Driving evaluation device, driving evaluation method, and driving evaluation program

Abstract

To provide a driving evaluation device, a driving evaluation method, and a driving evaluation program that can improve accuracy in evaluating a driver.SOLUTION: A driving evaluation device 1 detects a diamond mark drawn on a road from a captured image; and detects a crosswalk within a predetermined distance from where the diamond mark is detected. Furthermore, when, after the crosswalk within the predetermined distance from the diamond mark is detected and the presence of a target pedestrian who satisfies a predetermined condition is determined and an own vehicle is determined to be in a stop state, and then the own vehicle comes into a start state, the driving evaluation device 1 performs evaluation processing based on whether or not the presence of the target pedestrian that satisfies the predetermined condition is determined.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a driving evaluation device, a driving evaluation method, and a driving evaluation program. [Background technology]

[0002] A driving evaluation device has been proposed that gives a positive evaluation to a driver when the driver gives priority to pedestrians at a crosswalk without traffic lights (see, for example, Patent Document 1). After the vehicle stops, this device analyzes the captured image to detect a crosswalk, and if the driver gives priority to pedestrians at the crosswalk, it determines that this is a considerate scene and adds points to the driver's evaluation. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-95104 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the driving evaluation device described in Patent Document 1 cannot determine whether or not a traffic light is on when a crosswalk is detected. As a result, the driving evaluation device described in Patent Document 1 will give additional points even if the driver simply waits for a pedestrian to cross the crosswalk while waiting for the traffic light to turn. Furthermore, the driving evaluation device described in Patent Document 1 gives priority to pedestrians based on the length of the stop time, so even in a scene where points should be added for giving priority to pedestrians, points will not be added if the stop time is short. As such, the driving evaluation device described in Patent Document 1 has room for improvement in terms of appropriate driver evaluation.

[0005] The present invention has been made to solve such problems, and its purpose is to provide a driving evaluation device, a driving evaluation method, and a driving evaluation program that can improve the accuracy of driver evaluation. [Means for solving the problem]

[0006] The driving evaluation device of the present invention comprises a first detection means that detects a diamond mark painted on a road on which the vehicle is traveling based on an image acquired by an imaging means that images the area ahead of the vehicle; a second detection means that detects a crosswalk within a predetermined distance from the location where the diamond mark is detected by the first detection means based on the image acquired by the imaging means; a target determination means that determines the presence of a target pedestrian who meets predetermined conditions from the image acquired by the imaging means; a start / stop determination means that determines whether the vehicle is in a starting state or a stopped state; and an evaluation means that performs an evaluation process for the driver of the vehicle.The evaluation means performs the evaluation process based on whether the target pedestrian has been determined to be present by the target determination means when the crosswalk is detected by the second detection means, the target determination means has determined the presence of a target pedestrian, and the start / stop determination means has determined that the vehicle is in a stopped state, and then the start / stop determination means determines that the vehicle has entered a starting state.

[0007] The driving evaluation method of the driving evaluation device according to the present invention comprises a first detection step of detecting a diamond mark painted on a road on which the vehicle is traveling, based on an image acquired by an imaging means that images the area ahead of the vehicle; a second detection step of detecting a crosswalk within a predetermined distance from the location where the diamond mark was detected in the first detection step, based on the image acquired by the imaging means; a target determination step of determining the presence of a target pedestrian who satisfies predetermined conditions from the image acquired by the imaging means; a start / stop determination step of determining whether the vehicle is in a starting state or a stopped state; and an evaluation step of performing an evaluation process for the driver of the vehicle. In the evaluation step, the evaluation process is performed based on whether the presence of a target pedestrian has been determined in the target determination step, which is performed when the crosswalk is detected in the second detection step, the presence of a target pedestrian has been determined in the target determination step, and the vehicle has been determined to be in a stopped state in the start / stop determination step, and then the vehicle has been determined to be in a starting state in the start / stop determination step.

[0008] A driving evaluation program according to the present invention is a driving evaluation program for causing a computer to function as the driving evaluation device described above. [Effects of the Invention]

[0009] According to the present invention, it is possible to improve the accuracy of driver evaluation. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing an outline of a driving evaluation device according to an embodiment of the present invention; [Figure 2] 2 is a diagram showing an example of a captured image obtained by the camera shown in FIG. 1. FIG. [Figure 3] This is the first figure showing the driving evaluation method of the driving evaluation device of this embodiment, where (a) shows the state at the first time, (b) shows the state at the second time, (c) shows the state at the third time, and (d) shows the state at the fourth time. [Figure 4] FIG. 2 is a second diagram showing the driving evaluation method of the driving evaluation device according to the present embodiment. [Figure 5] FIG. 10 is a third diagram showing the driving evaluation method of the driving evaluation device according to the present embodiment. [Figure 6] 1 is a first flowchart showing a driving evaluation method according to the present embodiment, illustrating a process related to evaluation at a pedestrian crossing without traffic lights. [Figure 7] 10 is a second flowchart showing the driving evaluation method according to the present embodiment, illustrating processing related to a departure violation. DETAILED DESCRIPTION OF THE INVENTION

[0011] The present invention will be described below in accordance with preferred embodiments. Note that the present invention is not limited to the embodiments shown below and can be modified as appropriate without departing from the spirit of the present invention. In addition, in the embodiments shown below, some components are omitted from illustration and description, but it goes without saying that publicly known or well-known technologies are applied as appropriate to the details of the omitted technologies within the scope of the content described below.

[0012] FIG. 1 is a block diagram showing an outline of a driving evaluation device according to this embodiment. The driving evaluation device 1 shown in FIG. 1 is mounted on a vehicle and evaluates the driving performance of a driver while the vehicle is traveling. This driving evaluation device 1 is configured, for example, as part of the functions of a digital driving recorder. In this embodiment, the driving evaluation device 1 is configured to determine whether a driver has given priority to a pedestrian at a crosswalk without a traffic light, that is, whether the driver stopped the vehicle at the crosswalk without a traffic light and maintained the stopped state until the pedestrian passed through the crosswalk. This driving evaluation device 1 includes a control unit 10, a vehicle speed acquisition unit 20, a storage unit 30, and a communication unit 40.

[0013] The control unit 10 controls the entire driving evaluation device 1, and evaluates the driver based on information on captured images acquired by a camera (imaging means) 2 that captures images ahead of the vehicle, information from the vehicle speed acquisition unit 20, and brake signals. The control unit 10 includes a first detection unit (first detection means) 11, a second detection unit (second detection means) 12, and a third detection unit 13.

[0014] The first detection unit 11 detects a diamond mark (a diamond-shaped road marking, see symbol D in FIG. 3(a) described later) painted on the road on which the host vehicle is traveling, based on the captured image acquired by the camera 2. The first detection unit 11, for example, performs a white line detection process on the captured image to identify the traveling path of the host vehicle. On the identified traveling path, the first detection unit 11 detects a diamond mark painted on the road surface just before a crosswalk without traffic lights, for example, by template matching. Note that the diamond mark may be detected using other methods, such as AI (artificial intelligence), rather than template matching.

[0015] The second detection unit 12 detects crosswalks within a predetermined distance from the location where the diamond mark is detected by the first detection unit 11, based on the captured image acquired by the camera 2. Here, the diamond mark is drawn 30 m and 50 m before the crosswalk without a traffic light. Therefore, the predetermined distance is set to a value that reflects this distance, and the second detection unit 12 detects crosswalks without a traffic light by detecting crosswalks within the predetermined distance from the location where the diamond mark is detected.

[0016] The vehicle speed acquisition unit 20 acquires information about the vehicle speed. This vehicle speed acquisition unit 20 receives vehicle speed information from a calculation unit (not shown) that calculates the vehicle speed based on, for example, a vehicle speed pulse. Note that the vehicle speed acquisition unit 20 may receive the vehicle speed pulse and perform calculations to acquire the vehicle speed information, or may acquire the vehicle speed information by calculating the vehicle speed from changes in the vehicle position based on radio waves from an artificial satellite.

[0017] The vehicle speed acquisition unit 20 transmits the acquired vehicle speed information to the control unit 10. The vehicle speed information transmitted to the control unit 10 is used by the second detection unit 12 to determine whether or not the vehicle is within a predetermined distance.

[0018] Figure 2 is a diagram showing an example of a captured image obtained by the camera 2 shown in Figure 1. The second detection unit 12 detects a crosswalk PC from a range corresponding to a first distance L1 or more and a second distance L2 or less from the vehicle on the captured image. In this case, the second detection unit 12 detects a crosswalk PC where a straight line SL corresponding to the innermost edge of the pattern of the crosswalk PC (intermittent white lines crossing the road) is in a range of the first distance L1 or more and the second distance L2 or less.

[0019] The third detection unit 13 shown in FIG. 1 detects pedestrians from within the captured image acquired by the camera 2. The third detection unit 13 is configured with, for example, an AI-based person detection function, and detects pedestrian H (see FIG. 2) from within the captured image. Note that in the example shown in FIG. 2, the pedestrian H is a person, but is not limited to this, and a person riding a bicycle, a tricycle, or the like may also be detected as pedestrian H. Furthermore, the third detection unit 13 may detect pedestrian H using other methods such as template matching without relying on AI.

[0020] Furthermore, the control unit 10 includes a first judgment unit (object judgment means) 14, a second judgment unit 15, a third judgment unit (start / stop judgment means) 16, a fourth judgment unit 17, and a fifth judgment unit 18. In addition, the control unit 10 includes an evaluation unit (evaluation means) 19, an event recording unit 10a, a timing unit 10b, and a speaker control unit 10c.

[0021] The first determination unit 14 determines the presence of a target pedestrian EP (see FIG. 2), which is a pedestrian H that satisfies a predetermined condition, among the pedestrians H detected by the third detection unit 13. The target pedestrian EP is an object whose walking on the crosswalk PC should be given priority over the movement of the vehicle at a crosswalk PC that is not equipped with traffic lights. The first determination unit 14 determines the target pedestrian EP based on a target area A shown in FIG. 2.

[0022] The target area A has a first area A1 and a second area A2. The first area A1 is, for example, an area in front of the host vehicle, and has a width equal to the host vehicle width and a depth (vertical width) extending a specific distance from the front end of the host vehicle. In FIG. 2, the first area A1 is indicated by narrow-pitch diagonal hatching. The second area A2 is an area in front of the host vehicle, and has a width extending a specified distance from the left or right edge of the host vehicle to a position in the left-right direction, and a depth extending a specific distance from the front end of the host vehicle. In FIG. 2, the second area A2 is indicated by dot hatching.

[0023] When a pedestrian H determined to be present within the first area A1 is moving, the first determination unit 14 determines that the predetermined condition is satisfied and determines that the pedestrian H is a target pedestrian EP. The movement is determined, for example, based on the position of the pedestrian H in the current captured image and the position of the pedestrian H in the previous captured image. The position of the pedestrian H may be determined by the head, the center of gravity of the body, and the position relative to the road surface. For example, the pedestrian H1 located slightly in front of the center of the captured image shown in FIG. 2 is located within the first area A1. Therefore, if the pedestrian H1 is moving, the first determination unit 14 determines that the pedestrian H1 is a target pedestrian EP. Note that the first determination unit 14 may determine that the pedestrian H determined to be present within the first area A1 is a target pedestrian EP even if the pedestrian H is not moving.

[0024] Furthermore, the first determination unit 14 determines that a pedestrian H determined to be present within the second area A2 satisfies a predetermined condition when the pedestrian H is moving toward the center as viewed in the width direction of the image, and determines that the pedestrian H is a target pedestrian EP. For example, pedestrian H2 located slightly to the right of the center of the captured image shown in FIG. 2 is located within the second area A2 and is moving toward the left (toward the center). Therefore, the first determination unit 14 determines that pedestrian H2 is a target pedestrian EP. Similarly, the first determination unit 14 determines that pedestrian H3 located slightly to the left of the center of the captured image is also a target pedestrian EP.

[0025] On the other hand, pedestrian H4, who is present in the second area A2 but is not moving toward the center, and pedestrian H5, who is stationary, do not satisfy the predetermined condition. Therefore, the first determination unit 14 will not determine these pedestrians H4 and H5 to be target pedestrians EP. Furthermore, the first determination unit 14 will not determine pedestrian H6, who is not present in the first area A1 or the second area A2, to be a target pedestrian EP.

[0026] The second determination unit 15 shown in Fig. 1 determines whether a pedestrian H is present within a danger zone. For example, the second determination unit 15 determines whether a pedestrian H (hereinafter referred to as a dangerous pedestrian DH) is present within the danger zone DA, which is an area of ​​the first area A1 that is a first distance L1 or less from the vehicle. In the example shown in Fig. 2, only the pedestrian H1 is determined to be a dangerous pedestrian DH.

[0027] Here, the dangerous pedestrian DH determined by the second judgment unit 15 is a different concept from the target pedestrian EP determined by the first judgment unit 14. That is, the target pedestrian EP determined by the first judgment unit 14 is an object that should be given priority at a crosswalk PC without traffic lights, whereas the dangerous pedestrian DH determined by the second judgment unit 15 indicates an object with a relatively high possibility of collision if the host vehicle starts moving. Note that, because the dangerous area DA and the target area A overlap, it is possible that a pedestrian, like pedestrian H1, may be determined to be both a target pedestrian EP and a dangerous pedestrian DH by both the first judgment unit 14 and the second judgment unit 15.

[0028] 1 determines whether the host vehicle is in a starting state or a stopped state based on the brake signal and information from the vehicle speed acquisition unit 20. This third determination unit 16 determines that the host vehicle has transitioned from a traveling state to a stopped state when the brake signal is turned on and the vehicle speed becomes zero. Furthermore, the third determination unit 16 determines that the host vehicle has transitioned from a stopped state to a starting state when the brake signal is turned off or the vehicle speed exceeds zero while the host vehicle is in a stopped state.

[0029] The fourth determination unit 17 determines the start and end of the monitoring state. The fifth determination unit 18 determines whether the driver of the host vehicle has committed a departure violation. The fourth determination unit 17 determines that the monitoring state has started when a specific time (for example, 3 to 10 seconds) has elapsed since the third determination unit 16 determined that the host vehicle is in a stopped state. The fifth determination unit 18 determines that a departure violation has occurred when the host vehicle starts despite the presence of a dangerous pedestrian DH determined by the second determination unit 15 during the monitoring state. Furthermore, the fourth determination unit 17 determines that the monitoring state has ended when the second determination unit 15 can no longer detect the presence of a dangerous pedestrian DH.

[0030] The evaluation unit 19 executes evaluation processing related to the driving of the driver of the host vehicle. In particular, the evaluation unit 19 according to this embodiment executes positive evaluation processing (GOOD evaluation processing) when priority is given to pedestrian H at a crosswalk PC without traffic lights. Note that the evaluation unit 19 may execute negative evaluation processing (BAD evaluation processing) when priority is not given to pedestrian H at a crosswalk PC without traffic lights. Furthermore, the evaluation unit 19 may execute negative evaluation processing when the host vehicle starts moving despite the second determination unit 15 determining the presence of a dangerous pedestrian DH.

[0031] The event recording unit 10a records sudden stops, sudden starts, and the like as events, and in this embodiment also records as events scenes where pedestrian H should have priority at a crosswalk PC without traffic lights.

[0032] The storage unit 30 is an internal memory that stores a driving evaluation program for causing the driving evaluation device 1 according to this embodiment to function. The storage unit 30 may also function as a device that records the above-mentioned events. The events may be recorded on a memory card that is inserted into the driving evaluation device 1. The communication unit 40 transmits information such as the vehicle's position information, driving speed, and event information to an external server.

[0033] The timer unit 10b is a functional unit that measures time, for example, timing a specific time when the monitoring state starts and timing a timeout, which will be described later. The speaker control unit 10c controls the output of voice and sound from the speaker 3. This speaker control unit 10c issues an alarm when a departure violation occurs, or provides a voice notification to the driver when a GOOD evaluation is given, for example.

[0034] Next, the driving evaluation method of the driving evaluation device 1 according to this embodiment will be described with reference to Figures 2 to 5. Figure 3 is a first diagram showing the driving evaluation method of the driving evaluation device 1 according to this embodiment, where (a) shows the state at a first time, (b) shows the state at a second time, (c) shows the state at a third time, and (d) shows the state at a fourth time. Figure 4 is a second diagram showing the driving evaluation method of the driving evaluation device 1 according to this embodiment.

[0035] First, assume that the host vehicle is traveling on a road. In this state, camera 2 shown in FIG. 1 captures an image of the area ahead of the vehicle. Then, as shown in FIG. 3(a), assume that the host vehicle V approaches a diamond mark D at a first time point. In this case, first detection unit 11 (see FIG. 1) detects diamond mark D from the captured image. As a result, second detection unit 12 (see FIG. 1) executes a process of detecting a pedestrian crossing PC within a predetermined distance from diamond mark D.

[0036] Then, as shown in FIG. 3(b), it is assumed that the host vehicle V travels close to a pedestrian crossing PC within a predetermined distance from the diamond mark D. At this time, the second detection unit 12 detects the pedestrian crossing PC. Note that, because the detected pedestrian crossing PC is located within the predetermined distance from the diamond mark D, it can be said that the pedestrian crossing PC is a pedestrian crossing PC without a traffic light. Thereafter, the third detection unit 13 (see FIG. 1) detects a pedestrian H (see FIG. 2) from the captured image. Note that the detection of the pedestrian H is not limited to after the detection of the pedestrian crossing PC, and may be performed before the detection of the pedestrian crossing PC.

[0037] Here, as shown in FIG. 3(c), if a pedestrian H is present near the crosswalk PC, the driver stops the host vehicle V in front of the crosswalk PC. At this time, the vehicle speed becomes zero by applying the brakes or the like. The third determination unit 16 determines this situation based on the brake signal and information from the vehicle speed acquisition unit 20, and determines that the host vehicle V is in a stopped state. Furthermore, in this state, the first determination unit 14 determines the presence of a target pedestrian EP.

[0038] Furthermore, the fourth judgment unit 17 (see Figure 1) judges whether a specific time (for example, 3 seconds or more and 10 seconds or less) has passed since the stop state based on information from the third judgment unit 16 and the timing unit 10b, and if it determines that the time has passed, it transitions to a monitoring state.

[0039] In the monitoring state, the second determination unit 15 (see FIG. 1) determines whether a dangerous pedestrian DH exists in the dangerous area DA (see FIG. 2). In addition, in the state where the second determination unit 15 determines whether a dangerous pedestrian DH (see FIG. 2 or FIG. 4) exists, the third determination unit 16 determines whether the host vehicle V has entered a starting state.

[0040] Then, as shown in Fig. 4, when the host vehicle V starts moving in a state in which it is determined that a dangerous pedestrian DH is present, the fifth determination unit 18 (see Fig. 1) determines that a starting violation has occurred. As a result, the speaker control unit 10c issues an alarm from the speaker 3 (see Fig. 1). In this case, the evaluation unit 19 does not execute a positive evaluation process for the driver's driving. Note that in this case, the evaluation unit 19 may execute a negative evaluation process.

[0041] On the other hand, as shown in FIG. 3(d), suppose that the target pedestrian EP passes through the crosswalk PC and then the host vehicle V starts moving. In this case, the target pedestrian EP, who was within the target area A in the state of FIG. 3(c), is located outside the target area A in the state of FIG. 3(d). Therefore, the first determination unit 14 determines that the target pedestrian EP is not present at the time of FIG. 3(d). Furthermore, if the host vehicle V starts moving in this state, the evaluation unit 19 determines that the host vehicle V has given priority to pedestrian H at the crosswalk PC where there are no traffic lights, and executes a positive evaluation process.

[0042] It is preferable that the evaluation unit 19 executes the evaluation process after the host vehicle V has passed the crosswalk PC. That is, it is preferable that the evaluation unit 19 executes the evaluation process after the crosswalk PC is no longer detected by the second detection unit 12 (for example, after it has gone out of frame). This is because an accurate driving evaluation can be performed in the following situations.

[0043] FIG. 5 is a third diagram illustrating the driving evaluation method of the driving evaluation device 1 according to this embodiment. As shown in FIG. 5, the number of pedestrians H passing through a crosswalk PC without traffic lights is not limited to one. Therefore, the first determination unit 14 may determine the presence of a second target pedestrian (another target pedestrian) EP2 immediately after a first target pedestrian EP1 passes through the crosswalk PC. Moreover, in this case, the driver may overlook the presence of the second target pedestrian EP2 and start the host vehicle V. In such a case, if the evaluation unit 19 performs a positive evaluation process immediately after giving priority to the first target pedestrian EP1, the positive evaluation process will be executed even though the second target pedestrian EP2 was not actually given priority. However, in this embodiment, the evaluation unit 19 performs the evaluation process after the host vehicle V passes through the crosswalk PC, and therefore prohibits the positive evaluation process because the second target pedestrian EP2 was not given priority.

[0044] Next, the driving evaluation method according to this embodiment will be described in detail. Fig. 6 is a first flowchart showing the driving evaluation method according to this embodiment, illustrating the process related to evaluation at a crosswalk PC without traffic lights. Fig. 7 is a second flowchart showing the driving evaluation method according to this embodiment, illustrating the process related to a departure violation. The processes in Figs. 6 and 7 are executed in parallel, and both are repeatedly executed until the power of the driving evaluation device 1 is turned off.

[0045] 6, the control unit 10 first determines whether the first detection unit 11 has detected a diamond mark D on the road on which the vehicle V is traveling (S1). If the diamond mark D has not been detected (S1: NO), the control unit 10 repeats this process.

[0046] If the first detection unit 11 detects the diamond mark D (S1: YES), the driving evaluation device 1 determines whether the position of the vehicle V is within a predetermined distance from the diamond mark D (lost diamond mark D) (S2). If the position of the vehicle V is not within the predetermined distance from the diamond mark D (S2: NO), the process proceeds to step S1.

[0047] On the other hand, if the position of the vehicle V is within the predetermined distance from the diamond mark D (S2: YES), the control unit 10 determines whether the speed of the vehicle V is equal to or less than the predetermined speed and whether the second detection unit 12 has detected the crosswalk PC (S3). The predetermined speed is the speed immediately before the vehicle stops, and is set to a value of, for example, 20 km / h or less.

[0048] Here, detecting the crosswalk PC at a low speed means that there is a lower possibility of false detection than detecting the crosswalk PC at a high speed. Also, detecting the crosswalk PC at a low speed means that there is a high possibility that the driver has already confirmed the pedestrian H near the crosswalk PC. Therefore, by checking that the speed of the vehicle V is equal to or less than a predetermined speed in connection with the detection of the crosswalk PC, the driving evaluation device 1 can improve the detection accuracy of the crosswalk PC and determine that it is a situation where there is a high possibility that a pedestrian H is present near the crosswalk PC.

[0049] If the speed of the host vehicle V is not equal to or less than the predetermined speed (S3: NO), or if the second detection unit 12 has not detected the pedestrian crossing PC (S3: NO), the process proceeds to step S2. On the other hand, if the speed of the host vehicle V is equal to or less than the predetermined speed and the pedestrian crossing PC has been detected (S3: YES), the control unit 10 determines whether the host vehicle V is stopped (is in a stopped state) and whether the first determination unit 14 has detected the presence of a target pedestrian EP (S4).

[0050] If the host vehicle is not stopped (S4: NO), or if the first determination unit 14 has not detected the target pedestrian EP (S4: NO), the process proceeds to step S2.

[0051] When the host vehicle V stops and the first detection unit 11 detects a target pedestrian EP (S4: YES), the control unit 10 determines that an event in which the target pedestrian EP should be prioritized has started. Then, the control unit 10 determines whether the target pedestrian EP is no longer detected (S5). If the target pedestrian EP is detected (S5: NO), this process is repeated until the target pedestrian EP is no longer detected. On the other hand, if the target pedestrian EP is no longer detected (S5: YES), the control unit 10 determines whether the host vehicle V has entered a start state and passed the pedestrian crossing PC (S6). Here, the control unit 10 determines whether the host vehicle V has passed the pedestrian crossing PC based on whether the host vehicle has traveled a certain distance since losing the pedestrian crossing PC from the captured image.

[0052] If the vehicle V has not passed through the crosswalk PC (S6: NO), this process is repeated until it is determined that the vehicle V has passed through the crosswalk PC.

[0053] On the other hand, if the host vehicle V has passed the pedestrian crossing PC (S6: YES), the control unit 10 determines whether a specified time (for example, several tens of seconds or more) has passed since the monitoring state was established (S7). If the specified time has passed (S7: YES), the process proceeds to step S1. Note that in steps S5 and S6, it is possible that the host vehicle V may pass the pedestrian crossing PC while the target pedestrian EP is still detected. For this reason, in such a case, it is preferable to proceed to step S1.

[0054] Here, when a specified time of several tens of seconds or more has elapsed since the monitoring state was entered, it may be the case that the driver has stopped the vehicle V on the shoulder in front of a crosswalk PC without traffic lights, rather than giving priority to pedestrian H at a crosswalk PC without traffic lights. In such a case, it is clear that the driver has no intention of giving priority to pedestrian H. Therefore, when a time (i.e., the specified time) that is clearly longer than the time for giving priority to target pedestrian EP has elapsed, the evaluation unit 19 returns the process to step S1 to prohibit the execution of the evaluation process.

[0055] In this process, the control unit 10 determines that a specified time has elapsed since the monitoring state was entered. The monitoring state is a state to which the vehicle V is transitioned when a specific time has elapsed since the vehicle V was placed in a stopped state. Therefore, in the process of step S7, the control unit 10 can be said to determine whether a predetermined time (specific time + specified time) has elapsed since the vehicle V was determined to be in a stopped state.

[0056] If the specified time has not elapsed in the processing of step S7 (S7: NO), the control unit determines whether a starting violation has occurred (S8). A starting violation is determined according to the flowchart shown in FIG. 7. If a starting violation has occurred (S8: YES), the processing proceeds to step S1. In this case, the evaluation unit 19 may execute a negative evaluation process.

[0057] If there is no departure violation (S8: NO), the evaluation unit 19 determines that the driver gave priority to pedestrian H at the crosswalk PC and executes a positive evaluation process (S9). Then, the event recording unit 10a records the fact that the driver gave priority to pedestrian H as an event (S10), and the speaker control unit 10c outputs a voice indicating that a positive evaluation was made (S11). Thereafter, the process shown in FIG. 6 ends.

[0058] 7, the control unit 10 determines whether a specific time has elapsed while the host vehicle V is stopped (S21). That is, the control unit 10 determines whether a specific time has elapsed while the brake signal is on and the vehicle speed is zero (S21). If the specific time has not elapsed while the brake signal is on and the vehicle speed is zero (S21: NO), this process is repeated until the specific time has elapsed while the host vehicle V is stopped.

[0059] On the other hand, if a specific time has elapsed in these states (S21: YES), the monitoring state is started (S22). Then, the control unit 10 determines whether a dangerous pedestrian DH has been detected by the second detection unit 12 (S23). If a dangerous pedestrian DH has not been detected (S23: NO), the process proceeds to step S21.

[0060] If a dangerous pedestrian DH is detected (S23: YES), the control unit 10 determines whether the dangerous pedestrian DH is no longer detected (S24). If the dangerous pedestrian DH is no longer detected (S24: YES), it is assumed that the dangerous pedestrian DH has crossed the crosswalk PC. Therefore, the monitoring state ends (S25), and the process shown in FIG. 7 ends.

[0061] On the other hand, if the dangerous pedestrian DH remains detected (S24: NO), the control unit 10 determines whether the host vehicle V has entered a starting state, i.e., whether the brake signal has been turned off or the vehicle speed has exceeded zero (S26). If the brake signal remains on and the vehicle speed remains zero (S26: NO), the process proceeds to step S24.

[0062] On the other hand, if the brake signal is off or the vehicle speed exceeds zero (S26: YES), the host vehicle V is in a starting state despite the presence of the dangerous pedestrian DH. Therefore, the speaker control unit 10c issues a warning alarm from the speaker 3, and the fifth determination unit 18 determines that the host vehicle V has violated the starting rule (S27).

[0063] Next, the control unit 10 determines again whether the dangerous pedestrian DH is no longer detected (S28). If the dangerous pedestrian DH remains detected (S28: NO), the control unit 10 determines whether the host vehicle V has come to a stop, i.e., whether the brake signal is on or the vehicle speed has become zero (S29).

[0064] If the brake signal is on or the vehicle speed becomes zero (S29: YES), it is highly likely that the driver started the vehicle V without noticing the dangerous pedestrian DH, and then noticed the dangerous pedestrian DH due to the advance warning and stopped the vehicle V. Therefore, the speaker control unit 10c issues a main warning from the speaker 3 (S30) to notify the driver that the previous driving was dangerous. After that, the processing shown in FIG. 7 ends.

[0065] If the brake signal remains off and the host vehicle V remains in a starting state (S29: NO), the process proceeds to step S28. If the dangerous pedestrian DH is no longer detected in step S28 (S28: YES), it can be said that the host vehicle V remains in a starting state without coming to a stop and the dangerous pedestrian DH is no longer detected. In other words, it can be said that the driver has started the host vehicle V, judging that the pedestrian H is about to leave the dangerous area DA. Therefore, the monitoring state ends (S25), and the process shown in FIG. 7 ends.

[0066] In this way, the driving evaluation device 1, driving evaluation method, and driving evaluation program according to this embodiment detect a diamond mark D painted on the road on which the vehicle V is traveling, and detect a crosswalk PC within a predetermined distance from the point where the diamond mark D was detected. Here, the diamond mark D is painted a certain distance before the crosswalk PC without a traffic light. Therefore, by detecting the crosswalk PC within the predetermined distance from the point where the diamond mark D was detected, it is possible to identify the crosswalk PC without a traffic light without performing image analysis to determine whether or not there is a traffic light.

[0067] Furthermore, after the crosswalk PC is detected, the target pedestrian EP is detected, and the host vehicle V is determined to be in a stopped state, the evaluation process is performed based on whether the presence of the target pedestrian EP was determined when the host vehicle V was determined to be in a starting state. Therefore, a case is identified in which the crosswalk PC is detected, the target pedestrian EP is detected, and the host vehicle V is in a stopped state, and the host vehicle should wait for the target pedestrian EP to pass the crosswalk PC. Then, the evaluation is performed based on the fact whether the target pedestrian EP was present when the host vehicle V entered a starting state. Therefore, it is possible to determine a situation in which pedestrian H is given priority without relying on the length of time the host vehicle V is stopped.

[0068] From the above, it is possible to more accurately judge the situation where pedestrian H has priority at a crosswalk PC without traffic lights, thereby improving the accuracy of driver evaluation.

[0069] Furthermore, the evaluation process is executed after the host vehicle V is determined to be in a starting state and is no longer able to detect the pedestrian crossing PC that had been detected. Therefore, the evaluation process is executed after determining that the host vehicle V has passed the pedestrian crossing PC, in which case the host vehicle V is in a starting state and is no longer able to detect the pedestrian crossing PC. Therefore, the evaluation process is not executed before the host vehicle V passes the pedestrian crossing PC. This prevents a situation where, for example, a first target pedestrian EP1 is given priority, but a second target pedestrian EP2 is detected immediately thereafter and is no longer given priority, and a positive evaluation is made immediately after the first target pedestrian EP1 is given priority, even though a positive evaluation would not normally be possible. This further improves the driver's evaluation accuracy.

[0070] Furthermore, when the vehicle V is in a starting state, if the evaluation unit 19 cannot determine the presence of a first target pedestrian EP1 but determines the presence of a new second target pedestrian EP2, the evaluation unit 19 prohibits the execution of a positive evaluation process. Therefore, if a certain pedestrian H has been given priority at a crosswalk PC without traffic lights, but another pedestrian H should also be given priority, the execution of a positive evaluation can be prohibited, thereby further improving the accuracy of the driver's evaluation. In this case, a negative evaluation process may be additionally executed.

[0071] Furthermore, if a predetermined time has elapsed since it was determined that the host vehicle V is in a stopped state, the execution of the evaluation process is prohibited. Here, the vehicle tends to be stopped for a long time when the driver does not intend to give priority to pedestrian H, such as when stopping the vehicle on the shoulder in front of a crosswalk PC without traffic lights. Therefore, by prohibiting the evaluation process in such cases, it is possible to prevent the evaluation process from being performed when the driver does not intend to give priority to pedestrian H, and further improve the accuracy of the driver's evaluation.

[0072] Note that prohibiting the execution of the evaluation process after a predetermined time has elapsed also produces the following incidental effect. For example, the second detection unit 12 may erroneously detect an object reflected on the windshield as a crosswalk PC. Here, the device 1 determines that the crosswalk PC has been passed after traveling a certain distance after losing the crosswalk PC in the captured image. However, if the device 1 continues to erroneously detect something reflected on the windshield as a crosswalk PC, the crosswalk PC will not disappear from the captured image. In this case, even if the vehicle V is stopped and a pedestrian H crosses in front of the vehicle V when the crosswalk PC is not actually present, the device may determine that priority has been given to the pedestrian H. However, by determining the elapse of a predetermined time, as in this embodiment, the possibility of making an erroneous evaluation can be reduced.

[0073] Additionally, the second detection unit 12 detects the crosswalk PC from a range corresponding to a first distance L1 or more and a second distance L2 or less from the vehicle V. This narrows the detection range of the crosswalk PC, making it less likely to mistakenly detect a distant crosswalk PC when, for example, a crosswalk PC without traffic lights exists in succession in front of the vehicle V and in the distance. Therefore, the accuracy of the driver's evaluation can be improved along with the improvement in the detection accuracy of the crosswalk PC.

[0074] Although the present invention has been described above based on the embodiments, the present invention is not limited to the above embodiments, and modifications may be made without departing from the spirit of the present invention, and other techniques may be appropriately combined to the extent possible. Furthermore, publicly known or well-known techniques may be combined to the extent possible.

[0075] For example, in the above embodiment, the driving evaluation device 1 enters a monitoring state when the vehicle V remains stopped for a specific period of time and determines that a starting violation has occurred. However, if possible, the driving evaluation device 1 may use an infrared sensor or the like to determine a situation in which the vehicle V has approached an obstacle more closely than necessary without entering the monitoring state, and determine that a starting violation has occurred when the vehicle V has approached an obstacle more closely than necessary. Also, a starting violation may be determined by other methods. Furthermore, the driving evaluation device 1 may evaluate the driver's driving without determining that a starting violation has occurred.

[0076] In the above embodiment, the driving evaluation device 1 is assumed not to reflect the fact that the driver gave priority to pedestrian H at a crosswalk PC without a traffic light in the daily report or performance evaluation of the truck driver or taxi driver. However, this is not limiting, and the fact may be reflected in the daily report or performance evaluation. [Explanation of symbols]

[0077] 1: Driving evaluation device 2: Camera (imaging means) 10: Control section 11: First detection unit (first detection means) 12: Second detection unit (second detection means) 13: Third detection unit 14: First judgment unit (object judgment means) 15:Second Judgment Department 16:Third judgment part (start / stop judgment means) 17: 4th Judgment Department 18: 5th Judgment Department 19: Evaluation section (evaluation means) 20:Vehicle speed acquisition part 30: Storage section 40: Communications Department A: Target area A1: Area 1 A2: Area 2 D: Diamond mark DA: Danger Area DH: Dangerous Pedestrian EP: Target pedestrian EP1: First target pedestrian EP2: Second target pedestrian H: Pedestrian L1: 1st distance L2: 2nd distance PC: Crosswalk SL: Straight line V: Vehicle

Claims

1. a first detection means for detecting a diamond mark painted on a road on which the vehicle is traveling based on an image captured by an imaging means for capturing an image of a road ahead of the vehicle; a second detection means for detecting a crosswalk within a predetermined distance from the location where the diamond mark is detected by the first detection means, based on the captured image obtained by the imaging means; a target determination means for determining the presence of a target pedestrian who satisfies a predetermined condition from the captured image acquired by the imaging means; a start / stop determination means for determining whether the host vehicle is in a starting state or a stopped state; evaluation means for executing an evaluation process for the driver of the vehicle; The evaluation means performs an evaluation process based on whether or not the object determination means has determined the presence of a target pedestrian when the second detection means detects the crosswalk, the object determination means determines the presence of a target pedestrian, and the start / stop determination means determines that the host vehicle is in a stopped state, and then the start / stop determination means determines that the host vehicle is in a starting state. A driving evaluation device characterized by:

2. The evaluation means executes the evaluation process after the start / stop determination means determines that the host vehicle is in a starting state and the crosswalk that was detected by the second detection means can no longer be detected.

2. The driving evaluation device according to claim 1.

3. When the start / stop determination means determines that the vehicle is in a starting state, if the target determination means no longer determines the presence of a first target pedestrian, and if the target determination means determines the presence of a second target pedestrian that is a new target pedestrian different from the first target pedestrian, the evaluation means prohibits the execution of a positive evaluation process.

2. The driving evaluation device according to claim 1.

4. The evaluation means prohibits execution of the evaluation process when a predetermined time has elapsed since the second detection means detected the crosswalk and the start / stop determination means determined that the vehicle is stopped.

2. The driving evaluation device according to claim 1.

5. The second detection means detects a crosswalk from a range corresponding to a first distance or more and a second distance or less from the vehicle on the captured image acquired by the imaging means.

2. The driving evaluation device according to claim 1.

6. a first detection step of detecting a diamond mark painted on a road on which the vehicle is traveling based on an image captured by an imaging means that captures an image of a road ahead of the vehicle; a second detection step of detecting a crosswalk within a predetermined distance from the location where the diamond mark was detected in the first detection step, based on the captured image acquired by the imaging means; a target determination step of determining whether there is a target pedestrian who satisfies a predetermined condition from the captured image acquired by the imaging means; a start / stop determination step of determining whether the host vehicle is in a starting state or a stopped state; an evaluation step of executing an evaluation process for the driver of the vehicle; In the evaluation step, the crosswalk is detected in the second detection step, the presence of a target pedestrian is determined in the target determination step, and after the host vehicle is determined to be in a stopped state in the start / stop determination step, the evaluation process is performed based on whether or not the presence of a target pedestrian is determined in the target determination step, which is executed when the host vehicle is determined to be in a starting state in the start / stop determination step. A driving evaluation method for a driving evaluation device.

7. A driving evaluation program for causing a computer to function as the driving evaluation device according to claim 1.

Images