JP6915995B2 - Drive recorder - Google Patents

Description

The present invention relates to a drive recorder, an image recording method, and a monitoring device.

A drive recorder that records an image taken by a camera mounted on a vehicle is known (for example, Patent Document 1). By displaying the image recorded by the drive recorder on the display device, the driver can verify his / her driving situation.

Japanese Unexamined Patent Publication No. 2016-28332

When recording an image taken by a camera mounted on a vehicle on a drive recorder, the image recorded on the drive recorder is an image for protecting the privacy of the person reflected in the image, the owner of another vehicle, etc. It is preferable to apply the treatment.

However, if image processing is applied to the image recorded on the drive recorder to protect privacy, it may be difficult for the driver who sees the image displayed on the display device to grasp the situation such as the driving situation. be.

Even when image processing is performed on an image taken by a fixed-point camera to protect privacy, it may be difficult for an observer who sees the image displayed on the display device to grasp the situation.

An object of the present invention is to provide a drive recorder, an image recording method, and a monitoring device capable of generating an image in which it is easy to grasp a situation such as a driving situation while protecting the privacy of an object reflected in the image. ..

The drive recorder according to the present invention is a drive recorder that records an image taken by a camera mounted on a vehicle, and has an image extraction unit that extracts an area in which another vehicle is photographed from the image and another vehicle. The area is provided with an image processing unit that performs image processing for privacy protection that reflects the characteristics of other vehicles.

Preferably, the image extraction unit extracts an area in which a person is photographed from the image, and the image processing unit performs image processing for privacy protection that reflects the characteristics of the person in the area in which the person is imaged. To give.

Preferably, the image processing unit superimposes and displays the occupant characteristic image reflecting the characteristics of the occupants of the other vehicle on the area where the other vehicle subjected to the image processing for privacy protection is captured.

Preferably, the image processing unit generates a traveling information image that reflects information on the traveling of another vehicle.

Preferably, the image processing unit generates a biometric information image that reflects the biometric information of the driver of another vehicle.

The image recording method according to the present invention is an image recording method for recording an image taken by a camera mounted on a vehicle on a drive recorder, in which the drive recorder extracts an area in which another vehicle is captured from the image. The drive recorder is provided with a step of performing image processing for privacy protection reflecting the characteristics of the other vehicle on the area where the other vehicle is photographed.

The monitoring device according to the present invention has an image extraction unit that extracts an area in which an object is photographed from the image taken by a camera, and a privacy protection that reflects the characteristics of the object in the area in which the object is photographed. It is provided with an image processing unit for performing the image processing of the above.

According to the present invention, it is possible to generate an image in which it is easy to grasp a situation such as a driving situation while protecting the privacy of the object reflected in the image.

It is the schematic of the vehicle which arranged the drive recorder by this invention. It is a figure for demonstrating an example of viewing the image recorded in the drive recorder by this invention. It is a figure which shows the structure of one Embodiment of the drive recorder by this invention. It is a figure which shows an example of the person image feature amount table used in the drive recorder by this invention. It is a figure which shows an example of the background image feature amount table used in the drive recorder by this invention. It is a figure for demonstrating an example of image processing for privacy protection performed by the drive recorder by this invention. It is a figure for demonstrating another example of image processing for privacy protection performed in the drive recorder by this invention. It is a flowchart which shows an example of the image processing procedure for privacy protection of the drive recorder shown in FIG. It is a figure which shows the structure of one Embodiment of the monitoring device by this invention.

The drive recorder, image recording method, and monitoring device according to the present invention will be described in detail with reference to the drawings. In the drawings, the same components are designated by the same reference numerals.
FIG. 1 is a schematic view of a vehicle in which a drive recorder according to the present invention is arranged. The drive recorder 2 mounted on the vehicle 1 such as an automobile takes a picture of the periphery of the car 1 with a built-in camera, and records the image obtained by taking the picture. In the present embodiment, image recording is continuously performed from the start to the end of the drive of the drive recorder 2 regardless of the occurrence of events such as sudden braking and collision. Further, in the present embodiment, the driver can verify his / her own driving situation by displaying the image recorded by the drive recorder 2 on the display device 3.

FIG. 2 is a diagram for explaining an example of viewing an image recorded on a drive recorder according to the present invention. As shown in FIG. 2, the data recorded on the memory card 4 by the drive recorder 2 can be input to the computer 5. In this case, the driver or the like can view the recorded image on the display of the computer 5 and edit the image.

FIG. 3 is a diagram showing a configuration of an embodiment of a drive recorder according to the present invention. When the ACC switch 6 is turned on, the drive recorder 2 is supplied with electric power from the battery 7 provided in the vehicle 1 to start driving, and when the ACC switch 6 is turned off, the electric power supplied from the battery 7 is cut off. Stop.

The drive recorder 2 includes a RAM 21, a camera 22, a processing circuit 23, a microphone 24, a non-volatile memory 25, a CPU 26, an operation unit 27, an acceleration sensor 28, a card slot 29, and an image output unit 30. , And a timekeeping circuit 31.

The RAM 21 has a storage area for work of arithmetic processing of the CPU 26, an image storage area for temporarily storing image data, and a voice storage area for temporarily storing voice data.

The camera 22 includes a lens and an image pickup device, and continuously takes an image while the drive recorder 2 is being driven.

The processing circuit 23 is a hardware circuit that processes an image acquired by the camera 22. The processing circuit 23 performs image processing such as luminance correction, contrast correction, and compression on the image signal input from the camera 22, and generates image data in a JPEG format or the like. The image processed in the processing circuit 23 is stored in the image storage area of the RAM 21.

The microphone 24 collects sound in the vicinity of the vehicle 1 and acquires voice data. The acquired voice data is stored in the voice storage area of the RAM 21.

The non-volatile memory 25 is composed of, for example, a flash memory. The non-volatile memory 25 stores data used for various controls in addition to the program as firmware.

The CPU 26 controls each part of the drive recorder 2 by executing arithmetic processing according to a program stored in the non-volatile memory 25. In the present embodiment, the CPU 26 has an image extraction unit 26a, an image processing unit 26b, and an event detection unit 26c.

The image extraction unit 26a extracts an area (hereinafter, “object image”) in which a detection object to be protected for privacy is captured from the image. In the present embodiment, the detection target is another vehicle and a person. Hereinafter, the area in which the other vehicle is shown is referred to as an "other vehicle image", and the area in which the person is shown is referred to as a "person image". For example, the image extraction unit 26a uses a classifier (not shown) set for each type (for example, vehicle type) such as AdaBoost and a support vector machine learned in advance for detecting the image of the object to be detected. Extract the image. For example, the image extraction unit 26a sets a detection window for the image, and extracts the feature amount used by the classifier from the area in which the detection window is set. The feature amount is, for example, a HOG feature amount. Then, the image extraction unit 26a obtains the result of whether or not the object image exists in the area where the detection window is set by inputting the extracted feature amount into the classifier. The image extraction unit 26a determines whether or not the detection target is present in the image by repeating the above process while changing the position of the detection window, and if the detection target is present, the detection target is present. The image is extracted, associated with the type, and stored in the storage area of the RAM 21. The above-mentioned processing is an example, and other processing can be applied to extract the image of the object to be detected, that is, the image of another vehicle and the image of a person.

The image processing unit 26b performs image processing for privacy protection that reflects the characteristics of the person on the person image for which the type is specified, and privacy protection that reflects the characteristics of the other vehicle for the image of the other vehicle for which the type is specified. Perform image processing for.

In the present embodiment, the image processing unit 26b reflects the characteristics of the person image on the person image in order to perform image processing for privacy protection that reflects the characteristics of the person on the person image for which the type is specified. Convert to a person feature image (avatar image).

For example, when the type includes gender (male or female) and age (adult or child), the image processing unit 26b displays the person image as a person characteristic image reflecting the gender (male or female) and age (adult or child). Convert to. When the type includes face type, age, height and clothing color, the image processing unit 26b displays the person image as a person reflecting the gender (male or female), age (adult or child) height and clothing color. Convert to a feature image.

FIG. 4 shows an example of a table showing the relationship between the type and the avatar of the character feature image. The table T1 shown in FIG. 4 is stored in the non-volatile memory 25, and the avatar XXXXX of the character feature image representing an adult male is stored in association with the person type A1 representing an adult male, and is stored in association with an adult female. The avatar YYYYYY of the person feature image representing the above is stored in association with the person type A2 representing an adult female.

In addition, the image processing unit 26b reflects the characteristics of the other vehicle image on the other vehicle image in order to perform image processing for privacy protection that reflects the characteristics of the other vehicle on the other vehicle image for which the type is specified. Convert to other vehicle feature image. For example, when the type includes a vehicle type and a vehicle color, the image processing unit 26b converts the other vehicle image into another vehicle-to-feature image that reflects the vehicle type and the vehicle color.

FIG. 5 shows an example of a table showing the relationship between the vehicle type and the code of the other vehicle feature image. The table T2 shown in FIG. 5 stores the non-volatile memory 25, and the avatar RRRRRRR of the other vehicle feature image representing the red car is stored in association with the vehicle type B1 representing the red car, and is white. The avatar WWWWW of the other vehicle feature image representing the automobile is stored in association with the vehicle type B2 representing the white automobile.

Further, when the vehicle 1 receives a signal related to other vehicle information such as the model year of the other vehicle from the other vehicle by inter-vehicle communication, the image processing unit 26b reflects the other vehicle image, the vehicle type, the color of the vehicle, and the other vehicle information. Convert to another vehicle-to-feature image. In this case, the non-volatile memory 25 corresponds the avatarized other vehicle-to-characteristic image with the type code (type code B1, type code B2, etc.) and other vehicle information corresponding to the vehicle in the table T2 of FIG. Attach and memorize.

FIG. 6 is a diagram for explaining an example of image processing for privacy protection performed by the drive recorder according to the present invention. As shown in FIG. 6A, when the image includes another vehicle image X corresponding to a red automobile, a person image Y corresponding to an adult man, and a person image Z corresponding to an adult woman, the image processing unit 26b The other vehicle image X, the person image Y, and the person image Z are converted into the other vehicle feature image X', the person feature image Y', and the person feature image Z', respectively.

The other vehicle feature image X'is another vehicle feature image indicating that the vehicle is a red vehicle, and the avatarized other vehicle feature image X'is stored in the non-volatile memory 25 in association with the type B1 shown in FIG. It is remembered.

The person feature image Y'is a person feature image indicating that the person is an adult male, and the avatarized person feature image Y'is stored in the non-volatile memory 25 in association with the type A1 shown in FIG. There is.

The person feature image Z'is a person feature image indicating that the person is an adult woman, and the avatarized person feature image Z'is stored in the non-volatile memory 25 in association with the type A2 shown in FIG. There is.

Further, when the image processing unit 26b converts the other vehicle image X, the person image Y, and the person image Z into the other vehicle feature image X', the person feature image Y', and the person feature image Z', the other vehicle image X , Person image Y and person image Z are encrypted, respectively. Then, the image processing unit 26b stores the encrypted other vehicle image X, the person image Y, and the person image Z in the storage area of the RAM 21 in association with the corresponding image subjected to the image processing for privacy protection.

The image processing unit 26b may superimpose and display the occupant feature image reflecting the characteristics of the occupants of the other vehicle on the other vehicle feature image. The occupant feature image is, for example, an avatarized person feature image. This makes it possible to identify what kind of person was driving (or riding) the vehicle. Further, when the image processing unit 26b extracts a mark image (for example, a young leaf mark image) attached to another vehicle, the image processing unit 26b superimposes a sprout image on the other vehicle feature image as a occupant feature image reflecting the mark. It may be displayed. In this case, the avatarized sprout image is stored in the non-volatile memory 25. The mark image such as the sprout image is an example of the occupant feature image.

In addition, the image processing unit 26b may generate a traveling information image that reflects information on the traveling of another vehicle. For example, when the vehicle 1 acquires the continuous driving time of the driver of another vehicle by inter-vehicle communication, the image processing unit 26b displays the continuous driving time together with the other vehicle feature image. The continuous operation time is an example of information on the running of another vehicle. Further, when the vehicle 1 receives a signal indicating that the other vehicle is an autonomous driving vehicle by inter-vehicle communication, the image processing unit 26b displays a traveling information image indicating the autonomous driving vehicle together with the other vehicle feature image. Let me. Further, when the vehicle 1 receives a signal indicating dangerous driving (for example, a signal indicating a speed violation) from another vehicle by inter-vehicle communication, the image processing unit displays a traveling information image indicating dangerous driving. Displayed together with other vehicle feature images.

In addition, the image processing unit 26b may generate a biometric information image that reflects the biometric information of the driver of another vehicle. For example, when the driver of another vehicle wears a wearable device that measures the heart rate, and a signal related to information such as the heart rate acquired by the wearable device is transmitted from the other vehicle to the vehicle 1 via the device of the other vehicle. , The image processing unit 26b generates a biometric information image reflecting the biometric information of the driver of the other vehicle (for example, a biometric information image indicating that the heart rate of the driver of the other vehicle is zero), and features the other vehicle. Display with the image.

In addition, the image processing unit 26b may perform blurring processing in order to perform image processing for privacy protection. In this case, the blurred image is colored to distinguish between the blurred images. Further, in order to perform image processing for privacy protection, the image processing unit 26b uses a license plate of a specific vehicle instead of converting another vehicle image into another vehicle feature image for a specific vehicle (for example, a truck). You may erase it from the image.

FIG. 7 is a diagram for explaining another example of image processing for privacy protection performed by the drive recorder according to the present invention. As shown in FIG. 7A, when the other vehicle image T1 corresponding to the truck is included in the image, the image extraction unit 26a extracts the license plate image Tr2 from the other vehicle image Tr1, and the image processing unit 26b extracts the license plate image Tr2. Convert Tr2 to a black-painted image Tr2'.

The image processing unit 26b encrypts the license plate image Tr2 when converting the license plate image Tr2 into the black-painted image Tr2'. Then, the image processing unit 26b stores the encrypted license plate image Tr2 in the storage area of the RAM 21 in association with the corresponding image subjected to the image processing for privacy protection.

Further, the image processing unit 26b may not perform image processing for privacy protection on a specific vehicle (for example, a police car). Further, the image processing unit 26b may adjust the image so that the number of other vehicle images included in the image does not exceed a predetermined number (for example, 5). For example, the image processing unit 26b adjusts the image so that the number of other vehicle images included in the image is 4 or less when the number of other vehicle images included in the image is 7 (for example, enlargement of the image). May be done.

The event detection unit 26c detects the occurrence of an event for the vehicle 1. That is, the event detection unit 26c detects whether or not an event has occurred for the vehicle 1 based on the acceleration detected by the acceleration sensor 28. When the event detection unit 26c detects the occurrence of an event, the event detection unit 26c associates the information regarding the detection of the event occurrence (for example, the event occurrence time) with the corresponding image (for example, the image taken at the event occurrence time) in the non-volatile memory. Store in 25.

The operation unit 27 includes a plurality of buttons for receiving various instructions from the driver and the like. The content of the instruction from the user received by the operation unit 27 is input to the CPU 26 as an electric signal. The memory card 4 can be attached to and detached from the card slot 29, and the image data, audio data, and the like stored in the RAM 21 can be read and written to and from the attached memory card 4. The image output unit 30 converts the image data into a video signal conforming to a method such as NTSC and outputs the image data to the display device 3. The timekeeping circuit 31 transmits the current time data to the CPU 26.

FIG. 8 is a flowchart showing an example of an image processing procedure for protecting the privacy of the drive recorder shown in FIG. This flow is executed mainly by the control unit 26 in cooperation with each element of the drive recorder 2 based on the program stored in advance in the non-volatile memory 25. This flow is performed in non-real time or in real time.

First, the image extraction unit 26a determines whether or not another vehicle image has been extracted from the image (step S1). When the other vehicle image is extracted from the image (Yes in step S1), the image processing unit 26b specifies the type of the other vehicle image (step S2). Then, the image processing unit 26b performs image processing for privacy protection that reflects the characteristics of the other vehicle (step S3).

For example, when the image processing unit 26b specifies in step S2 that the type of the other vehicle image is the background type B1 (red vehicle) of FIG. 5, the image processing unit 26b displays the vehicle image as a red vehicle. It is converted into another vehicle feature image (for example, another vehicle feature image X'in FIG. 6) showing that.

After performing image processing for protecting the privacy of other vehicles in step S3, or when the image of another vehicle is not extracted from the image (No in step S1), has the image extraction unit 26a extracted a person image from the image? Whether or not it is determined (step S4). When a person image is extracted from the image (Yes in step S4), the image processing unit 26b specifies the person type of the person image (step S5). Then, the image processing unit 26b performs image processing for privacy protection that reflects the characteristics of the person (step S6).

For example, when the image processing unit 26b specifies in step S5 that the type of the person image is the person type A1 (adult male) in FIG. 4, the image processing unit 26b sets the person image to be an adult male. Is converted into a person feature image representing (for example, the person feature image Y'in FIG. 6).

After performing image processing for privacy protection for a person in step S3, or when another vehicle image is not extracted from the image (No in step S1), the control unit 26 ends the processing.

According to the present embodiment, by performing image processing for privacy protection that reflects the characteristics of the other vehicle, the driving situation while protecting the privacy of the driver, owner, etc. of the other vehicle reflected in the image. It is possible to generate an image that makes it easy to grasp the situation such as.

For example, if there is a testimony that "I saw a red car running", whether or not the image of other vehicle features indicating that it is a red car is included in the image after privacy protection processing By checking, it is possible to confirm whether the testimony is correct while protecting privacy.

In addition, the other vehicle image is converted into another vehicle-to-characteristic image that reflects the vehicle type, vehicle color, and model year. If there is a testimony that "I saw", the privacy is protected by checking whether the other vehicle feature image showing the MM of the 10-year red company T is included in the image after the privacy protection process. At the same time, it is possible to confirm whether the testimony is correct.

In addition, by performing image processing for privacy protection that reflects the characteristics of the person, the drive recorder 2 records an image that can image the situation such as the driving situation while protecting the privacy of the person reflected in the image. be able to.

In addition, by superimposing an occupant feature image that reflects the characteristics of the occupants (for example, the driver) of another vehicle on the image, the occupants of the other vehicle can protect the privacy of the occupants of the other vehicle reflected in the image. It is possible to generate an image that makes it easy to grasp the situation. For example, a person image corresponding to a driver of another vehicle is extracted, and the extracted person image is converted into a person characteristic image representing an adult man dressed in blue. If there is a testimony that "I was driving", I protect my privacy by checking whether a person feature image representing an adult man dressed in blue is included in the image after privacy protection processing. At the same time, it is possible to confirm whether the testimony is correct. The determination of whether or not the person image is the driver of another vehicle is made by, for example, determining whether or not the position of the head of the person image is at the position of the windshield of the other vehicle image by the image processing unit 26b. Will be done.

Further, by generating a traveling information image that reflects information on the traveling of another vehicle, it is possible to generate an image that makes it easy to grasp the traveling of the other vehicle while protecting the privacy of the other vehicle reflected in the image.

In addition, by generating a biometric information image that reflects the biometric information of the driver of another vehicle, an image that makes it easy to grasp the biometric information of the driver of the other vehicle while protecting the privacy of the other vehicle reflected in the image is created. Can be generated.

FIG. 9 is a diagram showing a configuration of an embodiment of a monitoring device according to the present invention. The monitoring device 8 is supplied with electric power from an external power source (not shown) and exchanges various information via the Internet network. The monitoring device 8 includes a RAM 21, a camera 22, a processing circuit 23, a microphone 24, a non-volatile memory 25, a CPU 26', an image output unit 30, and a timekeeping circuit 31. The CPU 26'has an image extraction unit 26a and an image processing unit 26b.

When the camera 22 is a fixed-point camera and the camera 22 captures a procession of a store, an attraction, or the like, the monitoring device 8 distributes an image obtained by converting a person image into a person feature image on the Web in real time. The portrait image is an example of the target image. The CPU 26'may calculate the number of people in the line and the waiting time from the number of human feature images, and change the display according to the waiting time. The display is changed by superimposing the waiting time on the image, changing the person feature image in the image, the color of the building, and the like.

When the camera 22 is a fixed-point camera and the camera 22 photographs the inside of a bus or a train, the monitoring device 8 distributes an image obtained by converting a person image into a person feature image on the Web in real time. The person who sees the image distributed on the Web can check the congestion condition of the bus or train to be boarded, the vacant seat information, the age group of the user, and the like.

When the camera 22 is a fixed-point camera and the camera 22 captures a procession at a checkout in a supermarket or the like, the monitoring device 8 displays an image obtained by converting a person image into a person feature image on the display device 3. The CPU 26'may calculate the waiting time from the information on the number of baskets and the amount of products in the basket, and superimpose and display the waiting time on the image.

According to the present embodiment, by performing image processing for privacy protection that reflects the characteristics of the target such as a person or a vehicle, the image is taken by the camera 22 while protecting the privacy of the person or the like reflected in the image. It is possible to generate an image that makes it easy to grasp the situation of the place.

1 Vehicle 2 Drive recorder 3 Display device 4 Memory card 5 Computer 6 ACC switch 7 Battery 8 Monitoring device 21 RAM
22 Camera 23 Processing circuit 24 Microphone 25 Non-volatile memory 26,26'CPU
26a Image extraction unit 26b Image processing unit 26c Event detection unit 27 Operation unit 28 Accelerometer 29 Card slot 30 Image output unit 31 Timekeeping circuit

Claims (7)

A drive recorder that records images taken by a camera mounted on a vehicle.
An image extraction unit that extracts an area in which another vehicle is captured from the image, and an image extraction unit.
An image processing unit that converts the area in which the other vehicle is captured into another vehicle feature image that reflects the features related to the type of the other vehicle.
Equipped with a,
The image processing unit is a drive recorder that superimposes and displays an occupant feature image that reflects the characteristics of the occupants of another vehicle on the other vehicle feature image. The drive recorder according to claim 1, further comprising a storage unit that stores an area in which the other vehicle is copied in association with the type of the other vehicle. When the type of the other vehicle is specified by the image processing unit and the vehicle receives a signal relating to the type of the other vehicle from the other vehicle, the image processing unit obtains the area in which the other vehicle is copied. The method 1 or 2 for converting into another vehicle feature image that reflects the type of the other vehicle specified by the image processing unit and the characteristics related to the type of the other vehicle specified by the signal received from the other vehicle. Drive recorder. The image extraction unit extracts an area in which a person is captured from the image.
The drive recorder according to any one of claims 1 to 3, wherein the image processing unit converts an area in which the person is copied into a person feature image that reflects features related to the type of the person. The drive recorder according to any one of claims 1 to 4 , wherein the image processing unit generates a traveling information image reflecting information on traveling of another vehicle. The drive recorder according to any one of claims 1 to 5 , wherein the image processing unit generates a biometric information image that reflects biometric information of a driver of another vehicle. A drive recorder that records images taken by a camera mounted on a vehicle.
An image extraction unit that extracts an area in which another vehicle is captured from the image, and an image extraction unit.
An image processing unit that converts the area in which the other vehicle is captured into another vehicle feature image that reflects the features related to the type of the other vehicle.
Equipped with a,
When the type of the other vehicle is specified by the image processing unit and the vehicle receives a signal relating to the type of the other vehicle from the other vehicle, the image processing unit obtains the area in which the other vehicle is copied. A drive recorder that converts into another vehicle feature image that reflects the characteristics related to the type of the other vehicle specified by the image processing unit and the type of the other vehicle specified by the signal received from the other vehicle.

Images