JP4893593B2 - Vehicle driving support device - Google Patents

Description

  The present invention relates to a vehicle driving support device, and more particularly to a vehicle driving support device that informs a driver of the presence of an obstacle so that the driver can avoid a collision with the obstacle.

  2. Description of the Related Art Conventionally, there has been known a vehicle driving support device that is configured to issue an alarm or the like to call a driver's attention when approaching a predetermined distance when there is an obstacle in front of the vehicle.

  In recent years, a plurality of model images are prepared in advance in such a system, and a so-called pattern matching process for comparing the model image with an image photographed by an in-vehicle camera or the like is executed, whereby an obstacle is detected. Some are detected (see Patent Document 1).

JP 2007-58805 A

  However, when detecting an obstacle in such a method, it is extremely presumed that various obstacle patterns are preliminarily assumed in accordance with various traveling environments and stored in a database in the in-vehicle control device. Have difficulty.

Therefore, there is a possibility that something Re in does not fit obstacle previously prepared model image appears, it is impossible to improve the recognition accuracy of the obstacle from these circumstances or.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a vehicle driving support device capable of enhancing a database for pattern matching used for pattern matching processing and thereby realizing improvement in obstacle recognition accuracy.

The vehicle driving support device of the present invention is a vehicle driving support device that informs the driver of the presence of an obstacle, and detects the obstacle by comparing it with stored pattern matching data. Obstacle notifying means for informing the driver's line of sight toward the obstacle direction, gaze direction detecting means for detecting the driver's line of sight or face direction, and the obstacle targeted for notification by the obstacle notifying means And a mismatch level detecting means for detecting a mismatch level of the line of sight based on a mismatch frequency indicating a frequency at which the driver's line of sight does not match, or a mismatch duration, and a level of mismatch of the line of sight is predetermined. when it is less than the value storage means for adding storage and estimation means for estimating the driver's line of sight direction of the obstacle, the estimated the obstacle as a new pattern matching data It is those with a.

According to this configuration, when the disagreement level between the direction of the obstacle to be notified and the driver's line-of-sight direction is equal to or higher than a predetermined value, the obstacle in the driver's line-of-sight direction is estimated, and the estimated Since the obstacle is additionally stored as new pattern matching data, the database can be enhanced by accumulating the pattern matching data, and the obstacle recognition accuracy can be improved.
In the present invention, the “mismatch frequency” means the number of times that the sight line direction mismatch occurs per unit time.

  In one embodiment of the present invention, an imaging means for photographing the periphery of the vehicle, a means for presenting a photographed image photographed when the discrepancy level is equal to or greater than a predetermined value, and whether additional storage is necessary or not to the driver. Means for determining, and performs the processing related to the additional storage based on the result of the determination.

  According to this configuration, it is possible to store only pattern matching data that is considered necessary for the driver, thereby improving the accuracy of obstacle recognition.

  In one embodiment of the present invention, the image processing apparatus includes an imaging unit that captures the periphery of the vehicle, and a unit that presents a captured image captured when the mismatch level is equal to or greater than a predetermined value to the driver. In addition to the obstacle image in the line-of-sight direction included in the captured image presented in Fig. 1, the risk determination result of the obstacle is also stored.

  According to this configuration, since it is possible to determine the degree of danger at the time of pattern matching, the recognition accuracy of the obstacle can be improved, and notification according to the degree of danger becomes possible.

  In one embodiment of the present invention, the risk is determined by the driver.

  According to this configuration, it is possible to recognize the degree of danger more accurately, thereby improving the recognition accuracy of the obstacle.

  In one embodiment of the present invention, the image processing apparatus includes an imaging unit that captures the periphery of the vehicle, and a unit that presents a captured image captured when the mismatch level is equal to or greater than a predetermined value to the driver. The obstacle type determination result of the obstacle is stored together with the obstacle image in the line-of-sight direction included in the captured image presented in FIG.

  According to this configuration, obstacle recognition according to the characteristics of the type can be performed, and the recognition accuracy can be improved.

  In one embodiment of the present invention, the determination of the type is made by the driver.

  According to this structure, it becomes possible to recognize a classification more correctly, and this can improve the recognition accuracy of an obstruction more.

  In an embodiment of the present invention, the image presentation to the driver is executed when the vehicle stops for the first time after the mismatch level becomes a predetermined value or more.

  According to this configuration, it is possible to reliably avoid the confirmation of the display contents during driving of the vehicle and ensure safety during driving. Furthermore, before the driver forgets the presence of the obstacle, the process related to the additional storage can be executed.

  In one embodiment of the present invention, at the time of the additional storage, a vehicle driving environment when the mismatch level becomes a predetermined value or more is also stored, and when the pattern matching data is read, the data is stored in the driving environment. It is read out accordingly.

  According to this configuration, it is possible to finely perform obstacle recognition by reading pattern matching data used for recognition according to changes in the driving environment, thereby improving obstacle recognition accuracy. .

According to the present invention, when the mismatch level between the direction of the obstacle to be notified and the driver's line-of-sight direction is equal to or greater than a predetermined value, the obstacle in the driver's line-of-sight direction is estimated, and the estimated Since the obstacle is additionally stored as new pattern matching data, it is possible to enhance the database by accumulating the pattern matching data, thereby improving the obstacle recognition accuracy.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic diagram of an automobile employing a vehicle driving support apparatus according to an embodiment of the present invention, FIG. 2 is a system block diagram of the vehicle driving support apparatus, and FIG. 3 is an obstacle direction and a driver's line-of-sight direction. FIG. 4 is a flowchart for explaining a method for extracting an obstacle image in the driver's line-of-sight direction, and FIG. 5 is a diagram for pattern-matching the obstacle image in the driver's line-of-sight direction. FIG. 6 is a flowchart illustrating a method of additionally storing data as data, and FIG. 6 is a diagram illustrating a display example when additionally storing an obstacle image in the line of sight of the driver as pattern matching data.

  As shown in FIG. 1, a plurality of sensor devices are installed in an automobile 1, a vehicle speed sensor 2 that detects a vehicle speed, an obstacle detection camera 3 that detects an obstacle ahead of the vehicle, and a traveling environment. A raindrop sensor 4 as an environment detection sensor to detect and a line-of-sight detection sensor 5 for detecting the driver's line-of-sight direction are provided.

  The obstacle detection camera 3 is a known stereo camera provided in the vicinity of a sun visor (not shown) in the front of the passenger compartment, and picks up the periphery (particularly in front) of the automobile 1 and obtains a stereo image obtained as a result. This is to extract and specify obstacles such as other vehicles and pedestrians that should avoid contact from the data.

  The automobile 1 also includes a window display actuator 6 and a confirmation display / input operation combined display 7 (hereinafter abbreviated as a display 7) as information display means for the driver.

  These sensors (2, 4, 5), the obstacle detection camera 3, and the information display means (6, 7) are respectively connected to a control device 10 comprising a CPU or the like as shown in the block diagram of FIG. Each sensor (2, 4, 5) transmits each detection signal to the control device 10, and the obstacle detection camera 3 transmits the captured stereo image data to the control device 10. Each information display means (6, 7) is configured to perform a predetermined operation in response to an operation signal transmitted from the control device 10.

  The vehicle speed sensor 2 transmits a vehicle speed signal to the control device 10, and the raindrop sensor 4 detects a rainfall state based on the pressure of the raindrop contacting the vehicle 1, and transmits the detection signal to the control device 10. Yes.

  The line-of-sight detection sensor 5 detects the driver's face direction by, for example, detecting the line of sight based on the position of the driver's black eye or photographing the driver's face with a face image capturing camera. From the line of sight. Generally, the former is provided near the sun visor, and the latter is provided in a steering column (not shown).

  The line-of-sight detection sensor 5 transmits a detection signal corresponding to the line-of-sight detection result to the control device 10.

  Further, the display 7 can transmit an operation signal transmitted by a driver's operation to the control device 10.

  A storage device 8 is connected to the control device 10, and various data such as pattern matching data including obstacle image data and accompanying data described later are written and read by the control device 10. It is possible.

  In addition, the control device 10 includes an obstacle extraction processing unit 11, an obstacle direction detection processing unit 12, and a detected obstacle display processing unit 13. The obstacle extraction processing unit 11 detects the presence of obstacles such as other vehicles, pedestrians, and two-wheeled vehicles around the automobile 1 (particularly in front) based on the image data from the obstacle detection camera 3.

  Further, in the present embodiment, the obstacle direction detection processing unit 12 is based on the image data obtained from the obstacle detection camera 3, and the traveling direction of the automobile 1 and the detected obstacle I1 (see FIG. 3). The angle difference θA from the direction, that is, the direction of the obstacle seen from the automobile 1 is detected.

  In addition, when the obstacle I1 is detected, the detected obstacle display processing unit 13 is activated to give a warning to the driver. At this time, the detected obstacle display processing unit 13 transmits an operation signal to the window display actuator 6, and the window display actuator 6 displays a warning on the front window 9 (see FIG. 1) based on the operation signal. Do.

  Specifically, when the window display actuator 6 receives the operation signal, the front window 9 is irradiated with predetermined light for warning display. For example, a frame-like light surrounding the obstacle image existing in front of the vehicle is superimposed on the front window 9 so that the driver's line of sight is directed toward the obstacle. The presence of the obstacle I1 is notified and a warning is given.

  By the way, in the present embodiment, in the storage device 8, a plurality of model images (hereinafter referred to as pattern matching data) reflecting obstacle characteristics (for example, color, edge information, luminance distribution, aspect ratio, distance distribution, etc.) are provided. The pattern is stored, and the obstacle extraction processing unit 11 detects the obstacle in the image by comparing the image data photographed by the obstacle detection camera 3 with the pattern matching data. .

  Further, in the present embodiment, the direction comparison processing unit 14 provided in the control device 10 receives the detection signal transmitted from the line-of-sight detection sensor 5 via the driver line-of-sight direction detection processing unit 15. The direction comparison processing unit 14 compares the driver's gaze direction detected by the gaze detection sensor 5 with the direction of the obstacle I1 detected by the obstacle direction detection processing unit 12 and targeted for notification. It has become.

  Then, when it is determined by comparison between the two that the direction deviation is equal to or greater than a predetermined value, it is determined that the driver is looking away due to a mismatch between the driver's line-of-sight direction and the direction of the obstacle I1. It has become. When the inconsistency level in the line-of-sight direction exceeds a predetermined value, an object existing in the driver's line-of-sight direction is estimated as an obstacle I2 (see FIG. 3), and this is added as new pattern matching data. It can be memorized.

Accordingly, next, the obstacle image extraction processing executed by the control unit 10, and, for data addition memory processing pattern matching further 4 will be described with reference to the flowchart of FIG.

Also not a, in step S1, reads various signals. These signals various sensors described above (2,4,5), read from and obstacle detection camera 3 and the like.
Next, in step S2, the obstacle extraction processing unit 11 executes an obstacle recognition process by comparing the captured image data with the pattern matching data in the storage device 8, and the degree of coincidence between the two is greater than or equal to a predetermined value. When it is, it is determined that there is an obstacle (step S3: YES), the process proceeds to step S4, and when the degree of coincidence between both is less than a predetermined value, it is determined that there is no obstacle. (Step S3: NO), the process returns to step S2.

When it is determined in step S3 that an obstacle is present, in the next step S4, the detected obstacle display processing unit 13 transmits an operation signal to the window display actuator 6, and the front window 9 (FIG. 1) Display a warning above.

  In step S5, the direction of the obstacle I1 seen from the automobile 1 is calculated by the obstacle direction detection processing unit 12, and here, the traveling direction of the automobile 1 and the direction in which the detected obstacle I1 is located are calculated. An angle difference θA is calculated.

  Next, the process proceeds to step S <b> 6, and based on the driver's line-of-sight direction detected by the line-of-sight detection sensor 5, the driver's line-of-sight direction detection processing unit 15 provided in the control device 10 detects the traveling direction of the automobile 1 and the detected direction. An angle difference θB (see FIG. 3) with the driver's line-of-sight direction is calculated.

  And it transfers to step S7, and based on the calculation result of angle difference (theta) A, (theta) B calculated in step S5, S6 mentioned above, the direction comparison process part 14 is the direction of the obstruction I1 used as alert | reporting object, The relative deviation from the driver's line-of-sight direction is calculated. For example, in the case shown in FIG. 3, the left side when the traveling direction of the automobile 1 is viewed is defined as a positive direction, and the right side is defined as a negative direction, and the absolute value of the difference between the two angle differences is calculated. In the case shown in the figure, the displacement of the direction = | θA − (− θB) | = θA + θB, resulting in the sum of the difference between the two angles.

  In the direction comparison processing unit 14, a threshold value for determining the magnitude of the above-described direction deviation is set in advance, and in step S7, the direction deviation θA + θB is compared with the threshold value α. . Here, when the deviation θA + θB in this direction exceeds the threshold value α (step S7: YES), the driver's line-of-sight direction does not match the direction of the obstacle I1, and the driver looks away. On the other hand, if it is less than or equal to the threshold value α (step S7: NO), it is determined that the driver is viewing the obstacle I1, and the process returns to step S2.

Furthermore, if the direction comparison processing unit 14 determines that the driver is looking away in step S7, the direction comparison processing unit 14 proceeds to step S8 and determines the inconsistency level of the line-of-sight direction. Specifically, direction comparing unit 14 measures the gaze direction of mismatch frequencies, i.e. the number per unit time viewing direction mismatch occurs N, continues and mismatch condition time T The magnitudes of these numerical values N and T are defined as the mismatch level.

Then, the direction comparison processing unit 14 compares these numerical values N and T with preset threshold values N0 and T0. Numerical N, when it exceeds the threshold value N0, T0 what Re or is preset T (Step S8: YES), another obstacle has not been detected as a warning target is present, the driver gazes it The direction comparison processing unit 14 transmits a determination result that the mismatch level is large to the temporary registration processing unit 16 provided in the control device 10.

  On the other hand, when both the numerical values N and T are equal to or less than the threshold values N0 and T0 (step S8: NO), the direction comparison processing unit 14 determines that the mismatch level is small and returns to the process of step S6. Subsequently, the calculation process of the driver's line-of-sight direction is executed.

  When it is determined in step S8 that the disagreement level is large, the temporary registration processing unit 16 determines the driver from the image data of the obstacle detection camera 3 at that time based on the driver's line-of-sight direction. The obstacle I2 (see FIG. 3) being watched is estimated. Then, temporary registration processing for temporarily registering and saving the image data including the obstacle I2 in the storage device 8 is executed (step S9).

  In addition to the image data, the temporary registration processing unit 16 receives a detection signal related to the travel environment (for example, travel environment data such as a rainfall state or a time zone) from the raindrop sensor 4 or the like via the travel environment detection processing unit 17. In step S9, this detection signal is stored as accompanying data.

Next, when the temporary registration process is executed, the process proceeds to step S10, and based on the detection signal of the vehicle speed sensor 2, whether or not the vehicle speed is 0, that is, the vehicle 1 is not in operation. whether is detected via the car both stop detection processing unit 18, when it is determined that the vehicle speed is zero (step S10: YES), the process proceeds to step S11 and subsequent steps shown in FIG. 5, the vehicle speed is 0 If it is determined that it is not (step S10: NO), the process returns to step S2.

Here, when it is determined in step S10 that the vehicle speed is 0, in step S11 in FIG. 5, the temporarily registered image display determination processing unit 19 provided in the control device 10 sends a predetermined operation signal to the display 7. As shown in FIG. 6A, the image data including the obstacle I2 that has been provisionally registered in step S9 is displayed.

  At this time, the display 7 is displayed so as to surround the obstacle I2 estimated by the temporary registration processing unit 16 in a frame shape, and subsequently the screen is switched to a screen as shown in FIG. In the display screen shown in FIG. 6B, only the image data of the obstacle I2 is extracted from the image data and displayed on the display unit X1 in the confirmation display unit 7a, and accompanying data corresponding to the obstacle I2 is displayed. Is displayed on the display unit X2.

  On the display 7, the temporary registration image display determination processing unit 19 simultaneously displays the touch panel type operation button display units X 3, X 4, and X 5 constituting the regular registration operation input unit 7 b in addition to the confirmation display unit 7 a. It is like that.

  By the way, in the storage device 8, for each stored obstacle (pattern matching data), the danger level and type of the obstacle are set. Types include, for example, pedestrians (children), pedestrians (adults), pedestrians (elderly persons), motorcycles, etc., and pedestrians (children) → pedestrians (elderly persons) → pedestrians (adults) → The degree of risk is set so that the frequency decreases in the order of the two-wheeled vehicle.

  Here, in FIG. 6B, “pedestrian (child)” is displayed by default in the “type” column of the display unit X2 of the regular registration operation input unit 7b, and “1” is displayed in the “risk level” column. "Is displayed by default.

  The driver is allowed to determine whether or not the display content about the type and the risk level corresponds to the obstacle I2 in the accompanying data described above, and as a result, the content of the default display is If it is determined that it corresponds accurately, the operation button display unit X3 on which “Register” is displayed is pressed.

  At this time, the regular registration determination processing unit 20 provided in the control device 10 determines that it is not necessary to change the content of the default display, and the regular registration processing unit 21 determines the obstacle I2 based on this determination. The image data is additionally stored as new pattern matching data.

Specifically, the regular registration determination processing unit 20 detects whether or not the operation button display unit X3 has been operated by the driver in step S12. If it is determined that the operation of the operation button display section X3 is made (step S12: YES), the process proceeds to step S13, temporarily registered image data, and stores the associated data in the default display of the content device 8 A regular registration process for saving (additionally storing) regular registration is executed, and the process proceeds to step S14 described later.

  On the other hand, if the driver determines that the default displayed content does not accurately correspond to the obstacle I2, the driver operates the operation button display unit X4 on which “data correction” is displayed, and performs control. If it is determined that the device 10 detects an obstacle but does not need to be recognized as an obstacle, the driver presses the operation button display unit X5 on which “delete” is displayed.

  Here, if it is determined that the content of the default display does not correspond accurately, the driver can correct the display content by operating the operation button display unit X4. If the regular registration determination processing unit 20 determines in step S12 that the regular registration operation has not been performed (step S12: NO), the process proceeds to step S15, and whether the operation button display unit X4 has performed an operation for correcting accompanying data. Determine whether or not.

If it is determined in step S15 that the accompanying data correction operation has been performed (step S15: YES), the regular registration determination processing unit 20 proceeds to step S16, and overwrites and corrects the accompanying data based on the content operated by the driver. To do. Then, the normal registration processing unit 21 executes the regular registration process for normalizing registers store additional data after the image data and correction of obstruction I2 in the storage device 8 proceeds to step S14.

  Further, when the regular registration determination processing unit 20 determines that the accompanying data correction operation is not performed in step S15 (step S15: NO), it is determined that the driver has pressed the remaining operation button display unit X5. Then, the process proceeds to step S17. In step S17, the regular registration determination processing unit 20 executes a temporary registration data deletion process in which the operation button display unit X5 is operated, and the process proceeds to step S14.

Thus, for a single provisional registration data, but will be something Re one of the processes of steps S13, S16, S17 are executed, when a plurality of data are temporarily registered in the vehicle traveling What Re or processing of sequential steps S13, S16, S17 for each provisional registration data is performed.

  In step S <b> 14, when the above-described processing for one temporary registration data is executed, it is determined one by one whether or not all the temporary registration data is displayed on the display 7.

  Here, if there is temporary registration data that has not yet been displayed and the registration or deletion operation for all the image data has not been completed (step S14: NO), the process returns to step S11 to return to the temporary registration image display determination processing unit. 19 displays the next temporary registration data that has not been displayed on the display 7 and repeats the processes of steps S11 to S17. When all the image data is displayed and the registration or deletion operation is completed (step S14: YES), the process is returned.

  In the storage device 8, as shown in FIG. 2, an existing registration data area and an additional registration data area are set. In the existing registration data area, data preset by the manufacturer at the time of manufacturing the automobile 1 is stored, and the data stored by the additional storage described above is stored in the additional registration data area.

Therefore, in step S2 described above, the existing registration data, and both the additional registration data added registered so far have become available as the pattern matching data.

  As described above, in this embodiment, when the mismatch level between the direction of the obstacle targeted for notification and the driver's line-of-sight direction is equal to or higher than a predetermined value, the obstacle I2 in the driver's line-of-sight direction is newly set. Therefore, the database can be enhanced by accumulating the pattern matching data in the storage device 8, and the obstacle recognition accuracy can be improved.

  In addition, in steps S12 to S17, the driver is made to determine whether or not additional registration of temporary registration data is necessary, and the processing related to the additional storage is performed based on the determination result of the driver. Only the data for pattern matching that is considered necessary for the user can be stored, and the accuracy of obstacle recognition in the control device 10 can be improved.

  In addition, when the pattern matching data is additionally stored, the information on the driving environment stored in step S2 is also stored as accompanying data. However, when the pattern matching data is read out, the driving at that time is stored. The pattern matching data corresponding to the environment can be narrowed down and read out.

  For example, when it rains, pedestrians wear umbrellas and wear rain pants, so pattern matching data peculiar to rain is necessary.In the evening and at night, obstacles and surrounding scenery Since there is a special situation that it becomes difficult to distinguish, pattern matching data corresponding to the time zone is also required. In the present embodiment, it is possible to finely perform obstacle recognition by reading pattern matching data used for recognition according to such changes in the driving environment, thereby improving the obstacle recognition accuracy. Can do.

  Further, in step S11, the image data when the line-of-sight direction does not match is displayed to the driver, and the accompanying data related to the risk determination result of the obstacle I2 included in the image data is additionally stored in the storage device 8. At the time of pattern matching, it becomes possible to determine the degree of danger together, thereby improving the recognition accuracy of the obstacle.

  For this reason, when obstacles such as pedestrians, bicycles, motorcycles, etc. appear at the same time while the vehicle is running, warnings are displayed in order from the obstacles with the highest risk, or a predetermined number of the objects with the highest risk are recognized Warning according to the degree of danger.

  In addition, the additional data related to the type determination result of the obstacle I2 is additionally stored in the storage device 8, so that the obstacle can be recognized according to the characteristics of the type at the time of pattern matching, and the recognition accuracy can be improved. Can do.

For example, the obstacle is a pedestrian, bicycle, what Re in which was the one of the bikes, and even, in the case of a pedestrian, the pedestrian it is you are walking while holding down whether a child, a baby carriage and cart, etc. Obstacles can be recognized according to the characteristics of the type, such as whether it is a guardian or an elderly person, and the recognition accuracy can be improved.

  Further, in the present embodiment, in steps S15 and S16, it is possible to correct each of the accompanying data by the driver, and finally the risk and type are determined by the driver. It becomes possible to recognize the degree and type more accurately, thereby improving the recognition accuracy of the obstacle.

  In addition, in step S10 after temporarily registering the image data when the line-of-sight direction does not match, when it is first detected that the vehicle speed is 0, the temporary registration data display process is executed. Confirmation of display contents during driving of the vehicle can be reliably avoided and safety during driving can be ensured. Furthermore, before the driver forgets the presence of the obstacle, the process related to the additional storage can be executed.

  In the above-described embodiment, the provisional registration processing unit 16 estimates the presence of the obstacle I2, and the image data of the obstacle I2 thus estimated is set as a target for additional storage. Is not necessarily limited to this.

  For example, as shown in FIGS. 7 and 8, a known input interface such as a touch pen 50 is separately provided, and the driver is obstructed as an additional storage target for image data displayed on the display 7 using the touch pen 50. The object I2 may be directly designated.

  In this case, the coordinates of the display 7 touched by the driver with the touch pen 50 can be transmitted as an operation signal to the temporary registration screen display determination processing unit 19 ′ of the control device 10.

  As shown in FIG. 8A, the driver draws a frame that surrounds the obstacle I2 on the display 7 with the touch pen 50 so that the obstacle I2 can be designated as an additional storage target. It has become. As shown in FIG. 8B, the temporary registration screen display determination processing unit 19 ′ extracts and displays an image within the frame drawn by the touch pen 50, and the extracted image data needs to be additionally stored. Let the driver determine NO.

  FIG. 9 is a flowchart for explaining another method for additionally storing an obstacle image in the driver's line of sight as pattern matching data. The provisional registration screen display determination processing unit 19 ′ determines whether or not the driver has designated the obstacle I2 with the touch pen 50 in step S52 after the process of step S51 corresponding to step S11 of FIG. judge. If it is determined that the driver has specified the obstacle I2 (step S52: YES), the processing of steps S53 to S58 corresponding to steps S12 to S17 in FIG. 5 is executed to specify the obstacle I2. If not (step S52: NO), the process of step S52 is repeated, and the operation state of the touch pen 50 by the driver is continuously monitored.

By the way, in embodiment mentioned above, when the vehicle speed sensor 2 detects that the vehicle speed is 0, the temporary registration data display process is executed, but the present invention is not necessarily limited to this. For example, when further consideration for safety during driving is required, display during a pause such as waiting for a signal may be avoided. For example , an operation of a side brake (not shown) is detected. Thus, the process corresponding to step S10 may be executed based on the detection result.

  Further, when it is detected that the key position is ACC (accessory) during the engine stop operation, the key operation may be forcibly stopped, and the processing corresponding to step S11 may be executed under this state. . Alternatively, after the engine has been stopped once, the process corresponding to step S11 may be executed when the engine start operation is performed next time.

  In the embodiment described above, the regular registration operation input unit 7b is configured by the operation button display units X3 to X5 displayed on the display 7. For example, a dedicated operation button is provided on the outer frame portion of the display 7. It may be arranged.

As for the display 7, for example , a display of a navigation device that guides the route to the set destination while displaying the travel route of the automobile 1 together with the map information may be used. A display may be provided.

  Further, when the automobile 1 includes the above-described navigation device, for example, as shown in FIG. 10, the temporary registration processing unit 16 can receive a detection signal from the navigation device 60 via the traveling environment detection processing unit 17 ′. The vehicle position information transmitted from the navigation device 60 may be stored as accompanying data.

For example, forest roads covered with trees and places where sunlight is blocked, such as in tunnels, are always dark, even during the day . Difficult to distinguish between obstacles and surrounding scenery. Therefore, as described above, if the positional information data from the navigation device 60 is additionally stored as accompanying data, the pattern matching pattern is used in accordance with the change in the location where the automobile 1 is traveling based on the positional information from the navigation device 60. Data can be read out, thereby improving obstacle recognition accuracy.

  In addition, in other embodiment shown in FIGS. 7-10, about the component similar to the first embodiment demonstrated with reference to FIGS. 1-6, the same code | symbol is attached | subjected and the detailed description is abbreviate | omitted. To do.

  In the above-described embodiment, in steps S16 and S17, the data can be corrected and deleted as temporary registration data before being additionally stored. However, the existing registration data in the storage device 8 can also be corrected. Such correction and deletion may be performed.

In correspondence between the configuration of the present invention and the above-described embodiment,
The obstacle notification means of the present invention corresponds to the window display actuator 6,
Similarly,
The line-of-sight direction detection means corresponds to the line-of-sight detection sensor 5,
The mismatch level detection means corresponds to the direction comparison processing unit 14 of the control device 10 that executes step S8,
The storage means corresponds to the storage device 8,
The imaging means corresponds to the obstacle detection camera 3,
The means for presenting the photographed image to the driver corresponds to the confirmation display portion 7a of the confirmation display / input operation combined display 7,
Means for determining the necessity of additional storage corresponds to the regular registration operation input unit 7b of the confirmation display / input operation combined display 7.
The present invention is not limited only to the configuration of the above-described embodiment, and many embodiments can be obtained.

1 is a schematic diagram of an automobile that employs a vehicle driving support apparatus according to an embodiment of the present invention. The system block diagram of the driving assistance device for vehicles. Explanatory drawing for demonstrating the method of comparing the direction of an obstruction, and a driver | operator's gaze direction. The flowchart explaining the method of extracting the obstruction image of a driver | operator's gaze direction. The flowchart explaining the method of additionally storing the obstacle image of a driver | operator's gaze direction as pattern matching data. The figure which shows the example of a display at the time of additionally storing the obstacle image of a driver | operator's gaze direction as data for pattern matching. The system block diagram of the driving assistance device for vehicles concerning other embodiments of the present invention. The figure which shows the operation example at the time of additionally storing the obstacle image of a driver | operator's gaze direction as pattern matching data. The flowchart explaining the method of additionally storing the obstacle image of a driver | operator's gaze direction as pattern matching data in other embodiment of this invention. The system block diagram of the driving assistance device for vehicles concerning other embodiments of the present invention.

3. Obstacle detection camera (imaging means)
5: Gaze detection sensor (gaze direction detection means)
6 ... Window display actuator (obstacle notification means)
7: Display for both confirmation display and input operation 8: Storage device (storage means)
10: Control device 14: Direction comparison processing unit (mismatch level detection means)
19, 19 '... Temporarily registered image display determination processing unit
S9: Estimating means

Claims (8)

A vehicle driving support device that informs the driver of the presence of an obstacle,
Obstacle notifying means for detecting an obstacle by comparing it with stored pattern matching data and notifying the driver's line of sight toward the obstacle,
Gaze direction detection means for detecting the driver's gaze or face direction;
Detecting the inconsistency level of the line-of-sight direction based on the inconsistency frequency indicating the frequency that the direction of the obstacle to be notified by the obstacle informing means and the line-of-sight direction of the driver do not match, or the mismatch duration time A mismatch level detection means for
An estimation means for estimating an obstacle in the gaze direction of the driver when the gaze direction mismatch level is a predetermined value or more ;
Vehicular driving support apparatus and a storage means for adding storing estimated the obstacle as a new pattern matching data. Imaging means for photographing the periphery of the vehicle;
Means for presenting a driver with a photographed image taken when the mismatch level is equal to or higher than a predetermined value;
Means for allowing the driver to determine whether additional storage is necessary,
The vehicle driving support device according to claim 1, wherein the process related to the additional storage is executed based on a result of the determination. Imaging means for photographing the periphery of the vehicle;
Means for presenting a photographed image taken when the mismatch level is equal to or higher than a predetermined value to the driver,
The vehicle driving support device according to claim 1, wherein, at the time of the additional storage, the obstacle degree determination result of the obstacle is stored together with the obstacle image in the line-of-sight direction included in the photographed image presented to the driver. The vehicle driving support device according to claim 3, wherein the determination of the degree of risk is made by a driver. Imaging means for photographing the periphery of the vehicle;
Means for presenting a photographed image taken when the mismatch level is equal to or higher than a predetermined value to the driver,
The vehicular driving support apparatus according to claim 1, wherein, at the time of the additional storage, the obstacle type determination result of the obstacle is stored together with the obstacle image in the line-of-sight direction included in the photographed image presented to the driver. The vehicle driving support device according to claim 5, wherein the determination of the type is made by a driver. The vehicle driving support device according to any one of claims 2 to 6, wherein the image presentation to the driver is executed when the vehicle stops for the first time after the mismatch level becomes a predetermined value or more. At the time of the additional storage, the vehicle traveling environment when the inconsistency level becomes a predetermined value or more is also stored.
The vehicle driving support device according to claim 1, wherein when the pattern matching data is read, the data is read according to a driving environment.

Images