 detects by image recognition that the detection object P assigned the second ID has re-entered the detection area A1 from the blind spot area A2, the controller 3 deletes the second ID while retaining the first ID of the detection object P assigned the second ID.

[0047] For example, as shown in FIG. 7, when a detection object P assigned a first ID "101" and a second ID "201" re-enters the detection area A1, the controller 3 deletes the second ID "201" while retaining the first ID "101" of the detection object P.

[0048] Specifically, when a detection object Q enters the detection area A1 from the blind spot area A2, the detection object Q may be a detection object P with the first ID "101" that has re-entered the detection area A1 from the blind spot area A2, or it may be a completely different detection object that has entered the blind spot area A2 from the left or right of the blind spot area A2 and then entered the detection area A1.

[0049] Therefore, the controller 3 detects the characteristics of the detection object Q that has entered the detection area A1 from the blind spot area A2 by image recognition. If the characteristics are the same as the characteristics of the detection object P to which the first ID "101" has been assigned (stored in the storage unit 2 as the detection object information 21), the controller 3 determines that the detection object Q is the detection object P to which the first ID "101" has been assigned. Then, the controller 3 treats the detection object Q as the detection object P, and deletes the second ID "201" of the detection object P while retaining the first ID "101" of the detection object P.

[0050] On the other hand, if the characteristics of the detection object Q differ from the characteristics of the detection object P stored in the memory unit 2, the controller 3 assigns a new first ID (for example, "102") to the detection object Q. In this case, the detection object P to which the first ID "101" has been assigned still has the second ID "201" and is still staying in the blind spot area A2.

[0051] In this way, when the object to be detected P re-enters the detection area A1 from the blind spot area A2, the controller 3 can accurately determine that the object to be detected P is not present in the blind spot area A2, since the object to be detected P with the second ID is no longer present.

[0052] Furthermore, when a detection object Q different from the detection object P enters the detection area A1 from the blind spot area A2, the controller 3 determines that the detection object P assigned the second ID is still present in the blind spot area A2, and therefore can accurately determine that the detection object P is present in the blind spot area A2.

[0053] Note that the detected object P in the blind spot area A2 may exit the blind spot area A2 without re-entering the detection area A1. For example, the detected object P may exit the blind spot area A2 from the right or left end of the blind spot area A2.

[0054] In this case, the controller 3 does not delete the second ID. Therefore, because the detection object P assigned the second ID is present, the controller 3 may erroneously determine that the detection object P is present in the blind spot area A2, even though the detection object P is not actually present in the blind spot area A2.

[0055] Therefore, when the detection object P moves from the detection area A1 to the blind spot area A2, the controller 3 requests the user of the vehicle to visually check the blind spot area A2. In other words, when the detection object P assigned the first ID and the second ID is present, the controller 3 requests the user of the vehicle to visually check the blind spot area A2.

[0056] At this time, the controller 3 outputs the warning in the second mode. For this reason, the controller 3 causes the warning device 12 to output a voice message or text message stating, for example, as shown in Fig. 8, "There is an obstacle in the blind spot area ahead! Please visually check the blind spot area. The warning will be canceled upon visual check." This allows the controller 3 to prompt the user to visually check the blind spot area A2.

[0057] After that, for example, if the user of the vehicle visually checks the blind spot area A2 in response to a request, the user will request that the person move out of the blind spot area A2 if the detected object P is a person. Also, if the detected object P is an animal or a fallen object, the user will take action to remove the detected object P from the blind spot area A2.

[0058] 9, when the controller 3 detects by image recognition that the user U of the host vehicle has entered the detection area A1, that is, when the controller 3 detects that the user U has visually confirmed the blind spot area A2, the controller 3 cancels the warning in the second mode. As a result, the controller 3 can cancel the warning in the second mode after the detection object P is no longer present in the blind spot area A2 and there is no longer any risk of a collision between the detection object P and the host vehicle.

[0059] In addition, when the controller 3 detects through image recognition that the user U of the vehicle has entered the detection area A1, that is, that the user U has visually confirmed the blind spot area A2, the controller 3 deletes the first ID and the second ID of the detection object P to which the first ID and the second ID have been assigned.

[0060] For example, as shown in Fig. 9, when the detection object P moves out of the blind spot area A2 from the right end of the blind spot area A2, the second ID of the detection object P is deleted. As a result, both the first ID and the second ID of the detection object P become "none."

[0061] As a result, the detection object P assigned the second ID no longer exists. Therefore, the controller 3 can prevent the controller 3 from continuing to hold an unnecessary second ID even though the detection object P does not exist in the blind spot area A2. Furthermore, the controller 3 can prevent the controller 3 from erroneously determining that a detection object P that does not exist in the blind spot area A2 exists because the controller 3 has held an unnecessary second ID.

[0062] [4. Processes performed by Controller 3] Next, a process executed by the controller 3 according to the embodiment will be described with reference to Fig. 10 and Fig. 11. Fig. 10 and Fig. 11 are flowcharts showing an example of a process executed by the controller 3 according to the embodiment. The controller 3 repeatedly executes the process shown in Fig. 10 while the information processing device 1 is powered on.

[0063] When the information processing device 1 is powered on, the controller 3 determines whether or not a detection target P has entered the detection area A1 by image recognition (step S101), as shown in Fig. 10. If the controller 3 determines that the detection target P has not entered the detection area A1 (step S101, No), it repeats the determination of step S101.

[0064] Then, when the controller 3 determines that the detection object P has entered the detection area A1 (step S101, Yes), it assigns a first ID to the detection object P that the detection object P has entered (step S102) and outputs an alarm in a first manner (step S103).

[0065] Next, the controller 3 determines whether the detection target P has moved outside the detection area A1 by image recognition (step S104). If the controller 3 determines that the detection target P has not moved outside the detection area A1 (step S104, No), that is, if the controller 3 determines that the detection target P continues to exist in the detection area A1, the controller 3 repeats the determination process of step S104.

[0066] If the controller 3 determines that the detection object P has moved out of the detection area A1 (step S104, Yes), the controller 3 cancels the alarm in the first mode (step S105). Next, the controller 3 determines by image recognition whether the detection object P has moved into the blind spot area A2 (step S106).

[0067] When the controller 3 determines that the detection object P has not moved into the blind spot area A2 (step S106, No), that is, when the detection object P has moved out of the detection area A1 without entering the blind spot area A2, the controller 3 deletes the first ID of the detection object P (step S108). After that, the controller 3 ends the process and starts the process again from step S101.

[0068] Furthermore, if the controller 3 determines that the object to be detected P has moved into the blind spot area A2 (step S106, Yes), it executes the blind spot area monitoring process (step S107), then ends the process and starts the process again from step S101.

[0069] Next, a specific example of the blind spot area monitoring process will be described with reference to Fig. 11. As shown in Fig. 11, when the blind spot area monitoring process is started, the controller 3 assigns a second ID to the detection target object P while keeping the first ID (step S201).

[0070] Next, the controller 3 determines whether the host vehicle has moved forward by image recognition (202). If the controller 3 determines that the host vehicle has not moved forward (No in step S202), the controller 3 proceeds to step S208.

[0071] Furthermore, when the controller 3 determines that the host vehicle has moved forward (step S202, Yes), it outputs an alarm in a second manner (step S203) and requests the user U to visually check the blind spot area A2 (step S204). Next, the controller 3 determines whether the user U has entered the detection area A1 by image recognition (step S205).

[0072] If the controller 3 determines that the user U has not entered the detection area A1 (step S205, No), it repeats the determination process of step S205. If the controller 3 determines that the user U has entered the detection area A1 (step S205, Yes), it cancels the alarm in the second mode (step S206) and deletes the first ID and the second ID of the detection object P (step S207).

[0073] Next, the controller 3 determines whether the detection object P has re-entered the detection area A1 by image recognition (step S208). If the controller 3 determines that the detection object P has not re-entered the detection area A1 (step S208, No), the process proceeds to step S202.

[0074] Furthermore, if the controller 3 determines that the object to be detected P has re-entered the detection area A1 (step S208, Yes), it deletes the second ID of the object to be detected P while retaining the first ID (step S209), and terminates the blind spot area monitoring process.

[0075] In the above embodiment, the controller 3 determines whether or not there is a detection target P in the blind spot area A2 based on the presence or absence of the second ID, but this is just one example. The controller 3 may be configured to estimate the detection target P in the blind spot area A2 based on the sound in the blind spot area A2 received from the microphone 13 in addition to the presence or absence of the second ID.

[0076] In this case, the controller 3 is equipped with a voice recognition AI that has been trained to estimate the source of a voice from the voice, thereby enabling the controller 3 to more accurately estimate the detection target P present in the blind spot area A2.

[0077] In the above embodiment, the case where the information processing device 1 detects and monitors the detection object P present in front of the vehicle has been described, but this is just one example. The information processing device 1 can improve the accuracy of determining the detection object P present in the blind spot area on the side or rear of the vehicle by performing the same processing as in the above embodiment on the video input from a camera that captures the side or rear of the vehicle.

[0078] Further advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents. [Explanation of symbols]

[0079] 1. Information processing equipment 2 Storage section 3 Controller 11 Camera 12 Alarm device 13. Mike 21 Information on detected objects 22 User Information A Footage A1 detection area A2 Blind spot area P, Q Detection target U User

Claims

1. When detecting by image recognition that a detection object has entered a detection area set for an image captured by a camera, assigning a first ID to the detection object that has entered the detection area; When it is detected by image recognition that the detection object has moved from the detection area to a blind spot area set below the detection area, a second ID is assigned to the detection object to which the first ID has been assigned while the first ID is maintained; When it is detected by image recognition that the detection object to which the second ID has been assigned has re-entered the detection area from the blind spot area, the second ID is deleted while the first ID of the detection object to which the second ID has been assigned is maintained. Information processing device.

2. When it is detected by image recognition that a user of the vehicle equipped with the camera has entered the detection area, the first ID and the second ID of the detection object to which the first ID and the second ID have been assigned are deleted. The information processing device according to claim 1 .

3. outputting an alarm in a first manner when the object to be detected is assigned the first ID and is not assigned the second ID; When the detection object assigned the first ID and the second ID is present and the vehicle equipped with the camera detects movement toward the detection object assigned the first ID and the second ID, an alarm is output in a second manner different from the first manner. The information processing device according to claim 1 .

4. When it is detected by image recognition that a user of a vehicle equipped with the camera has entered the detection area, the warning in the second mode is cancelled. The information processing device according to claim 3 .

5. When the detection object assigned the first ID and the second ID is present, a user of the vehicle equipped with the camera is requested to visually check the blind spot area. The information processing device according to claim 1 .

6. A detection target present in the blind spot area is estimated based on audio information acquired by a directional microphone that acquires audio emitted from the blind spot area. The information processing device according to claim 1 .

7. a step of assigning a first ID to the detection object that has entered a detection area set for an image captured by a camera when the detection object has entered the detection area by image recognition; a step of assigning a second ID to the detection object to which the first ID has been assigned while maintaining the first ID, when it is detected by image recognition that the detection object has moved from the detection area to a blind spot area set below the detection area; When it is detected by image recognition that the detection object to which the second ID has been assigned has re-entered the detection area from the blind spot area, deleting the second ID while retaining the first ID of the detection object to which the second ID has been assigned; An information processing program that causes a computer to execute the above.

Images