JP5239952B2 - Safe driving monitoring device - Google Patents

Description

  The present invention relates to a safe driving monitoring apparatus that monitors the safety of a person riding a two-wheeled vehicle such as a bicycle or a motorcycle traveling on a road.

  Conventionally, there has been an apparatus that determines and detects whether a moving object traveling on a road is an automobile (four-wheeled vehicle), a motorcycle (two-wheeled vehicle), a bicycle (two-wheeled vehicle), or a pedestrian (Patent Document 1). 2). Further, Patent Document 1 describes a configuration that outputs a warning to that effect when a detected object type is not appropriate, such as when a bicycle is detected at the entrance of an automobile-only road.

Japanese Patent No. 38437533 Japanese Patent No. 4120430

  However, Patent Documents 1 and 2 detect the type of a moving object traveling on a road, and are on an automobile (four-wheeled vehicle), a motorcycle (two-wheeled vehicle), a bicycle (two-wheeled vehicle) or the like. It was not intended to monitor human safety.

  On the other hand, bicycle accidents have been increasing recently. In particular, for children and infants under 13 years of age, discussions have been made on the requirement of wearing a helmet when riding a bicycle. There are also local governments and schools and other organizations that are already active in enlightening wearing helmets.

  The object of the present invention is to promote the wearing of a helmet for a person riding a motorcycle, particularly a child / infant riding a bicycle, thereby raising the safety awareness of the person riding the motorcycle, An object of the present invention is to provide a safe driving monitoring device that improves the wearing rate of a helmet.

  The safe driving monitoring apparatus of the present invention is configured as follows to achieve the above object.

  In this invention, an image processing means processes the captured image of the imaging device which is imaging the road, and detects the face and head of a person riding on the two-wheeled vehicle traveling on the road. Further, the helmet wearing determination means determines whether or not the helmet is worn based on the ratio between the width of the face of the person riding on the motorcycle and the width of the head. Then, the output means outputs a warning to a person who is not wearing a helmet. For example, the output means outputs, as a warning output, a voice message that prompts the user to wear a helmet “Dangerous, let's wear a helmet when riding a bicycle”. Thereby, the consciousness with respect to the safety | security of the person who rides a two-wheeled vehicle can be raised, and the improvement of the wearing rate of a helmet can be aimed at.

  In addition, the age of this person is estimated from the face image of the person riding in the motorcycle, and a warning message such as a voice message prompting the user to wear a helmet is output only to children and infants under 13 years old. May be. In addition, when a motorcycle is not a bicycle but a motorcycle, it is mandatory to wear a helmet under the Road Traffic Law, so a voice message prompting the user to wear a helmet is output regardless of the age of the passenger. Just do it.

  According to the present invention, it is possible to urge a person who rides a motorcycle, particularly a child / infant riding a bicycle, to wear a helmet. In addition, this makes it possible for a person riding in a motorcycle to raise awareness of safety and improve the helmet wearing rate.

It is a figure which shows the structure of the principal part of a safe driving | operation monitoring apparatus. It is a flowchart which shows operation | movement of a safe driving | operation monitoring apparatus. It is a figure explaining the process which determines the wearing presence or absence of a helmet. It is a figure which shows the bicycle with which the infant and guardian are riding two persons on a bicycle.

  Hereinafter, a safe driving monitoring apparatus according to an embodiment of the present invention will be described.

  FIG. 1 is a diagram showing a configuration of a main part of the safe driving monitoring apparatus. The safe driving monitoring apparatus 1 includes a control unit 11, an image processing unit 12, a captured image storage unit 13, and an output unit 14. The control unit 11 controls the operation of each part of the safe driving monitoring apparatus 1 main body.

  The image processing unit 12 processes a captured image input from the camera 2. The camera 2 is a general video camera, and inputs frame images of several tens of frames (for example, 30 frames) per second to the image processing unit 12. The camera 2 is attached so that the target area on the road fits in the imaging area. The image processing unit 12 sequentially processes the frame images input from the camera 12 and detects the person being imaged. The image processing unit 12 also determines whether the detected person is riding a motorcycle, whether the two-wheeled vehicle is a bicycle or a motorcycle, the age estimation of the detected person, and whether the detected person wears a helmet. Judgment about whether or not is also made.

  The captured image storage unit 13 selectively stores the frame image input from the camera 2. The control unit 11 selects a frame image to be stored in the captured image storage unit 13. A speaker 3 is connected to the output unit 14. The output unit 14 outputs an audio signal related to an audio message notified from the speaker 3. The speaker 3 notifies a voice message at a volume that can be heard by a person located within the imaging area of the camera 2. The voice message notified from the speaker 3 is determined by the control unit 11 based on the determination result of the image processing unit 12.

  Hereinafter, the operation of the safe driving monitoring apparatus 1 will be described in detail.

  FIG. 2 is a flowchart showing the operation of the safe driving monitoring apparatus. In the safe driving monitoring apparatus 1, the image processing unit 12 sequentially processes the frame images input from the camera 2. The image processing unit 12 detects a person captured in the frame image to be processed (s1). In s1, feature points of face parts such as eyes, nose and mouth are extracted from the entire frame image to be processed, and a person is imaged when the feature points of these face parts can be extracted. Is determined. If no person can be detected in s1, the image processing unit 12 updates the frame image to be processed to an image after one frame, and performs the process related to s1.

  In s1, if there are a plurality of persons captured in the frame image to be processed, the plurality of persons are detected.

  The safe driving monitoring device 1 determines whether or not the person detected in s1 is on a two-wheeled vehicle (bicycle or motorcycle) (s2). In s2, it determines by detecting whether the tire which is the characteristic of a two-wheeled vehicle is imaged under the person detected by s1. The tire may be detected by, for example, pattern matching with a model image of a motorcycle tire. If the safe driving monitoring apparatus 1 determines that the person is not on the motorcycle in s2, for example, a pedestrian, the process returns to s1.

  If the safe driving monitoring apparatus 1 determines that the person is riding a motorcycle in s2, the safe driving monitoring apparatus 1 determines whether the motorcycle is a bicycle or a motorcycle (s3). In s3, if the width of the tire detected in s2 is shorter than a predetermined length, it is determined that the vehicle is a bicycle. In other words, if the width of the tire detected in s2 is equal to or longer than a predetermined length, it is determined that the vehicle is a motorcycle. If the safe driving monitoring device 1 determines that the vehicle is a motorcycle at s3, the safe driving monitoring device 1 executes processing after s6 described later.

  If the safe driving monitoring device 1 determines that it is a bicycle in s3, it estimates the age of the person detected in s1 (s4). In s4, feature quantities such as the corners of the eyes, a fringe line, a mouth, and a chin are extracted from the face image of the person detected in s1, and the age of the person is estimated using the extracted feature quantities. A technique for processing the captured image of the person and estimating the age and gender of the person uses a neural network and is known in the art, so detailed description thereof is omitted here. If the age estimated in s4 is 13 years or older (if it is not a child / infant) (s5), the safe driving monitoring apparatus 1 returns to s1 and repeats the above-described processing.

  When the safe driving monitoring device 1 determines that the vehicle is an s3 motorcycle, and when the age estimated in s4 is less than 13 years old (if it is a child / infant), the person detected in s1 wears a helmet It is determined whether or not (s6).

  FIG. 3 is a diagram illustrating a process for determining whether the helmet is worn / not worn. FIGS. 3A to 3C show face images of a person wearing a helmet, and FIGS. 3D to 3F show face images of a person not wearing a helmet. Yes. In s6, as indicated by broken lines in FIGS. 3A and 3D, the upper part of the eyes of the person detected in s1 is detected as the head of this person. Further, the width a of the human face detected in s1 is detected. Here, the position for detecting the width of the face is set to the height of the eyes of the person as shown in FIGS. Further, as shown in FIGS. 3C and 3F, the maximum width of the head detected for the person is detected as the width b of the head of the person. Whether or not the safe driving monitoring device 1 is wearing a helmet depending on whether the ratio (b / a) between the width b of the head detected here and the width a of the face is greater than or equal to a predetermined value. Determine. Specifically, if this ratio (b / a) is equal to or greater than a predetermined value, it is determined that a helmet is worn, and if this ratio (b / a) is less than a predetermined value, the helmet Determine that you are not wearing.

  Note that the above-described face width a and head width b may be detected in the captured image without considering the imaging magnification of the camera 2 or the like.

  Further, in order to prevent erroneous determination due to the hairstyle, the determination relating to s6 may be performed in consideration of pattern matching with the model image of the helmet, gloss of the head, color of the head, and the like.

  If the safe driving monitoring device 1 determines that the helmet is worn in s6, the safe driving monitoring device 1 returns to s1 and repeats the above-described processing. On the other hand, if it determines with not wearing a helmet by s6, the frame image which was the process target this time will be memorize | stored in the captured image memory | storage part 13 (s7). Moreover, the safe driving monitoring apparatus 1 outputs the audio | voice signal concerning the audio | voice message which urges | wears a helmet from the output part 14 as a warning output with respect to the person who determined not wearing the helmet this time (s8). This audio signal is input to the speaker 3. As a result, a voice message that prompts the person who has determined not to wear the helmet this time to wear the helmet is notified. In s8, a voice message is selected according to the determination result of s3 described above. For example, if it is determined that it is a bicycle in s3, a voice message such as “Because it is dangerous, wear a helmet when riding a bicycle” is selected, and if it is determined that it is a motorcycle in s3 Selects a voice message, such as “You are required to wear a helmet under the Road Traffic Law.”

  As a result, it is possible to urge children and infants riding on bicycles to wear helmets. It can also warn people who ride a motorcycle without wearing a helmet in violation of the Road Traffic Law. Therefore, it is possible to increase the safety awareness of the person riding the motorcycle and improve the helmet wearing rate.

  The safe driving monitoring device 1 may have a function of detecting whether or not the bicycle is a two-seater when it is determined that the bicycle is on board in s3. For example, as shown in FIG. 4, the guardian may have a function of detecting whether or not the bicycle is carrying two children and infants. This function can be realized by a function of determining whether or not there are two detected faces for one bicycle. In the case of a two-seater bicycle, if at least one of them is a child / infant who does not wear a helmet, it may be configured to output a voice signal related to a voice message prompting the user to wear the helmet in s8. . In this case, for example, a voice message such as “Let's wear a helmet for the safety of children” may be selected.

  In addition, if the two people who are riding the bicycle are neither children nor infants, a voice message notifying that the two-seater bicycle is prohibited by the Road Traffic Law may be notified from the speaker 3. Good. For example, a voice message such as “two-seater bicycles are prohibited by the Road Traffic Law.

  Furthermore, a function of detecting a person riding a bicycle while talking on a mobile phone or a bicycle whose headlight is not turned on at night may be added by processing captured images. In this case, select a voice message such as "Use of a mobile phone is dangerous. Stop the bicycle and talk." Or "Let the headlights turn on for safety." What is necessary is just to set it as the structure alert | reported from the speaker 3. FIG.

  Furthermore, since the captured image of the camera 2 that caused the voice message to be notified to the detected person is stored in the captured image storage unit 13, a school that promotes wearing a helmet when riding a bicycle, This captured image can also be provided to a local government or the like.

1-Safe driving monitoring device 2-Camera 3-Speaker 11-Control unit 12-Image processing unit 13-Captured image storage unit 14-Output unit

Claims (3)

Image processing means for processing a picked-up image of an image pickup device that picks up a road, and detecting a face and a head of a person riding a two-wheeled vehicle traveling on the road;
Helmet wearing determination means for determining whether or not a helmet is worn based on the ratio of the width of the face of the person riding on the two-wheeled vehicle detected by the image processing means and the width of the head;
A safe driving monitoring device comprising: output means for outputting a warning when the helmet wearing judgment means judges that a helmet is not worn. From the face image of the person riding on the two-wheeled vehicle detected by the image processing means, an age estimation means for estimating the age of the person is provided,
The safe driving monitoring apparatus according to claim 1, wherein the output unit does not perform the warning output if the age estimated by the age estimating unit is equal to or greater than a predetermined age. The image processing means is means for detecting whether the two-wheeled vehicle traveling on the road is a bicycle or a motorcycle,
The output means is means for outputting the warning even when the age estimated by the age estimating means is equal to or higher than a predetermined age when the image processing means detects a motorcycle as a motorcycle. The safe driving monitoring apparatus according to 2.

Images