JP6089456B2 - Driving behavior diagnostic device, driving behavior diagnostic method, and driving behavior diagnostic program - Google Patents

Description

  The present invention relates to a driving behavior diagnostic device, a driving behavior diagnostic method, and a driving behavior diagnostic program for evaluating a driving behavior of a driver of a vehicle.

  In recent years, there are an increasing number of cases in which a camera is mounted inside a vehicle for the purpose of constantly recording the situation ahead of the vehicle and assisting safe driving. Furthermore, there is a need to analyze the video obtained from the in-vehicle camera and evaluate whether the driver is driving correctly at the point where the road sign and the road sign are installed.

  The “road sign” is a display device in which figures and characters are written and is installed at a predetermined height from the road. For example, road signs include signs that require a vehicle to stop or stop at a predetermined position. The point where the temporary stop is required refers to an intersection where the temporary stop is designated, an intersection where a red light is displayed, and the like. The point where the stop is required refers to, for example, a railroad crossing where a train passes soon. Hereinafter, the pause and stop are simply referred to as “stop”.

  The “road marking” is a figure directly displayed on the road surface with a white line or the like. Some road markings, for example, indicate maximum speed and prohibition of turning (U-turn).

  Hereinafter, “road sign” and “road marking” are collectively referred to as “instruction structure”. As described above, the structure instructs the driver to perform a predetermined driving action on a road in a predetermined position or a predetermined range (hereinafter simply referred to as “instructed position”). . “Driving action” means an action taken by the driver to control the speed or course of the vehicle. Hereinafter, a driver who is subject to a diagnosis of driving behavior is referred to as “diagnostic person”, and a vehicle in which the driving behavior of the subject of diagnosis is performed by the subject is referred to as “target vehicle”.

  It is very inefficient to visually evaluate the driving behavior of the person to be diagnosed as described above using a normal video reproduction tool. For example, when evaluating the driving behavior corresponding to the temporary stop sign, the following work is required.

  (1) Play the video from the in-vehicle camera and search for a scene that needs to be paused.

  (2) In the scene, the video frame corresponding to the position to be paused is specified.

  (3) In the vicinity of the frame specified in (2), it is checked whether or not the vehicle speed is 0 with reference to the vehicle speed information, and the presence or absence of a temporary stop is totaled.

  To perform such an operation, it is necessary to play back all the recorded videos. Therefore, the real time, that is, the time equivalent to the length of the video is required.

  As a means for making this work more efficient, it is conceivable to automatically extract a point where a driving action such as a temporary stop is required by a structure using an image recognition technique. That is, by detecting a road sign or the like instructing a temporary stop from the image, the operations (1) and (2) can be automated and the entire processing can be made efficient.

  Various devices for evaluating the driver's driving behavior have been proposed.

  For example, Patent Literature 1 discloses a driving diagnosis information providing apparatus that can diagnose whether or not a driver has paused a vehicle at a location where the vehicle should be paused. The device of Patent Document 1 determines whether or not it is a place to be temporarily stopped by using traffic information for VICS (Vehicle Information and Communication System Center (registered trademark)) service. Then, real-time driving diagnosis and driving tendency diagnosis are performed on the driver.

  Patent Document 2 discloses an image recognition device that can improve the recognition accuracy of a stop line that once exists at a stop point. In the apparatus of Patent Document 2, only a stop line at a stop point is extracted by image recognition of a combination of a stop line that once indicates a stop point and a character string adjacent to the stop line.

  Patent Document 3 also discloses a road marking recognition device that reduces misrecognition and recognition failure of road markings. In the apparatus of Patent Document 3, it is determined whether the mutual positional relationship between the first road marking and the second road marking is normal based on the position data of the first road marking and the second road marking.

JP 2010-038645 A (page 6-9, FIG. 11) JP 2009-099125 A (page 9-10, FIG. 1) JP 2007-052645 A (page 9-10, FIGS. 16 and 17)

  However, image recognition by image processing may cause false detection. Therefore, as in Patent Documents 1, 2, and 3, when driving behavior diagnosis as described above is performed using image recognition technology, there is a problem that the driving behavior of the diagnosed person may not be correctly evaluated. To do.

  In addition, the devices of Patent Literatures 1, 2, and 3 do not have a function that can confirm at a glance the appropriateness of the driving behavior at the predetermined structure and the designated position specified thereby. Below, the subject of each apparatus of patent documents 1, 2, and 3 is shown concretely.

  With the apparatus of Patent Document 1, real-time driving diagnosis and driving tendency diagnosis are possible. The real-time driving diagnosis by the device of Patent Document 1 is presentation of advice to a driver who is driving. Since the driver is driving, the validity of the diagnosis itself, confirmation of the reason for diagnosis, and details of the driving behavior in which the problem has occurred cannot be confirmed in detail. On the other hand, the driving tendency diagnosis is literally the presentation of the results of analyzing the driving tendency, and in the same way as the real-time driving diagnosis, confirm the details of the validity of the diagnosis, the reason for the diagnosis, and the content of the driving behavior that had the problem. I can't.

  Furthermore, the apparatus of Patent Document 1 uses traffic information for VICS service. Therefore, the vehicle needs to be equipped with a VICS information receiving device that receives VICS information and a car navigation device that includes map information. However, even if the vehicle is equipped with a VICS information receiving device, it is not always possible to receive VICS information transmitted by radio waves or optical signals. In the first place, information on whether or not it is a stop point is not provided for all roads in the country by the VICS service. As described above, there are various restrictions in the evaluation of whether or not there is a temporary stop at the stop point, for example, using the apparatus of Patent Document 1.

  In the device of Patent Document 2, the recognition accuracy of a stop line once at a stop point is improved by image recognition of a combination of a stop line that once indicates a stop point and a character string adjacent to the stop line. However, when the device of Patent Document 2 is used for driving diagnosis, either a method of performing real-time diagnosis during driving or a method of recording all images taken by an in-vehicle camera and checking all recorded images after driving is completed. It is necessary to take. Regarding the real-time diagnosis, there is a problem that the validity of diagnosis, the reason for diagnosis, the content of the operation operation in which the problem has occurred, and the like cannot be confirmed in detail, as in the technique of Patent Document 1. The method of confirming the recorded image of the in-vehicle camera has a problem that it is very inefficient because it must confirm the recorded image of the entire operation time.

The apparatus of Patent Document 3 reduces misrecognition and recognition omission of road markings by making a determination using the positional relationship between two road markings. However, Patent Document 3 also needs to adopt either a method for performing a real-time diagnosis during driving or a method for confirming a recorded image of a vehicle-mounted camera when used for driving diagnosis. There is a problem.
(Object of invention)
The present invention has been made in view of the technical problems as described above, and without using a special on-vehicle device such as a car navigation device or a receiving device, the result of diagnosis of driving behavior and confirmation of its validity can be confirmed. An object of the present invention is to provide a driving behavior diagnostic apparatus, a driving behavior diagnostic method, and a driving behavior diagnostic program that can be efficiently performed.

  The driving behavior diagnosis apparatus of the present invention is an information input for reading driving behavior information indicating a driving behavior of a subject with respect to the target vehicle, including a recorded image of the foreground of the subject vehicle taken from the subject vehicle driven by the subject. And means for detecting a structure instructing the person to be diagnosed to perform a predetermined driving action at a predetermined position from the recorded image and including identification information of the image in which the structure is detected. A first detection means for outputting a first detection result; and a second detection means for detecting a designated position from the recorded images and outputting a second detection result including identification information of the image from which the designated position is detected. Based on the detection means and the driving behavior information, the judgment means for determining the presence or absence of the driving behavior of the diagnosed person at the indicated position and outputting the determination result; the information input means; the first detection means; the second detection Start the means and judgment means First detection means, characterized in that it comprises a control means for inputting the driving behavior information to the second sensing means and determining means.

  The driving behavior diagnosis method of the present invention includes a foreground recorded image of a target vehicle captured from the target vehicle, which is included in the driving behavior information indicating the driving behavior of the diagnostic subject with respect to the target vehicle driven by the diagnostic subject. Detecting a structure instructing the subject to perform a predetermined driving action at a predetermined designated position, outputting a first detection result, and detecting the designated position from the recorded image The second detection result is output, the presence / absence of the driving behavior of the diagnosed person at the indicated position is determined based on the driving behavior information, and the determination result is output.

  The driving behavior diagnosis program of the present invention includes a computer included in the driving behavior diagnosis device, and includes a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnostic subject, and the driving behavior of the diagnostic subject with respect to the target vehicle. An information input means for reading driving behavior information indicating a structure and a structure detected from a recorded image that instructs the subject to perform a predetermined driving behavior at a predetermined designated position. A first detection means for outputting a first detection result including identification information of the image that has been detected; a second that includes the identification information of the image from which the indicated position is detected by detecting the indicated position from the recorded image; A second detection unit that outputs the detection result, a determination unit that determines the presence or absence of the driving behavior of the diagnosed person at the indicated position based on the driving behavior information, and outputs a determination result; an information input unit; 1's Knowledge means activates the second detection means and determination means, the first detection means and thereby functions as a control means for inputting the driving behavior information to the second sensing means and determination means.

  According to the present invention, it is possible to efficiently check the diagnosis result of driving behavior, its validity, and the content of driving behavior.

  That is, according to the present invention, it is possible to obtain a detection result of a structure that indicates a predetermined driving action and an indicated position from an image of an in-vehicle camera of a vehicle that is driven by a diagnosis subject. In addition, a determination result of the presence or absence of the instructed driving action at the instructed position is also obtained. Therefore, the person to be diagnosed can easily refer only to the image in which the structure is detected. Therefore, the person to be diagnosed can efficiently check the diagnosis result, the validity of the diagnosis result, and the content of the driving action.

It is a block diagram which shows the structure of the driving | operation diagnosis apparatus of the 1st Embodiment of this invention. It is a figure which shows an example of how to memorize | store the information which shows the presence or absence of the structure detection result information, stop line detection result information, and the presence or absence of the event of a temporary stop. It is explanatory drawing of the measuring method of the distance using the principle of triangulation. It is a figure which shows the state transition of a detection target, and its conditions. It is a figure explaining the internal structure of an image signal output means. It is a figure which shows an example of the information display example by an image signal output means. It is a figure which shows the flowchart which shows operation | movement of the driving behavior diagnostic apparatus of embodiment of this invention. It is a figure which shows the mode of the change of the detection target on the time-axis by a control means. It is a figure which shows the mode of the change of the detection target on the time-axis by the control means by another control rule. It is a figure which shows the hardware constitutions of this Embodiment. It is a block diagram which shows the structure of the driving | operation diagnosis apparatus of the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the driving | operation diagnosis system of the 2nd Embodiment of this invention. It is a flowchart which shows operation | movement of the information display apparatus 903 in the 2nd Embodiment of this invention.

(First embodiment)
Embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram illustrating a configuration of the driving behavior diagnosis apparatus according to the first embodiment. The driving behavior diagnosis apparatus 100 according to this embodiment includes an information input unit 101, an instruction structure detection unit 102, a stop line detection unit 103, a determination unit 104, a control unit 105, and an image signal output unit 106.

  The information input means 101 reads the in-vehicle camera video and vehicle speed data recorded in advance on a recording medium and outputs them to the control means 105 described later. The “in-vehicle camera video” is obtained by capturing and recording a foreground of a vehicle driven by a person to be diagnosed, that is, a foreground of a target vehicle having a driving action to be diagnosed. The in-vehicle camera video is composed of frames including image data corresponding to each time of a predetermined time interval. “Vehicle speed data” is data indicating the speed in the front-rear direction of the target vehicle, that is, the speed.

  The vehicle-mounted camera image and the vehicle speed data exist separately, and the time when each is acquired and measured, in other words, the data may be associated with each other and recorded. Alternatively, the vehicle speed data may be information in which information indicating the vehicle speed measured at the same time is embedded in an image of each frame. In short, as long as the vehicle speed data represents the vehicle speed of the target vehicle at the time when each image in the image data was taken, the format of the data is arbitrary.

  The pointing structure detection unit 102 detects a structure (pointing structure) that instructs to stop, including a temporary stop, using an image recognition technique from the image, and outputs structure detection result information. “Structure detection result information” refers to the identification information of the frame to be processed, the presence or absence of the pointing structure in the video of the frame, and the detection position of the pointing structure in the image when the pointing structure is detected It is. The detection position of the pointing structure can be expressed by, for example, coordinates of four corners of a rectangle surrounding the pointing structure. The pointing structure is, for example, a stop sign, a red light, a railroad crossing, or the like. Each of these is paired with a stop line indicating a position to be stopped.

  The object of detection of the pointing structure is a stop sign, a red light, a railroad crossing, and the like. For example, in Japan, stop signs and red signal detection can be based on existing methods such as generalized Hough transforms that detect red inverted triangles and red high-luminance circular regions, and pattern matching methods. it can. The level crossing can also be detected by a pattern matching method for a red high-luminance circular region or a yellow and black striped region.

  The pointing structure detection unit 102 is activated by the control unit 105 described later, and the control unit 105 is activated when the detection target is the pointing structure.

  The stop line detection means 103 detects a stop line from the images in the in-vehicle camera video using an image recognition technology, and outputs stop line detection result information. “Stop line detection result information” is the identification information of the frame to be processed, the presence / absence of the stop line in the video of the frame, and the detection position in the image of the stop line when the stop line is detected. . The detection position of the stop line can be expressed by, for example, coordinates of four corners of a rectangle surrounding the stop line. The detection of the stop line can be performed based on an existing image recognition method such as Hough transform.

  The stop line detection means 103 is activated by the control means 105 described later, and the control means 105 is activated when the detection target is a stop line.

  The determination unit 104 determines whether or not the target vehicle has stopped from vehicle speed data at a time corresponding to the currently focused frame in the in-vehicle camera video, and identifies the target frame identification information and whether or not there is a stop. The judgment result indicating is output. Whether or not the vehicle has stopped can be determined from whether or not the vehicle speed data is a value equal to or less than a predetermined threshold value.

  The determination unit 104 is activated by the control unit 105 described later, and the control unit 105 is activated when the detection target is the presence or absence of a stop event.

The control means 105 is responsible for the activation of the information input means 101, the pointing structure detection means 102, the stop line detection means 103, and the determination means 104, and the recording and input / output of information exchanged with these means. When starting each of the information input unit 101, the pointing structure detection unit 102, the stop line detection unit 103, and the determination unit 104, the control unit 105 operates as follows.
1) When activating the information input unit 101 The control unit 105 outputs information indicating a frame to be processed next, and acquires an image and vehicle speed data for the frame from the information input unit 101. The image and the vehicle speed data acquired here become processing target data of the pointing structure detection unit 102, the stop line detection unit 103, and the determination unit 104.
2) When activating the pointing structure detection unit 102 The control unit 105 outputs the image acquired from the information input unit 101 to the pointing structure detection unit 102. Then, the control unit 105 acquires the structure detection result information from the pointing structure detection unit 102, that is, information indicating the presence / absence of the pointing structure. Further, when the pointing structure detection unit 102 detects the pointing structure, the control unit 105 also acquires position information of the pointing structure in the image where the pointing structure is detected.
3) In the case of starting the stop line detection means 103 In this case as well, similarly, the control means 105 outputs an image to the stop line detection means 103 and acquires stop line detection result information, that is, information indicating the presence or absence of the stop line. . When the stop line detecting unit 103 detects a stop line, the control unit 105 also acquires position information of the stop line in the image where the stop line is detected.
4) When activating the determination unit 104 The control unit 105 outputs vehicle speed data to the determination unit 104, and acquires information indicating whether or not the target vehicle is stopped.

  As shown in FIG. 2, the structure detection result information, the stop line detection result information, and the stop presence / absence determination result are collectively stored for each portion to be stopped. When the stop line cannot be detected, the fact is expressed in the stop line detection result information.

  The control means 105 has a function of calculating the movement amount of the target vehicle in an arbitrary frame section by integrating the vehicle speed data with time. For example, the frame interval for calculating the movement amount may be determined from a frame in which the stop line detection unit 103 and the determination unit 104 are continuously activated to a frame at a certain time point. In this case, it is possible to calculate the moving distance after starting the stop line detecting means 103 continuously and the moving distance after starting the determining means 104 continuously. Alternatively, as the “second frame section”, for example, from the video frame in which the pointing structure is detected to the frame corresponding to the distance traveled by the target vehicle by 20 m may be determined. In this case, the currently input frame is included in the second frame section by comparing the moving distance after starting the stop line detecting means 103 continuously with the distance 20 m corresponding to the second frame section. It can be easily determined whether or not.

  The control means 105 may store camera internal / external parameter information. In this case, the control unit 105 can also calculate the distance from the in-vehicle camera to the pointing structure based on the structure detection information output from the pointing structure detection unit 102. The distance from the in-vehicle camera to the pointing structure is determined from the relationship between the internal parameters of the in-vehicle camera and the size of the object detected in the image using the standardized size and shape of the object. Is possible.

For example, assuming that the optical system of the camera can be expressed by a perspective projection model, the length of one side in the real world of a stop sign having an inverted triangle is 0.6 [m], and a pixel on one side of the stop sign in the image When the number is 60 [pixels], the inter-pixel pitch of the image sensor of the in-vehicle camera is 0.01 [mm], and the focal length of the optical system of the in-vehicle camera is 0.005 [m], the distance x from the in-vehicle camera to the pause indication is
60 × 0.00001: 0.6 = 0.005: x
Than,
x = 5 [m]
Can be calculated.

  If the distance between the in-vehicle camera and the vehicle front end is stored in the control means 105 in advance, the distance x can be converted to the distance from the vehicle front end to the temporary stop sign.

  Similarly, when the pointing structure is a red signal or a display of a railroad crossing breaker, the distance x from the in-vehicle camera to the pointing structure can be similarly determined from the diameter of the known circular lighting part.

  When the absolute size of the target object is unknown, it may be estimated from a change in the moving distance of the target vehicle and the length of the target object on the image in a certain time section.

For example, the length of the object on the image at the time interval t [s], the movement distance L [m] of the target vehicle during the time interval, and the time at both ends of the time interval is set to a1 [pixel] and a2 [pixel], respectively. And the distance x to the object at the end of the time interval is
x = L × a1 / (a2-a1)
It can also be expressed as

  The control unit 105 determines that the distance between the front end of the target vehicle and the stop line is within a predetermined range based on the result of the distance estimation result to the stop line and the movement distance calculation principle based on the above principle. It is possible to define a “section”. The first frame section is a range of the in-vehicle camera video indicating that the target vehicle has approached within a certain distance to the stop line. What is necessary is just to determine within 1 m, for example within the predetermined range.

  Further, based on the information indicating the stop line position included in the stop line detection result information output by the stop line detection means 103, the distance from the in-vehicle camera to the stop line can be measured based on the triangulation principle.

  FIG. 3 shows a method for measuring the distance from the in-vehicle camera to the stop line using the principle of triangulation. However, in order to simplify the explanation, it is assumed that the target vehicle is in contact with the horizontal ground and the angle (elevation angle) of the shooting direction of the camera is the horizontal direction.

  As can be seen from FIG. 3, from the height H from the road surface of the vehicle-mounted camera in the real world and the angle θ between the vertical direction and the detection direction of the stop line, the distance from the vehicle camera to the stop line is H · tan ( θ).

  By the way, the position of the image of the photographing object in the camera image is determined by the focal length of the camera, the inter-pixel pitch of the image sensor, the image size, and the direction of the photographing object. Since the focal length of the camera, the pitch between pixels of the image sensor, and the image size can be regarded as fixed values, the direction of the shooting object should be calculated if the position of the image of the shooting object in the camera image is determined Can do. Therefore, the relationship between the angle θ formed with the detection direction of the object to be photographed and the vertical coordinate value (y coordinate value) in the image of the object to be photographed is previously associated using a table or the like. be able to. The angle θ can be obtained by reading from the table the direction of the object to be photographed corresponding to the information indicating the stop line position included in the stop line detection result information output by the stop line detection means 103.

  Furthermore, the control means 105 sets only one detection target in the current frame from the past structure detection result information, stop line detection result information, and stop presence / absence determination result. Then, the control unit 105 determines a detection target for the next frame based on the structure detection result information, the stop line detection result information, or the stop presence / absence determination result acquired for the current frame. Here, the detection target refers to any of an indication structure, a stop line, and an event indicating whether or not there is a stop. Processing corresponding to each detection target is performed by the pointing structure detection unit 102, the stop line detection unit 103, and the determination unit 104, respectively.

  Hereinafter, the detection target setting method will be described in detail with reference to the drawings. FIG. 4 is a diagram illustrating an example of a transition rule to be detected.

  In the initial state, the driving behavior diagnosis apparatus of the present embodiment is in a state (state 0) in which the detection target is the pointing structure. In the state 0, when the pointing structure detecting unit 102 is activated and the pointing structure is detected, the driving behavior diagnosis apparatus 100 sets the detection target for the next frame as a stop line (state) as in (1). 1).

  In the state where the detection target is a stop line (state 1), when the stop line detection unit 103 detects the stop line, the driving behavior diagnosis apparatus 100 sets the detection target for the next frame as the target vehicle as shown in (2). Is set to a state (state 2) that indicates whether or not to stop.

  In the state where the detection target is whether or not there is a stop (state 2), when the determination unit 104 detects a stop, or when it is considered that the tip of the target vehicle has reached the vicinity of the stop line without detecting the stop As shown in (3), the driving behavior diagnosis apparatus 100 is set to state 0 in which the detection target for the next frame is the pointing structure. The fact that the tip of the target vehicle has reached the vicinity of the stop line can be determined by the fact that the processing target frame has reached the last frame of the first frame section.

  In the state where the detection target is a stop line (state 1), when the stop line cannot be detected while traveling a certain distance, as shown in (4), the state is changed to state 0 where the detection target of the next frame is the pointing structure. Transition. The fact that the stop line could not be detected while traveling a certain distance can be determined by the fact that the frame to be processed has reached the last frame of the second frame section.

  The image signal output means 106 is an output means for allowing a person to be diagnosed or a diagnostician who communicates a diagnosis result to the person to be diagnosed to determine the appropriateness of the stop at a glance, and includes three means. The three means are a structure detection result display means 401, a stop line detection result display means 402, and a vehicle speed information display means 403 as shown in FIG.

  The structure detection result display unit 401 displays an image of a frame including the pointing structure detected by the pointing structure detection unit 102 together with the detection position of the pointing structure. The stop line detection result display unit 402 displays the video of the frame including the stop line extracted by the stop line detection unit 103 together with the detection position of the stop line. The vehicle speed information display unit 403 displays a temporal change in the vehicle speed in the vicinity of the frame for which the presence or absence of the stop is determined by the determination unit 104.

  An example of a specific information display screen by the image signal output means 106 is shown in FIG. The display screen of FIG. 6 displays the presence or absence of a temporary stop at a temporary stop intersection in an easy-to-understand manner, and includes a sign detection frame image 501, a stop line detection frame image 502, and vehicle speed information 503.

  The sign detection frame image 501 and the sign detection result 504 are drawn by the structure detection result display means 401. The sign detection result 504 is displayed in the sign detection frame image 501 by surrounding the road sign with a rectangle.

  The stop line detection frame image 502 and the stop line detection result 505 are drawn by the detection line detection result display unit 402. The stop line detection result 505 is displayed by surrounding the road marking with a rectangle in the stop line detection frame image 502.

  The vehicle speed information 503 is a graph drawn by the vehicle speed information display means 403, the horizontal axis is the distance from the position where the target vehicle is detected to stop (hereinafter referred to as “stop detection position”), and the vertical axis is the target vehicle. Speed. The vehicle speed information 503 includes a vehicle speed curve 509 indicating the speed of the target vehicle and line segments indicating the frame position information 506, 507, and 508.

  A vehicle speed curve 509 indicates a vehicle speed from when the pointing structure is detected to when the first frame section ends.

  The frame position information 506 and 507 indicate the frame where the pointing structure and the stop line are detected by the distance from the stop detection position, respectively. The frame position information 508 indicates a stop detection position, that is, a position at a distance of 0 m.

  The image signal output means 106 may display the image recognition result and the vehicle speed data one by one as shown in FIG. 6 for the place where the stop is detected or the place where the stop is not stopped. Alternatively, a plurality of results may be displayed side by side on one screen. In order to display the image recognition result and vehicle speed data of other places that are not displayed, the driving behavior diagnosis apparatus 100 may be provided with an input device such as a keyboard so that the location to be displayed can be selected using the input device. .

  Moreover, not only the location where the stop was detected and the location where the stop was not detected, but also the location where the indicating structure was detected but the corresponding stop line could not be detected may be displayed. In this case, since the frame to be displayed as the stop line detection result is not determined, for example, an image of a frame in which it is estimated that the pointing structure is on the extension line of the front end of the target vehicle may be displayed. Further, as the time series information, a change in vehicle speed data from when the pointing structure is detected until the target vehicle travels a certain distance may be displayed.

  The image signal output means 106 detects the image recognition result for the frame immediately before changing the detection target, or the indication structure is the front end of the target vehicle when the stop line is not detected at the position where the stop is detected and the position where the stop is not stopped. An image of a frame that is assumed to be on the extension line of is displayed. Furthermore, it is also possible to use a keyboard so that the image recognition results for the frames before and after the displayed frame and the image of the frame can be replaced. In this way, particularly when a stop line cannot be detected, whether or not a stop line really exists can be confirmed by referring to the images of the previous and subsequent frames.

  Next, the operation of this embodiment will be described in detail with reference to the drawings. FIG. 7 is a flowchart showing the operation of the driving behavior diagnosis apparatus 100 of the present embodiment. This flowchart is equivalent to the control order of the control means 105. In the initial state, the detection target is set to the pointing structure.

  First, as the first process for each frame of data, the information input means 101 is activated to acquire an image and vehicle speed data (step S1).

  Next, in order to determine the timing for switching the detection target to the pointing structure again after the detection target becomes a stop line, the moving distance of the vehicle from the timing when the detection target is set as the current detection target to the present is obtained. (Step S2). This is because if the stop line cannot be detected because the stop line is unclear, it may not be possible to exit the state where the detection target is the stop line (state 1). Even in such a case, by obtaining the moving distance of the vehicle, it is possible to detect that the moving distance has become a certain value or more and get out of the stop line detection mode.

  Note that the timing at which the detection target is set is the timing at which the detection target is changed from the target before the change to the current detection target, and specifically, steps S9, S10, S14, and S16 described later. In the initial state, the time when this operation is activated may be regarded as the timing when the detection target is set.

  Next, it is determined whether or not the detection target for the current frame is a pointing structure (step S3).

  When the detection target is the pointing structure (step S3: Yes), the pointing structure detection means 102 is activated (step S4).

  If the detection target is not the pointing structure (step S3: No), it is determined whether the detection target is a stop line (step S5).

  If the detection target is a stop line (step S5: Yes), the stop line detection means 103 is activated (step S6).

  After executing Step S4 or Step S6, the control means 105 stores the structure detection result information or the stop line detection result information (Step S7).

  Next, it is determined whether or not a detection target, that is, a pointing structure or a stop line has been detected (step S8), and if detected, the detection target is changed (step S9). Specifically, when the current detection target is a pointing structure, the detection target is changed to a stop line. When the current detection target is a stop line, the detection target is changed depending on whether the target vehicle is stopped.

  If not detected (step S8: No), the detection target is reviewed (step S10). That is, if the frame being processed is the last frame of the second frame section, the detection target is changed from the stop line to the pointing structure. If the frame being processed is not the last frame of the second frame section, the detection target remains a stop line.

  After execution of step S9 and step S10, the presence / absence of unprocessed data is determined (step S17), and if there is unprocessed data, the process returns to step S1 to process the next frame.

  In step S5, when the detection target is not a stop line (step S5: No), it means that the detection target is the presence or absence of a stop of the target vehicle. In this case, the determination means is activated (step S11), and whether or not there is a stop is determined (step S12).

  If a stop is detected (step S12: Yes), a stop presence / absence determination result is stored (step S13). Then, the detection target is changed to the pointing structure (step S14), and if there is unprocessed data (step S17: Yes), the process returns to step S1.

  If no stop is detected in step S12 (step S12: No), it is verified whether it is the last frame of the first frame section (step S15).

  If the frame being processed is the last frame of the first frame section (step S15: Yes), it is determined that the target vehicle has not stopped. In this case, the fact that the target vehicle has not stopped is output, and the detection target for the next frame is changed to the pointing structure (step S16). If there is unprocessed data (step S17: Yes), the process returns to step S1.

  If the frame being processed is not the last frame of the first frame section (step S15: No), if there is unprocessed data (step S17: Yes), the process returns to step S1.

  After execution of steps S9, S10, S14, and S16, if there is no unprocessed data in step S17 (step S17: No), the image signal output means 106 is activated to display information on the result of image recognition and changes in vehicle speed data. (Step S18).

  As described above, the control unit 105 in the present embodiment changes the detection target in the next frame when the pointing structure detection unit 102 and the stop line detection unit 103 detect the pointing structure and the stop line, respectively.

  On the other hand, the control means 105 may once execute a process for detecting the detection target and change the detection target only when the detection target cannot be detected in a certain frame. This will be described with reference to FIGS.

  FIG. 8 is a schematic diagram showing a state of control of the control means 105 in the present embodiment. The horizontal axis is an axis indicating the arrangement of frames, that is, a time axis. The sections 705, 706, and 707 are frame sections that can detect the pointing structure, the stop line, and the stop event when the pointing structure detection unit, the stop line detection unit, and the determination unit are activated for each frame, respectively. Show. A section 701 to a section 704 indicate detection targets set by the control unit 105 in the present embodiment for the processing target video, respectively. As can be seen from FIG. 8, in the control means 105 in the present embodiment, the detection target is immediately changed from the next frame in which the pointing structure and the stop line can be detected.

  On the other hand, FIG. 9 is a schematic diagram showing a state in which the detection target is controlled to be changed for the first time in a frame in which the detection target cannot be detected after detecting the detection target for the same image and vehicle speed data. FIG. Sections 801 to 804 indicate detection targets set by the control unit 105 for the processing target video, as in FIG. Since it is assumed that the same image as that in FIG. 8 is a processing target, the sections 705, 706, and 707 are equal to those shown in FIG.

  As can be seen by comparing FIG. 8 and FIG. 9, the detection target is changed from the state in which the pointing structure and the stop line can be detected in FIG. 9 to the next to the frame in which they cannot be detected. Compared to the control method of FIG. 8, the control of FIG. 9 needs to store information of detection results of one frame past.

  However, in the control of FIG. 9, the distance to the stop line can be estimated more accurately by adopting the detection result of the pointing structure and the stop line for the last frame of the sections 705 and 706. Therefore, the control of FIG. 9 has an advantage that it is easier to estimate the frame when the target vehicle passes the stop line with higher accuracy. This is because the target vehicle is closer to the stop line in the last frame of the section 705 and the section 706. Therefore, the estimation of the distance to the stop line by triangulation is not easily influenced by the error of the camera parameter or the error of the detection position of the stop line.

  When the control unit 105 changes the detection target, the image signal output unit 106 displays the structure detection result and the stop line detection result. The structure detection result is displayed for a frame including the pointing structure detected last by the pointing structure detection unit 102, and the stop line detection result includes the stop line detected last by the stop line detection unit 103. Displayed for the frame to be displayed.

  In step S9 in FIG. 7, the control unit 105 of the present embodiment changes the detection target from the stop line to the presence / absence of the stop, and then the determination unit 104 is activated from the next frame.

  On the other hand, the control means 105 may determine whether or not the frame being processed belongs to the first frame section prior to the execution of step S11. If the frame being processed belongs to the first frame section, the stop determination means is activated (step S11). If not, the process may return to step S1 to process the next frame. In this way, the stop determination can be executed only for the frames belonging to the first frame section.

  Further, a display indicating the range of the first frame section may be added to the time-series information display in the image display example by the image signal output means shown in FIG. Thereby, the diagnostician or the diagnosed person can recognize at a glance the frame section in which the determination of the presence or absence of the stop is performed.

  As illustrated in FIG. 10, the driving behavior diagnosis apparatus 100 according to the present embodiment can be configured using a computer system 904 including a PC (Personal Computer) main body 901, a keyboard 902, and a display 903. In the driving behavior diagnosis apparatus 100 having the configuration shown in FIG. 10, the driving behavior diagnosis processing may be realized by causing the PC main body 901 to execute a computer program that realizes the function of the flowchart shown in FIG. 7. Alternatively, some functions may be realized by hardware, and similar functions may be realized by cooperative operation of the hardware and the software program.

  As described above, the driving behavior diagnosis apparatus 100 according to the present embodiment detects an instruction structure and a stop line for instructing a stop from an in-vehicle camera image in which a foreground of a target vehicle driven by a person to be diagnosed is recorded. After detecting the stop line, it is determined whether or not the target vehicle is stopped. And the detection result of a pointing structure and a stop line, and the judgment result of the presence or absence of the stop of a target vehicle are output as an image. Further, the frame that is the detection target is also specified.

  Therefore, a diagnostician who receives a diagnosis of driving behavior or a diagnostician who performs the diagnosis can easily find an image related to the diagnostic of driving behavior, such as whether or not the suspension is executed. Moreover, the content of actual driving behavior can also be confirmed. Therefore, there is an effect that it is possible to efficiently confirm the diagnosis result of driving behavior, its validity, and the specific content of driving behavior.

  In the embodiment, the evaluation of the driving behavior of stopping the target vehicle based on a structure that instructs a temporary stop, that is, a so-called “road sign” has been described as an example. The sign for instructing the temporary stop instructs the driver to temporarily stop the target vehicle on the stop line. As is clear from the description of the embodiment, the sign to which the present invention can be applied is not limited to a road sign that instructs a temporary stop.

  For example, it is also possible to evaluate driving behavior based on road signs indicating the maximum speed, overtaking prohibition, entry prohibition, etc., or road surface indications indicating the maximum speed, turn prohibition, etc., so-called “road markings”. That is, the road sign or road sign as described above (hereinafter collectively referred to as “sign etc.”) and the position where the sign etc. is installed are detected as the indicated position, and the driver instructs the sign etc. at the indicated position. It is possible to evaluate whether or not the driving action according to the vehicle has been taken based on the course and speed of the target vehicle.

  When the sign or the like is a sign for instructing a temporary stop, the indicated position where the driving action of the stop is instructed is clearly defined as one point of the stop line installation position. In some types of signs such as the maximum speed of the sign or the overtaking prohibition, the designated position is not defined as one point, but is a predetermined range of the road (hereinafter referred to as “indicated section”). In such a case, first, it is only necessary to detect a sign and the indicated position based on the position of the sign. Then, a sign that changes the instructed driving behavior or a road condition that is judged to be released from the driving behavior instruction may be detected to end the designated section.

  For example, in the case of a sign prohibiting entry, the position where the sign is detected may be determined as the designated position.

  In the case of the maximum speed sign, the range of the road until the next sign indicating a different maximum speed is found becomes the designated section. That is, it can be determined that the maximum speed v1 is indicated from the position at which the sign indicating the highest speed v1 is detected to the position where the mark indicating the different maximum speed v2 is detected.

  In the case of the overtaking prohibition sign, from the position where the sign indicating the overtaking prohibition is first detected to the position where the sign indicating the cancellation of the overtaking prohibition or the road marking indicating that the overtaking prohibition section is not detected, It can be determined as the designated section.

  In the case of a sign that prohibits turning, it may be determined that the position from the position where the sign indicating prohibition is detected first to the position where the intersection is detected is the designated section.

  As described above, even if the designated position is valid in the designated section of the road, the necessary information such as other signs, road conditions, etc. By detecting, the designated section can be determined. Therefore, the present invention can be widely applied to general signs or the like for instructing the driver of a predetermined driving action at the indicated position.

  In addition, the driving behavior diagnosis apparatus 100 may output the detection result of the signs and the installation position and the determination result of the presence / absence of the driving behavior as information as it is. That is, the driving behavior diagnosis apparatus 100 performs detection processing and determination processing, and does not perform screen display of detection results or determination results, generation of image data for screen display, or the like. In this case, the detection result and the determination result are configured and displayed on a screen of a predetermined format by an external information display device. This embodiment will be described later as a second embodiment.

  Alternatively, as shown in FIG. 6, the driving behavior diagnosis apparatus 100 may construct an image including a sign or the like and a detection result of the installation position, and a determination result of the presence or absence of the driving behavior, and output the image data. In this case, the external information display device only needs to display the image data as it is.

  From the above, it can be seen that the driving behavior diagnostic apparatus 110 having only the configuration essential to the present embodiment has the configuration shown in the block diagram of FIG. The driving behavior diagnosis apparatus 110 includes information input means 101, pointing structure detection means 102, pointing position detection means 113, determination means 104, and control means 105.

  The information input means 101 reads driving behavior information. The driving behavior information includes a recorded image of the foreground of the target vehicle, which is taken from the target vehicle driven by the subject. Further, the driving behavior information may include vehicle speed data that is speed information of the target vehicle associated with the shooting time of each frame constituting the recorded image.

  The control means 105 inputs driving behavior information from the information input means 101. In addition, the control unit 105 activates the pointing structure detection unit 102, the pointing position detection unit 113, and the determination unit 104. Then, the control unit 105 inputs driving behavior information to the pointing structure detection unit 102, the pointing position detection unit 113, and the determination unit 104.

  The pointing structure detection unit 102 detects a pointing structure that instructs the diagnosis subject to perform a predetermined driving action at a predetermined pointing position from the recorded image. The pointing structure is, for example, a road sign or road marking that temporarily indicates a stop or a maximum speed, and the pointing position is a position where the structure is set or a position designated by the structure. Then, the pointing structure detection unit 102 outputs the detection result as a first detection result (structure detection result information).

  The designated position detection unit 113 detects the designated position from the recorded image. The designated position detection unit 113 outputs a second detection result (detection result information).

  The indicated position may be a position on the road surface corresponding to the position where the structure is installed. When the structure is for instructing the driving behavior in the designated section including the designated position, the designated position detecting means 113 includes position information indicating that the current position of the target vehicle is within the designated section. 2 detection results are output.

  For example, when a road sign indicating the maximum speed is set, or when a road marking is installed on the road surface, the position where the road sign is installed and the instruction section after that position are detected as the indicated position. The Then, as the second detection result, position information indicating whether or not the target vehicle is in the designated section is output at each time of the recorded image.

  Alternatively, the designated position may be the position of the road marking when the road sign specifies the driving action and the road marking specifies the position where the driving action is performed. That is, when the driving action is designated as a temporary stop at the stop line by a combination of the stop road sign and the stop line road marking as in the above-described embodiment, the position of the road sign is set as the indicated position. Detected.

  The determination unit 104 determines the presence / absence of the driving behavior of the diagnosed person at the indicated position from the recorded image included in the driving behavior information, and outputs the determination result (the stop presence / absence determination result). When the structure is for instructing a driving action in a predetermined instruction section including the instruction position, the determination unit 104 outputs a determination result including vehicle state information indicating the motion state of the target vehicle in the instruction section.

  For example, when the structure is a road sign indicating that turning is prohibited, the determination unit 104 determines whether the target vehicle has turned from the recorded image. And the determination means 104 outputs the presence or absence of turning of a target vehicle as a driving | running state of a target vehicle.

Further, when the driving behavior information includes vehicle speed data, it is possible to determine whether the target vehicle is stopped or the traveling speed of the target vehicle as the driving state. Therefore, it is possible to determine the driving behavior of the diagnosed person with respect to the stop or the maximum speed sign.
(Second Embodiment)
A second embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 12 is a block diagram showing the configuration of the driving diagnosis system according to the second embodiment of the present invention. The driving diagnosis system 120 of this embodiment includes a driving behavior diagnosis device 130 and an information display device 140.

  The driving behavior diagnosis apparatus 130 includes an information input unit 101, an instruction structure detection unit 102, a stop line detection unit 103, a determination unit 104, a control unit 105, and a storage unit 131.

  The information display device 140 includes a storage unit 141, an information input unit 142, and an image signal output unit 143.

  The functions of the information input means 101, the pointing structure detection means 102, the stop line detection means 103, and the determination means 104 included in the driving behavior diagnosis apparatus 130 are included in the driving behavior diagnosis apparatus 100 of the first embodiment. It has the same function.

  The function of the control means 105 of the present embodiment is almost the same as that of the driving behavior diagnosis apparatus 100, but the results of the structure detection result information, the stop line detection result information, and the stop presence / absence determination result are revealed. The only difference is that it is output to the storage means 131 each time.

  The storage unit 131 includes a storage medium such as a hard disk or an SD card connected to the control unit 105 via a network, and has the same structure detection result information, stop line detection result information, and stop as in the first embodiment. The presence / absence determination result is stored.

  The information input unit 143 included in the information display device 140 has the same function as the information input unit 101 included in the driving behavior diagnosis device 130.

  The storage unit 141 may be a hard disk connected via a network. In this case, the storage unit 141 shares the same storage unit as the storage unit 131 in the driving behavior diagnosis apparatus 130. Alternatively, the storage unit 141 may be configured by a storage medium such as an SD card. In this case, the storage unit 141 uses a storage medium such as an SD card as the storage unit 131 extracted from the driving behavior diagnosis apparatus 130.

  The function of the image signal output unit 142 is almost the same as that of the image signal output unit 106 described in the first embodiment. In the second embodiment, in addition to the information from the information input unit 143, the information in the storage unit 141 is used. Also used. That is, the in-vehicle camera video and vehicle speed data from the information input unit 143 and the structure detection result information, the stop line detection result information, and the stop presence / absence determination result from the storage unit 141 are used. Then, by combining these pieces of information, a screen display as shown in FIG. 6 is performed so that the diagnosis person or the person to be diagnosed can judge the appropriateness of the stop at a glance.

  The operation of the driving diagnosis system of the second embodiment will be described.

  The operation of the driving behavior diagnosis system 120 in the present embodiment is nothing but to execute the steps S1 to S16 shown in FIG.

  However, it differs from the first embodiment only in that the storage means 901 is also set as the storage destination in steps S7, S13, and S16 for storing the structure detection result information, the stop line detection result information, and the stop presence / absence determination result. .

  A flowchart showing the operation of the information display device 140 is shown in FIG.

  First, the information display device 140 reads the structure detection result information, the stop line detection result information, and the stop presence / absence determination result from the storage unit 141 (step S21). Next, the information display device 140 reads the image and vehicle speed data in the second frame section at the location to be stopped (step S22). Then, the structure detection result information, stop line detection result information, stop presence / absence determination result, image and vehicle speed data are processed into a screen display as shown in FIG. 6 and displayed (step S23).

  The driving behavior diagnostic device 130 and the information display device 140 in the driving behavior diagnostic device of the second embodiment are configured using a computer system 904 including a PC main body 901, a keyboard 902, and a display 903 as shown in FIG. can do. At this time, information exchange between the driving behavior diagnosis device 130 and the information display device 140 can be realized in various forms.

  For example, the PC main body 901-1 (not shown) constituting the driving behavior diagnosis device 130 and the PC main body 901-2 (not shown) constituting the information display device 140 may be connected via a network. In this case, a computer program capable of realizing the functions of the flowcharts of steps S1 to S16 is executed in the PC main body 901-1 of the driving behavior diagnosis apparatus. In the PC main body 901-2 of the information display device, a computer program capable of realizing the functions of the flowcharts in steps S21 to S23 is executed.

  Or you may perform information exchange between PC main body 901-1 which comprises the driving action diagnostic apparatus 130, and PC main body 901-2 which comprises the information display apparatus 140 using a storage medium. That is, a storage medium such as an SD card is connected as the storage unit 131 to the PC main body 901-1 constituting the driving behavior diagnosis apparatus 130, and information is stored in the storage medium. Then, a storage medium for storing information is connected to the PC main body 901-2 constituting the information display device 140, and predetermined display is performed by the information display device 140.

  As described above, in the driving behavior diagnosis system 120 according to the second embodiment of the present invention, the driving behavior diagnostic device 130 diagnoses driving behavior and outputs the diagnostic result, and the information display device 140 displays the diagnostic result. That is, the driving action diagnosis process is divided into a diagnosis process and a diagnosis result display process, and each process is executed by a separate device. For this reason, the device used for each process has only the minimum necessary functions, and thus the configuration of the device is simplified.

  In addition, in order to refer to the diagnosis result, the person to be diagnosed has only to prepare a display device having only a display function, and there is an effect that it is not necessary to go to a place where the diagnosis device having the diagnosis function is installed.

  The driving behavior diagnosis program of the first embodiment and the driving behavior diagnosis program and information display program of the second embodiment are a semiconductor memory such as a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory, and the like. It may be stored in a non-transitory medium such as a device, an optical disk, a magnetic disk, or a magneto-optical disk.

A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.
(Appendix 1)
Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, a determination unit that determines the presence or absence of the driving behavior of the diagnosed person at the indicated position and outputs a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. Control means to
A driving behavior diagnosis apparatus comprising:
(Appendix 2)
When the structure is to instruct the driving behavior in a predetermined instruction section including the instruction position,
The second detection means outputs the second detection result including position information indicating that the current position of the target vehicle is within the designated section,
The driving behavior diagnosis apparatus according to appendix 1, wherein the determination unit outputs the determination result including vehicle state information indicating a motion state of the target vehicle in the instruction section.
(Appendix 3)
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
3. The driving behavior diagnosis apparatus according to appendix 1 or 2, wherein the determination result includes a third identification number indicating a third frame in which the presence or absence of the driving behavior is determined among the plurality of frames. .
(Appendix 4)
Inserting the first detection result, the second detection result and the determination result into the images of the first frame, the second frame and the third frame, and outputting a diagnosis result image The driving behavior diagnosis apparatus according to Supplementary Note 3, further comprising an image output unit.
(Appendix 5)
The control means has a period in which a distance between a front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames based on the second detection result and the vehicle speed data. Identify the first frame interval,
The image output means includes
Among the plurality of frames, the first detection result is a frame in a second frame section from the first frame to a third frame corresponding to a position where the target vehicle has advanced a predetermined distance. First display means for inserting
Second display means for inserting the second detection result into the second frame;
The driving behavior diagnosis apparatus according to appendix 4, further comprising third display means for inserting a change in the vehicle speed data and the determination result into a part or all of the first frame section.
(Appendix 6)
The first display means inserts the first detection result into the last first frame where the structure is detected,
The second display means inserts the second detection result into the last second frame where the stop line is detected. The third display means adds the first frame to the recorded image. The driving behavior diagnosis apparatus according to appendix 5, wherein a predetermined display indicating that the frame is a section is inserted (appendix 7)
The driving behavior information includes vehicle speed data of the target vehicle associated with the identification number,
The structure instructs the driver to perform the driving behavior for controlling the vehicle speed of the target vehicle at the indicated position,
The determination means determines the presence or absence of the driving behavior of the diagnosis subject at the indicated position based on the vehicle speed data, and outputs the determination result. The driving behavior diagnosis device described.
(Appendix 8)
The structure instructs the driving behavior that instructs the subject to stop for a predetermined period of the target vehicle at the indicated position;
The indicated position is an installation position of a stop line indicating a position on the road surface where the stop should be performed,
The driving behavior diagnosis apparatus according to appendix 7, wherein the determination unit determines whether or not the vehicle is stopped based on the vehicle speed data and outputs the determination result.
(Appendix 9)
The driving behavior diagnostic device according to appendix 2,
The driving behavior information is input, and the first frame, the second frame, and the determination result are displayed based on the first identification number, the second identification number, and the third identification number. A driving behavior diagnosis system including an information display device.
(Appendix 10)
From the recorded image of the foreground of the target vehicle captured from the target vehicle included in the driving behavior information indicating the driving behavior of the diagnostic subject with respect to the target vehicle driven by the diagnostic subject, On the other hand, it detects a structure instructing to perform a predetermined driving action at a predetermined instruction position, and outputs a first detection result,
Detecting the indicated position from the recorded image and outputting a second detection result,
A driving behavior diagnosing method, wherein the presence / absence of the driving behavior of the person to be diagnosed at the indicated position is determined based on the driving behavior information, and a determination result is output.
(Appendix 11)
A computer provided in the driving behavior diagnosis device,
Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, a determination unit that determines the presence or absence of the driving behavior of the diagnosed person at the indicated position and outputs a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. A driving behavior diagnosis program which is made to function as a control means.

DESCRIPTION OF SYMBOLS 100 Driving behavior diagnostic apparatus 101 Information input means 102 Instruction structure detection means 103 Stop line detection means 104 Judgment means 105 Image signal output means 106 Control means 110 Driving behavior diagnostic apparatus 113 Pointed position detection means 120 Driving behavior diagnosis system 130 Driving behavior diagnosis Device 131 Storage means 140 Information display device 141 Storage means 142 Image signal output means 143 Information input means 401 Instruction structure detection result display means 402 Stop line detection result display means 403 Vehicle speed information display means 501 Instruction structure detection result 502 Stop line detection Result 503 Time-series information 504 Pointed structure detection result 505 Stop line detection result 506 Frame that detected pointing structure 507 Frame that detected stop line 508 Frame that detected temporary stop 509 Vehicle speed information 701 Pointed structure Section 702 to be detected 702 Section to be detected as a stop line 703 Section to be detected as to whether or not there is a temporary stop 704 Section to be detected as pointing structure 705 Section of a frame capable of detecting the pointing structure 706 Stop line Detectable frame section 707 Frame section capable of detecting temporary stop 801 Section for indicating designated structure as detection target 802 Section for detecting stop line as target for detection 803 Section for detecting presence / absence of pause 804 Pointed structure 901 PC main body 902 Keyboard 903 Display 904 Computer system

Claims (10)

Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, determining means for determining the presence or absence of the predetermined driving behavior of the diagnosis subject at the indicated position, and outputting a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. and control means for,
When the structure is to instruct the driving behavior in a predetermined instruction section including the instruction position,
The second detection means outputs the second detection result including position information indicating that the current position of the target vehicle is within the designated section,
The determination means outputs the determination result including vehicle state information indicating a motion state of the target vehicle in the instruction section,
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Inserting the first detection result, the second detection result and the determination result into the images of the first frame, the second frame and the third frame, and outputting a diagnosis result image Image output means,
Based on the second detection result and the vehicle speed data, the control unit is configured to change the first frame in a period in which a distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Identify one frame interval,
The image output means includes
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. First display means for inserting the first detection result into a frame in a frame section of
Second display means for inserting the second detection result into the second frame;
A driving behavior diagnosis apparatus comprising: a third display means for inserting the change in the vehicle speed data and the determination result into a part or all of the first frame section . Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, determining means for determining the presence or absence of the predetermined driving behavior of the diagnosis subject at the indicated position, and outputting a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. Control means for
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Inserting the first detection result, the second detection result and the determination result into the images of the first frame, the second frame and the third frame, and outputting a diagnosis result image Image output means,
Based on the second detection result and the vehicle speed data, the control unit is configured to change the first frame in a period in which a distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Identify one frame interval,
The image output means includes
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. First display means for inserting the first detection result into a frame in a frame section of
Second display means for inserting the second detection result into the second frame;
Third display means for inserting the change in the vehicle speed data and the determination result into a part or all of the first frame section.
A driving behavior diagnosis apparatus characterized by the above. The first display means inserts the first detection result into the last first frame where the structure is detected,
The second display means inserts the second detection result into the last second frame where a stop line is detected.
The third display means inserts a predetermined display indicating the frame of the first frame section into the recorded image.
The driving behavior diagnosing device according to claim 1 or 2, characterized in that The driving behavior information includes vehicle speed data of the target vehicle associated with the identification number,
The structure instructs the driver to perform the driving behavior for controlling the vehicle speed of the target vehicle at the indicated position,
The determination means determines the presence or absence of the driving action for the control of the vehicle speed of the diagnosed person at the indicated position based on the vehicle speed data, and outputs the determination result.
The driving behavior diagnosis apparatus according to any one of claims 1 to 3. A driving behavior diagnosis device according to any one of claims 1 to 4,
The driving behavior information is input, and the first frame, the second frame, and the determination result are displayed based on the first identification number, the second identification number, and the third identification number. With information display device
A driving behavior diagnosis system. From the recorded image of the foreground of the target vehicle captured from the target vehicle included in the driving behavior information indicating the driving behavior of the diagnostic subject with respect to the target vehicle driven by the diagnostic subject, On the other hand, it detects a structure instructing to perform a predetermined driving action at a predetermined instruction position, and outputs a first detection result,
Detecting the indicated position from the recorded image and outputting a second detection result,
Based on the driving behavior information, determine the presence or absence of the predetermined driving behavior of the diagnosed person at the indicated position, and output a determination result,
When the structure is to instruct the driving behavior in a predetermined instruction section including the instruction position,
Outputting the second detection result including position information indicating that the current position of the target vehicle is within the designated section;
Outputting the determination result including vehicle state information indicating a motion state of the target vehicle in the indicated section;
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Inserting the first detection result, the second detection result, and the determination result into the images of the first frame, the second frame, and the third frame, and outputting a diagnosis result image ,
Based on the second detection result and the vehicle speed data, a first frame section in a period in which the distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Identify,
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. The first detection result is inserted into a frame in the frame section of
Inserting the second detection result into the second frame;
A change in the vehicle speed data and the determination result are inserted into a part or all of the first frame section.
A driving behavior diagnosis method characterized by the above. From the recorded image of the foreground of the target vehicle captured from the target vehicle included in the driving behavior information indicating the driving behavior of the diagnostic subject with respect to the target vehicle driven by the diagnostic subject, On the other hand, it detects a structure instructing to perform a predetermined driving action at a predetermined instruction position, and outputs a first detection result,
Detecting the indicated position from the recorded image and outputting a second detection result,
Based on the driving behavior information, determine the presence or absence of the predetermined driving behavior of the diagnosed person at the indicated position, and output a determination result,
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Inserting the first detection result, the second detection result, and the determination result into the images of the first frame, the second frame, and the third frame, and outputting a diagnosis result image ,
Based on the second detection result and the vehicle speed data, a first frame section in a period in which the distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Identify,
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. The first detection result is inserted into a frame in the frame section of
Inserting the second detection result into the second frame;
A change in the vehicle speed data and the determination result are inserted into a part or all of the first frame section.
A driving behavior diagnosis method characterized by the above. Inserting the first detection result into the last first frame where the structure is detected;
The second detection result is inserted into the second frame where the stop line is detected.
A predetermined display indicating that the frame is in the first frame section is inserted into the recorded image.
The method for diagnosing driving behavior according to claim 6 or 7. A computer provided in the driving behavior diagnosis device,
Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, determining means for determining the presence or absence of the predetermined driving behavior of the diagnosis subject at the indicated position, and outputting a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. Control means
Function as
When the structure is to instruct the driving behavior in a predetermined instruction section including the instruction position,
Further, the computer
Function as the second detection means for outputting the second detection result including position information indicating that the current position of the target vehicle is within the designated section;
Function as the determination means for outputting the determination result including vehicle state information indicating the motion state of the target vehicle in the indicated section;
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Further, the computer
Inserting the first detection result, the second detection result and the determination result into the images of the first frame, the second frame and the third frame, and outputting a diagnosis result image Function as image output means,
Based on the second detection result and the vehicle speed data, a first frame section in a period in which the distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Function as the control means to identify,
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. Functioning as a first display means for inserting the first detection result into a frame of the frame section of
Function as second display means for inserting the second detection result into the second frame;
It functions as third display means for inserting the change in the vehicle speed data and the determination result into a part or all of the first frame section.
A driving behavior diagnosis program characterized by that. A computer provided in the driving behavior diagnosis device,
Information input means for reading driving behavior information indicating the driving behavior of the diagnosed person with respect to the target vehicle, including a recorded image of the foreground of the target vehicle photographed from the target vehicle driven by the diagnosed person;
First, a structure instructing the subject to perform a predetermined driving action at a predetermined instruction position is detected from the recorded images, and identification information of an image in which the structure is detected is detected. First detection means for outputting a detection result of 1;
Second detection means for detecting the indicated position from the recorded images and outputting a second detection result including identification information of the image at which the indicated position is detected;
Based on the driving behavior information, determining means for determining the presence or absence of the predetermined driving behavior of the diagnosis subject at the indicated position, and outputting a determination result;
The information input means, the first detection means, the second detection means, and the determination means are activated, and the driving behavior information is input to the first detection means, the second detection means, and the determination means. Control means
Function as
The recorded image is composed of a plurality of frames with identification numbers attached thereto,
The first detection result includes a first identification number indicating a first frame in which the structure is detected among the plurality of frames.
The second detection result includes a second identification number indicating a second frame in which the indicated position is detected among the plurality of frames.
The determination result includes a third identification number indicating a third frame in which the presence or absence of the driving action is determined among the plurality of frames.
Further, the computer
Inserting the first detection result, the second detection result and the determination result into the images of the first frame, the second frame and the third frame, and outputting a diagnosis result image Function as image output means,
Based on the second detection result and the vehicle speed data, a first frame section in a period in which the distance between the front end of the target vehicle and the indicated position is within a predetermined range among the plurality of frames. Function as the control means to identify,
Of the plurality of frames, the second frame up to the third frame corresponding to the position where the target vehicle has advanced from the first frame by a predetermined distance, and the presence or absence of the predetermined driving action is determined. Functioning as a first display means for inserting the first detection result into a frame of the frame section of
Function as second display means for inserting the second detection result into the second frame;
It functions as third display means for inserting the change in the vehicle speed data and the determination result into a part or all of the first frame section.
A driving behavior diagnosis program characterized by that.

Images