JP6343769B2 - Lost and found prevention system, information provision method for passenger car crew, computer program - Google Patents

Description

  The present invention relates to a lost article prevention system that notifies passengers of passenger cars such as taxis and hires that there is a possibility that passengers are left behind.

In passenger cars, various mechanisms have been proposed in the past in order to prevent things left behind by passengers. For example, in Patent Document 1, the inside of a vehicle is imaged when the door is opened when the vehicle is empty and when the door is closed after payment, and the captured vehicle image is compared with a reference image. There has been proposed an in-vehicle left-behind prevention device that detects presence or absence. The reference image is an image when the passenger gets on. When it is detected that there is something left behind, it notifies the passenger or crew that there is something left behind.
Further, Patent Document 2 stores a list of articles that should be detected as forgotten items, discriminates the articles in the vehicle when a passenger gets off, and if the identified articles are included in the list, the articles are forgotten. An in-vehicle device that detects the above has been proposed. The discrimination of the article is performed based on the difference between the images taken by the imaging device before and after the passenger is boarded.

JP 2006-338535 A JP 2012-123497 A

The devices disclosed in Patent Documents 1 and 2 both capture the interior of a vehicle with an imaging device and detect forgotten things from the captured images. In general, the method of detecting forgotten things from an image captured by an imaging device is superior to the method of detecting forgotten things by a sensor or the like in terms of cost. However, since the inside of the vehicle is imaged by the imaging device, the passenger's face may be imaged. Imaging in a form that can identify a passenger is not preferable from the viewpoint of passenger privacy protection. Depending on the passenger, it is not preferable to take an image, and it is possible to have a sense of resistance.
Further, in imaging with visible light, it becomes difficult to detect forgotten objects in a shadow of some object as seen from the imaging apparatus or in a dark part such as a trunk room. Furthermore, in the image pick-up by the conventional image pickup device, the way of reflection varies depending on the weather (clear weather, cloudy weather), the time zone (daytime, nighttime), the traveling environment (whether there is a tunnel or the like), and the angle of sunlight. For this reason, an error may occur in the object recognition process due to unevenness in the amount of light at the time of imaging, or a shadow may be erroneously recognized as an object.

  It is an object of the present invention to provide a forgotten object prevention system capable of detecting forgotten things in a passenger car without being affected by weather, time zone, or the like while protecting passenger privacy.

The forgotten object prevention system of the present invention for solving the above problems is installed in a passenger car having a rented state where passengers exist and an empty state where no passengers exist. An imaging device that images the interior of the passenger car using infrared light of a specific wavelength with high resolution, and an object reflected in the first infrared image acquired from the imaging device in the empty state, and the analysis The result is held, and after changing from the unoccupied state to the rented state, the object reflected in the second infrared image obtained from the imaging device is analyzed when the passenger gets off, and the analysis result is retained. An analysis means for comparing with the analysis result, a determination means for determining whether an object other than a passenger exists in the vehicle according to a comparison content by the analysis means, and an object other than the passenger in the vehicle When it is determined to be present, toward the crew of the passenger car, characterized in that it comprises a warning means for left behind to warn that may have occurred, the.
This forgotten object prevention system can be realized by the cooperation of a computer program and a computer that can read the computer program. Moreover, the information provision method with respect to the passenger | crew of a passenger car can be implemented with this computer.

The present invention images the interior of a passenger car using, for example, an infrared ray having a specific wavelength that has an optimal resolution in accordance with the incident angle of the sunlight, the amount of light, and the weather. For this reason, it is possible to prevent the occurrence of a situation such as misidentifying a shadow as an object or not being able to recognize an object existing in a shadow portion, which may occur in a conventional imaging device that performs imaging using visible light. Further, since the object reflected in the infrared image is analyzed, passenger privacy is not harmed.
Also, because objects and passengers that are likely to be forgotten are discriminated based on the analysis results of the objects reflected in the infrared image, the processing load is reduced compared to holding images or comparing images. Can be significantly reduced. Therefore, it is possible to quickly determine whether or not there is a possibility that the passenger is left behind.

The whole block diagram of a thing-left-behind prevention system. The functional block diagram formed in a thing-left-behind prevention system. (A), (b) is explanatory drawing of a camera arrangement | positioning and an imaging range. Explanatory drawing of the timing of the process which a thing-left-behind prevention system performs. Processing procedure explanatory drawing of the thing-left-behind prevention system performed when a passenger gets off. Explanatory drawing of the process sequence of the thing-left-behind prevention system performed when it is an empty vehicle state. Explanatory drawing of the process sequence of the thing-left-behind prevention system performed in the rental state.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<System configuration>
FIG. 1 is a configuration diagram of the forgotten object prevention system of the present embodiment. The forgotten thing prevention system 1 is mounted on a taxi that is an example of a passenger car, and notifies the taxi crew of the presence or absence of a thing left behind by the passenger.

The forgotten object prevention system 1 includes a plurality of cameras 10 as imaging devices. The plurality of cameras 10 are arranged in various places in the taxi vehicle so as not to make a blind spot as much as possible. In this embodiment, the right seat camera 11 that captures the foot of the right seat of the rear seat, the left seat camera 12 that captures the foot of the left seat of the rear seat, the overhead camera 13 that captures the rear seat from the overhead, and the interior of the trunk A trunk camera 14 is provided in the vehicle.
However, the number and arrangement of the cameras 10 are not limited to this example, and are appropriately provided in consideration of cost and accuracy of detecting forgotten items.

  The camera 10 is an infrared camera having sensitivity to infrared rays, and takes a still image with an automatic shutter. Set so that shutter sound does not come out. An infrared camera is a camera having a wavelength longer than that of visible red (having a lower frequency) and sensitivity to an electromagnetic wave having a wavelength shorter than that of radio waves, and cannot be imaged by visible light, for example, inside a trunk, a floor mat of a rear seat, Even when the interior of the vehicle is dark like night, it is possible to take an image. The infrared camera can recognize the presence of a passenger by analyzing the relative size of the subject, but does not identify the passenger's face. Therefore, passenger privacy is not harmed.

The thing-left-behind prevention system 1 forms the main function regarding the thing-left-behind detection in the control part 20 which is a kind of computer. The control unit 20 includes an interface 21, a CPU (Central Processing Unit) 22, a RAM (Random Access Memory) 23, and a ROM (Read Only Memory) 24. In addition, an HDD (Hard Disk Drive) 25 that is a large-capacity recording medium is provided.
The interface 21, CPU 22, RAM 23, ROM 24, and HDD 25 are connected to each other via a bus so as to be able to transmit and receive data.

The interface 21 acquires an infrared image captured by the camera 10. The interface 21 is connected to the taximeter 30 and acquires state data representing the state of the taximeter 30 (rental, empty car, premium, payment). The taxi meter 30 is changed in its state (rental state, empty state, premium state, payment state) by the operation of the crew, but as described later, the state is forcibly changed by the control data from the control unit 20. There is also a case.
The interface 21 is also connected to the warning unit 40. The warning unit 40 is connected to a speaker 41 and a warning lamp 42.

  The CPU 22 reads the computer program of the present invention from the HDD 25, reads the control program necessary for control from the ROM 24, and executes these as the work area by the RAM 23, thereby executing processing related to passenger left behind detection.

FIG. 2 is a functional block diagram showing functions formed in the control unit 20 when the CPU 22 executes a computer program.
The control unit 20 functions as the image management unit 101, the notification unit 104, the meter management unit 105, the image analysis units 201 to 204 for each wavelength, the analysis result comparison units 205 to 208, the analysis result holding units 209 to 212, and the determination unit 300. To do.

  The meter management unit 105 performs management of the taximeter 30 and control of the image management unit 101 based on state data from the taximeter 30. Specifically, the state of the taxi is recognized based on the state data acquired from the taximeter 30 through the interface 21. For example, if the state data from the taximeter 30 represents the rented state, the state is recognized as a state in which a passenger is riding. Similarly, if the vehicle is in an empty state, the vehicle is recognized as being in a state where no passenger is present. If it is in the premium state, it is recognized as a state where a passenger is riding in the middle of the night. If it is a payment state, it will stop after a hiring state, and will recognize that it is the state which has received payment from the passenger.

Based on these recognition results, the meter management unit 105 sends to the image management unit 101 a camera control signal that determines the operation timing of the camera 10 (pressing of the shutter and the acquisition timing of an image captured thereby). In addition, when control data is received from the determination unit 300 described later, the state of the taximeter 30 is forcibly changed.
These functional operations in the meter management unit 105 will be described later.

  The image management unit 101 acquires an infrared image (electronic data for representing an image) from the camera 10 based on the camera control signal. The infrared image is repeatedly acquired at regular intervals, for example, every minute when the taxi is in an empty state. When the taxi stops to take passengers and changes to a rented state, the acquisition may be stopped. However, when monitoring the presence of passengers, the acquisition may be continued or the infrared image acquisition period may be set to 5 Get in minutes or more. When the taxi changes from an empty state to a rental state and further changes to a payment state, an infrared image is acquired at that point, and then image acquisition is stopped. In the following description, unless specifically distinguished, an infrared image acquired in an empty state or a rented state is referred to as a first infrared image, and an infrared image acquired when the payment state is changed is referred to as a second infrared image. .

  The acquired infrared image is input to the image management unit 101 as a set of images of a plurality of wavelengths. The wavelength A image analysis unit 201 analyzes the outline, arrangement, and the like of the object shown in the image of wavelength A among these multiple wavelength images, and causes the analysis result holding unit 209 to hold the analysis results. This process is repeated every time an infrared image is acquired.

  The analysis result comparison unit 205 compares the analysis result held in the analysis result holding unit 209 with the current analysis result. If the analysis results are the same, information indicating that there is no change in the vehicle (“no change”) is output to the determination unit 300. An object that is projected each time at the same position in the image is an object that is fixedly installed, so that the object can be held as a reference object, or the reference object can be masked and analyzed each time. The processing load for comparison may be reduced.

  On the other hand, when the analysis results are different, information (“changed”) indicating that the arrangement of the object in the vehicle has changed is output to the determination unit 300. The case where the analysis results are different is, for example, a case where an object different from the reference object is newly present, or a case where an object which is considered to be a passenger is present. When a passenger is reflected, the passenger usually has a large relative ratio in the infrared image, and in the case of a passenger, the body temperature is detected. Therefore, such an object is regarded as a passenger, and information indicating the presence of the passenger (“passenger present”) is output to the determination unit 300 in addition to “changed”.

  Similar processing is performed for the wavelength B image analysis unit 202, the wavelength C image analysis unit 203, the wavelength D image analysis unit 204, the analysis result comparison units 206, 207, 208, and the analysis result holding units 210, 211, 212. That is, the wavelength B image analysis unit 202 analyzes the outline, arrangement, and the like of the object shown in the image of the wavelength B, and causes the analysis result holding unit 210 to hold the analysis result. The wavelength C image analysis unit 203 analyzes the outline, arrangement, and the like of the object shown in the wavelength C image, and causes the analysis result holding unit 211 to hold the analysis result. The wavelength D image analysis unit 203 analyzes the contour, arrangement, and the like of the object shown in the image of the wavelength D, and causes the analysis result holding unit 212 to hold the analysis result. The analysis result comparison units 206, 207, and 208 compare the latest analysis result held in the corresponding analysis result holding unit with the current analysis result.

It should be noted that the image management unit 101 has one or several resolutions of wavelength A, wavelength B, wavelength C, and wavelength D that have higher resolution than other wavelengths in the imaging environment at the time of imaging (three or less in this example). It is desirable to notify the determination unit 300 that only the wavelength analysis result is valid.
The wavelength to be effective is individually determined based on the measurement result corresponding to the imaging environment such as the mounting position of the camera 10 in the vehicle, the performance of the camera 10, the traveling time zone, the weather, and the state of sunlight incident on the vehicle. Set. For example, as a result of analyzing the first infrared image or the second infrared image in various seasons, weather, time zones for each of the wavelength A, the wavelength B, the wavelength C, and the wavelength D, the analysis result of the wavelength that is the easiest to compare the analysis results Are those of wavelength A and wavelength D. In this case, the invalid flag is set for the wavelength B and the wavelength C, or the valid flag is set for the image analysis of the wavelength A and the wavelength D so that only the image analysis results of the wavelength A and the wavelength D are filtered.

  The discriminating unit 300 is a passenger in the vehicle according to the comparison of the status data input from the meter management unit 105, the information on the effective wavelength input from the image management unit 101, and the analysis result of the infrared image of the effective wavelength. Identify objects that may be forgotten passengers. That is, the presence or absence of the passenger or the object is determined by comparing the analysis results of the object reflected in the acquired infrared image, rather than simply comparing the infrared images before the passenger gets on and when getting off.

For example, when the taxi is in an empty state, the image analysis of the effective wavelength is for the wavelength A and the wavelength D, and the information “no change” is input from the analysis result comparison units 205 and 208, the object is also a passenger Is determined not to exist.
On the other hand, if the taxi changes from the rented state to the paid state and the passengers get off, but “changed” information is entered, the object that may be left behind by the passenger It is determined that it exists. In addition, when the taxi is in an empty state and “passenger” information is input, or when the taxi is in a rented state and “no change” information is input Determines that the taximeter 30 may be forgotten to be changed.

  When the state where there is a possibility that the taximeter 30 may be forgotten to be changed continues for a certain time (for example, 1 minute) or longer, the determination unit 300 outputs notification information indicating that fact to the notification unit 104 and controls data Is output to the taximeter management unit 105. The notification information is information generated for the purpose of notifying the crew, and the control data is data for changing the state of the taximeter 30. The control data is output when it is in the same state even after a certain period of time has elapsed after the notification information is output. The taximeter management unit 105 that has received the control data can forcibly change the state of the taximeter 30.

The notification unit 104 generates warning information based on the notification information from the meter management unit 105 and the determination result from the determination unit 300, and outputs this warning information to the warning unit 40. The content and data format of the warning information differ depending on whether the information is notification information or the content of the determination result of the determination unit 300.
In the case of the notification information, the warning information is generated as an operation signal of the warning lamp 42 so that the passenger is not informed. Due to this operation signal, the warning 42 flashes in red, for example. When the determination result of the determination unit 300 is “changed”, the warning information is a voice message “There is a possibility that something left behind is present”, and is output from the speaker 41 as synthesized speech.
The warning information may be output from an output device other than the speaker 41 or the warning 42.

<Camera layout>
FIG. 3 is an explanatory diagram of a specific arrangement of the camera 10 and an imaging part. FIG. 3A is a cross-sectional view when the inside of the taxi car is viewed from the side, and FIG. 3B is an overhead view of the inside of the taxi car. As described above, the four cameras 10 arranged in the present embodiment are the right seat camera 11, the left seat camera 12, the overhead camera 13, and the trunk camera 14.

  The right seat camera 11 is provided below the right seat in the front row seat, and is arranged on the floor mat in front of the rear seat right seat and between the rear seat right seat and the rear seat right door so as to capture an image. The left seat camera 12 is provided below the front seat left seat, and is disposed on the floor mat in front of the rear seat left seat and between the rear seat left seat and the rear seat left door so that an image can be taken. Each of the right seat camera 11 and the left seat camera 12 captures images below the seating surface of the rear seat. Thereby, the right seat camera 11 and the left seat camera 12 cannot image the face of the passenger who gets on the rear seat, and the passenger does not notice the presence of these cameras 11 and 12.

  The overhead camera 13 is provided on the ceiling on the rear seat, and is arranged so as to be able to capture an image on the seat surface of the rear seat. By disposing the overhead camera 13 behind the backrest portion of the rear seat, it is not possible to image the face of the passenger riding on the rear seat, and the passenger is not made aware of the presence of the camera 13. However, if it is too far behind, a blind spot is formed on the rear seat from the overhead camera 13. For this reason, it is not preferable to place the overhead camera 13 too far from the viewpoint of forgotten object detection. In this embodiment, the overhead camera 13 is provided above the headrest of the rear seat, and the entire seat surface of the rear seat is imaged without imaging the passenger's face.

  The trunk camera 14 is provided in the upper part in the trunk, and is arranged so as to be able to image the entire floor surface of the trunk.

  As described above, the camera 10 of the present embodiment is arranged so as to be able to image mainly the vicinity of the rear seat, the floor mat, and the trunk inside the taxi vehicle. Increase the number of cameras in order to increase the accuracy of detecting lost items. For example, two overhead cameras 13 may be provided on the rear seat right seat and the rear seat left seat. Conversely, the cost may be reduced by excluding the rear seat left seat periphery that is relatively easy to see by the crew from the imaging region (excluding the left seat camera 12). Alternatively, the camera 10 may be capable of panning and tilting, or a wide-angle lens, and the imaging region may be made variable to capture a wide range with a single camera.

<Operation mode: Information provision method>
Next, an operation mode example of the forgotten object prevention system 1 configured as described above will be described. Here, an example in which the forgotten object prevention system 1 is used as a tool for providing information to crew members will be described.
FIG. 4 is an explanatory diagram of the timing of information processing executed in the forgotten object prevention system 1. This information processing is appropriately performed depending on the state of the taximeter 30 operated mainly by the crew by events such as passenger boarding, getting off, and taxi departure. FIG. 4 shows an example in which there are no passengers in the empty state and there are passengers in the rental state.

  The process executed by the forgotten object prevention system 1 analyzes the outline, position, etc. of the object shown in the effective wavelength image in the first infrared image acquired from the camera 10 when the taxi is empty. The result is retained, and the object reflected in the second infrared image acquired from the camera 10 is analyzed when the passenger gets off after changing from the empty state to the rented state, and the analysis result is retained. Comparing with results.

  The analysis of the object reflected in the infrared image requires a great deal of processing. However, since the processing load of the forgotten object prevention system 1 is relatively small in the empty state, such an analysis can be performed in detail. Therefore, when the vehicle is in an empty state where it normally travels longer than the rented state, an analysis such as specifying the arrangement of the area around the rear seat, floor mat, and trunk is performed based on the first infrared image. , Keep the result. These processes are repeated during traveling. The period is normally about 1 minute as described above, but the interval may be automatically increased every time the empty state is continued for a certain period of time.

When the passenger gets off, the taxi meter 30 is changed from the rented state to the paid state by the crew. The process executed by the forgotten object prevention system 1 in the payment state analyzes the object reflected in the second infrared image, and compares the latest analysis result stored in advance with the analysis result by the first infrared image. Depending on the content of the comparison, it is possible to determine whether there are objects other than passengers in the vehicle, and if it is determined that objects other than passengers exist in the vehicle, there is a possibility that forgotten things have occurred for the crew Warning that there is sex.
The warning is given after the taxi meter 30 is in a rented state and before a lapse of a predetermined time when the passenger gets off, for example, within one minute immediately after getting off. Thereby, before a passenger leaves a taxi, a crew member can check the presence or absence of a thing left behind.

In addition, there is a possibility that the payment is made promptly by the passenger and the warning is not in time for departure. If a warning is issued after departure, even if the possibility of forgotten items is detected, it will not be prevented. Therefore, in order to reliably issue a warning before departure, a notification that there is no possibility of forgetfulness other than the warning may be given.
Or you may make it prevent that a warning is performed after departure by providing the control part 20 with the permission function which permits a departure only when a crew member responds with respect to warning or alerting | reporting information. The response by the crew in this case includes, for example, mute of the speaker 41 when the warning or notification information is output by the speaker 41, and reset of the warning light 42 when the warning light 42 is used. Thus, when there is an object that may be left behind, the taxi crew can be surely notified of the situation.

The processing procedure of the left-behind item prevention system 1 when providing information to the crew is as follows.
FIG. 5 is an explanatory diagram of the procedure for providing information when a passenger is about to get off. The thing-left-behind prevention system 1 confirms that the taximeter 30 is in a payment state. If it is not in the payment state, the information providing process is not performed (S10: N).
If the taxi meter 30 is in the payment state (S10: Y), it is time for the passengers to get off, so the control unit 20 says "Thank you for the ride. Please check the items around you so that there are no leftovers. The voice guidance such as “is performed by the speaker 41 (S11). With such voice guidance, first, the passengers are alerted to things left behind.

During the voice guidance, the forgotten object prevention system 1 determines the object that may be forgotten according to the comparison contents of the analysis result of the object reflected in the first infrared image and the analysis result of the object reflected in the second infrared image. The presence or absence is discriminated (S12). Specifically, the analysis result in the second infrared image is compared with the analysis result in the first infrared image held in advance. As a result, if both are reference objects and there is no change, there is no object that may be forgotten, so a notification of no change is made, and the information providing process is finished as it is (S12 “No”). .
When the object reflected in the second infrared image cannot be confirmed, for example, when the light from the outside of the vehicle is too strong (such as overexposure of the infrared image) (S12 “cannot be determined”), it is based on the second infrared image of the next cycle. Repeat the discrimination process. If it is still impossible to determine, it is determined that there is no object different from the reference object. If the object exists (S12: present), a warning is output (S13).

  Through the processing as described above, the crew can know the possibility of occurrence of forgotten things from the warning sound (voice) from the speaker 41 and the display of the warning light 42. You can call attention.

<Analysis process during travel>
6 and 7 are explanatory diagrams of processing procedures executed by the forgotten object prevention system 1 when a taxi is traveling. FIG. 6 shows an example when the taxi starts running after the taxi meter 30 changes from the paid state to the empty state, and FIG. 7 shows an example when the taxi meter 30 runs in the rented state. In FIG. 6, it is assumed that information on the reference object analyzed in the empty state is held as an analysis result.

  Referring to FIG. 6, when the forgotten object prevention system 1 recognizes that the taximeter 30 is in an empty state (S20: Y), it performs an object / passenger discrimination process (S21). That is, an infrared image is acquired from the camera 10, and the outline, arrangement, etc. of the object in the vehicle reflected in the infrared image are analyzed. The criterion for determining whether or not the object shown in the infrared image is a passenger is as described above. If the current analysis result is compared with the previous analysis result held in advance, if there is no change from the reference object, there is no object that may be forgotten, so no change notification is given (S21 “ No change ", S22), and the process returns to S20. In the notification of no change, the warning lamp 42 may be lit in green, or nothing may be displayed.

  In S21, when it is determined that there is no object corresponding to the passenger but there is an object different from the reference object (S21: “changed”, “no passenger”), the forgotten object prevention system 1 uses the contour or the like as the object Specify (S23). Since this state is an object discrimination failure in the information providing process of FIG. 5, a warning is issued again, and the process returns to S20.

In S21, when it is determined that there is a passenger (S21: “changed”, “passenger present”), the state of the taximeter 30 and the actual in-vehicle state are different. This is a phenomenon that generally occurs when the crew member forgets to switch the taximeter 30 from the empty state to the rented state or the premium state. Therefore, the forgotten object prevention system 1 notifies that there is a passenger (S24) and blinks the warning lamp 42. If the predetermined time has not elapsed, the process returns to S20 while the blinking state of the warning lamp 42 is continued. If it is the same after a lapse of a certain time, the state of the taximeter 30 is forcibly changed (S25: Y, S26).
As described above, the flashing of the warning light 42 can be eliminated when the crew member responds. Moreover, when it is not necessary to change the state of the taximeter 30 compulsorily, the process is abbreviate | omitted and it returns to S20 with a passenger presence notification.

Referring to FIG. 7, when the forgotten object prevention system 1 recognizes that the taximeter 30 is in a rented state (S30: Y), it determines whether or not there are passengers (S31).
In S31, when it is determined that there is a passenger (S31: “passenger present”), the taxi meter 30 and the actual in-vehicle state match, so a passenger presence notification is made (S32), and the process returns to S30. In the passenger presence notification, the warning lamp 42 may be lit in green, or nothing may be displayed.

In S31, when it is determined that there is no passenger (S31: “No passenger”), the state of the taximeter 30 and the actual in-vehicle state are different. This is a phenomenon that generally occurs when the crew member forgets to switch the taximeter 30 from the rented state to the empty state. Therefore, the forgotten object prevention system 1 notifies that there is no passenger (S33) and blinks the warning lamp 42. If the predetermined time has not elapsed, the process returns to S30 while the blinking state of the warning lamp 42 is continued. If it is the same after a certain time has elapsed, the state of the taximeter 30 is forcibly changed (S34: Y, S35).
As described above, when the crew member responds, the blinking state can be eliminated. Moreover, when it is not necessary to forcibly change the state of the taximeter 30, the process is omitted, and the process returns to S30 with the passenger absence notification.

[Effects of the embodiment]
As described above, in the forgotten object prevention system 1 of the present embodiment, the inside of a taxi is imaged using infrared rays of a specific wavelength that has a higher resolution than other wavelengths in the imaging environment when the shutter of the camera 10 is pressed. For this reason, it is possible to prevent the occurrence of a situation such as misidentifying a shadow as an object or not being able to recognize an object existing in a shadow portion, which may occur in a conventional imaging device that performs imaging using visible light.
Moreover, the analysis of the outline and arrangement of the object reflected in the infrared image is repeatedly performed, and the analysis results are compared with each other to determine an object that is likely to be a passenger or forgotten item. Therefore, the processing load is significantly reduced as compared with the case where the captured image itself is held and the images are compared with each other. As a result, it is possible to detect an object that may be forgotten in a short time until the passenger completes payment and gets off the vehicle.
Furthermore, since imaging is performed using infrared rays, it is possible not only to harm passengers' privacy, but also to identify objects that exist in dark places such as trunk rooms, lower seats of rear seats, or floor mats. Therefore, the detection accuracy can be remarkably improved as compared with the case where an imaging device using visible light is used.

[Modification]
Although this embodiment is an example when the forgotten object prevention system 1 is mounted in a taxi, it may be mounted in another passenger car such as a hire other than a taxi.
Further, in the present embodiment, as an example of analyzing an object reflected in an infrared image of a specific wavelength, the wavelength-specific image analysis units 201 to 204, the analysis result comparison units 205 to 208, and the analysis result holding unit 209 to FIG. 212 is provided, but is not limited to this example. For example, a filter (filtering circuit) that transmits only infrared rays having the specific wavelength may be provided.
A GPS receiver may be provided so that it is possible to identify when and where the passenger is left behind based on position data and time data derived from the GPS data. In this case, when an object that may be forgotten is determined, it is necessary to store information about the object in association with GPS data.

  DESCRIPTION OF SYMBOLS 1 ... Lost thing prevention system, 10 ... Camera, 11 ... Right seat camera, 12 ... Left seat camera, 13 ... Overhead camera, 14 ... Trunk camera, 20 ... Control part, 21 ... Interface, 22 ... CPU, 23 ... RAM, 24 ... ROM, 25 ... HDD, 30 ... Taximeter, 40 ... Warning section, 41 ... Speaker, 42 ... Warning light, 101 ... Image management section, 104 ... Notification section, 105 ... Meter management section, 201-204 An image analysis unit for each wavelength, 205 to 208, an analysis result comparison unit, 209 to 212, an analysis result holding unit.

Claims (3)

It is installed in a passenger car having a renting state where passengers exist and an empty state where no passengers exist, and in the imaging environment at that time, infrared rays of a specific wavelength that has a higher resolution than other wavelengths are used. An imaging device for imaging the interior of the vehicle;
When an object reflected in the first infrared image acquired from the imaging device in the empty state is analyzed, the analysis result is retained, and the passenger gets off after changing from the empty state to the rented state Analyzing the object reflected in the second infrared image acquired from the imaging device, and comparing the analysis result with the held analysis result;
According to the comparison content by the analysis means, a determination means for determining whether an object other than a passenger exists in the vehicle,
When it is determined that an object other than the passenger is present in the vehicle, the warning means for warning that there is a possibility that something left behind has occurred for the passenger of the passenger car,
A left-behind prevention system. It is installed in a passenger car having a renting state where passengers exist and an empty state where no passengers exist, and in the imaging environment at that time, infrared rays of a specific wavelength that has a higher resolution than other wavelengths are used. A method executed by a computer to which an imaging device for imaging the inside of a vehicle is connected,
Analyzing the object reflected in the first infrared image acquired from the imaging device when the passenger car is in an empty state, holding the analysis result, and after changing from the empty state to the rental state, Analyzing the object reflected in the second infrared image acquired from the imaging device when getting off the vehicle, and comparing the analysis result with the held analysis result;
Determining whether an object other than a passenger exists in the vehicle according to the comparison content of the analysis result;
When it is determined that an object other than the passenger is present in the vehicle, a warning is given to a passenger of the passenger car that there is a possibility that something left behind has occurred.
How to provide information to passenger car crews. It is installed in a passenger car having a renting state where passengers exist and an empty state where no passengers exist, and in the imaging environment at that time, infrared rays of a specific wavelength that has a higher resolution than other wavelengths are used. A computer program for operating a computer connected to an imaging device for imaging an interior of a vehicle as a forgotten object prevention system,
The computer,
When an object reflected in the first infrared image acquired from the imaging device in the empty state is analyzed, the analysis result is retained, and the passenger gets off after changing from the empty state to the rented state Analyzing the object reflected in the second infrared image acquired from the imaging device, and comparing the analysis result with the stored analysis result,
Discriminating means for discriminating whether an object other than a passenger exists in the vehicle according to the comparison content by the analyzing means,
A computer program that functions as a warning unit that warns a passenger in a passenger car that there is a possibility that something left behind has occurred when it is determined that an object other than the passenger is present in the vehicle.

Images