JP5604210B2 - Detection device and detection method - Google Patents

Description

  The present invention relates to a technique for detecting a shift in the mounting position of an imaging unit mounted on a vehicle.

  In recent years, in an apparatus, it is determined whether or not a vehicle deviates from a lane marking that divides a traveling lane on a road surface based on an image outside the vehicle acquired by an in-vehicle imaging unit. A technology for informing the effect has been developed. According to this technology, the user who drives the vehicle can avoid a collision between the host vehicle and a vehicle traveling beyond the lane marking, and can prevent an accident.

  This technique, as a premise, will not be effective unless the imaging unit is mounted at an appropriate position in the vehicle. In other words, if the mounting position of the imaging unit mounted on the vehicle is inappropriate, it is determined that the vehicle has deviated from the lane line even though the vehicle has not deviated from the lane line, and the user is notified erroneously. There is a risk of doing.

  In order to solve this problem, in the apparatus, a technique for detecting a displacement of the mounting position of the imaging unit based on an object reflected in an image outside the vehicle acquired by the vehicle-mounted imaging unit (hereinafter, this detection is referred to as “deviation detection”). Can be considered.

  As such a technique, for example, Patent Document 1 describes a technique in which a device detects a displacement based on a sign shown on a road surface reflected in an image outside the vehicle.

Japanese Patent Laid-Open No. 08-285534

  However, when the vehicle travels on a specific travel path, in order to detect the displacement based on the object shown in the image outside the vehicle acquired by the in-vehicle imaging unit, that is, the sign shown on the specific road surface reflected in the image outside the vehicle, There is a risk of misdetection.

  In other words, when the vehicle travels on a specific traveling road, for example, a sharp curve, an uphill road or a downhill road with a severe inclination angle, the shape of the road surface reflected in the outside image acquired by the in-vehicle imaging unit is linear. In addition to the shape, the signs shown on the road surface are also deformed according to this road surface, and if the device detects displacement based on the deformed sign, the imaging unit is mounted at an appropriate position However, the problem of misdetection occurs.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique for preventing misalignment detection in an apparatus.

  In order to solve the above-mentioned problem, the invention of claim 1 is a detection device that detects a shift in the mounting position of an imaging unit mounted on a vehicle, and a storage unit that stores a reference size of a lane marking on a road surface; The size of the lane marking on the road surface is derived based on input means for inputting an outside image including a road surface around the vehicle acquired by the imaging unit, and the image of the road surface on the vehicle side reflected in the image outside the vehicle. Deriving means, comparing means for comparing the size derived by the deriving means with the reference size stored in the storage means, and detecting means for detecting the deviation based on a comparison result of the comparing means And a notification means for notifying the user of the difference when the deviation is detected.

  The invention according to claim 2 is the detection device according to claim 1, wherein the comparison result is obtained when the left-right center of the inputted outside image does not coincide with the left-right center of the vehicle reflected in the outside image. It further has the 1st invalidation means which invalidates.

  The invention according to claim 3 is the detection device according to claim 1 or 2, wherein the detection means detects the deviation while the vehicle is traveling.

  According to a fourth aspect of the present invention, in the detection device according to any one of the first to third aspects, the partition line is a broken line, and the reference size is the length of the line in the broken line and the length of the interval. And a length including a line and an interval.

  The invention according to claim 5 is the detection device according to any one of claims 1 to 3, wherein the partition line is a solid line, and the reference size is a width of the solid line. .

  The invention according to claim 6 is the detection device according to any one of claims 1 to 5, wherein detection means for detecting an angle of the vehicle with respect to a horizontal direction, and a traveling path of the vehicle based on the angle. And a gradient determining means for determining whether or not is a gradient, and further comprising a second invalidating means for invalidating the comparing means when the gradient is determined by the gradient determination. To do.

  The invention according to claim 7 is the detection device according to any one of claims 1 to 6, further comprising departure determining means for determining whether or not the vehicle departs from the lane marking. The notification means notifies the user of the fact when the vehicle deviation is determined.

  The invention according to claim 8 is a detection method for detecting a displacement of a mounting position of an imaging unit mounted on a vehicle, and (a) an image outside the vehicle including a road surface around the vehicle acquired by the imaging unit. An input step, (b) a step of deriving the size of the lane marking on the road surface based on an image of the road surface on the vehicle side shown in the outside image, and (c) derived in the step (b). (D) a step of detecting the deviation based on a comparison result of the step (c), and (e) And a step of notifying the user when the deviation is detected.

  According to the first to eighth aspects of the present invention, the detection device derives the size of the lane marking on the road surface based on the image of the road surface on the vehicle side reflected in the image outside the vehicle including the road surface around the vehicle acquired by the imaging unit. And comparing the derived size of the lane marking with the reference size of the lane marking on the road surface stored in advance in the storage means to detect the displacement of the mounting position of the imaging unit and detect the displacement Inform the user to that effect. For this reason, it is possible to detect the misalignment of the mounting position of the imaging unit without making a mistake, and to inform the user of that fact.

  According to the invention of claim 2, the detection device detects a shift in the mounting position of the imaging unit mounted on the vehicle when the left-right center of the image outside the vehicle coincides with the left-right center of the vehicle shown in the image outside the vehicle. Therefore, it is possible to detect without misalignment of the mounting position in the horizontal direction and the vertical direction of the imaging unit.

  According to the invention of claim 3, since the detection device detects a shift in the mounting position of the imaging unit mounted on the vehicle while the vehicle is traveling, it is necessary to prepare settings and environment for the detection. The convenience of the user can be improved.

  According to the invention of claim 4, the detection device can select the reference size to be compared as any of the length of the line in the broken line, which is a partition line, the length of the interval, and the length of the line and the interval. In this way, since the shift of the mounting position of the imaging unit mounted on the vehicle is detected, the shift detection can be performed without error.

  According to the invention of claim 5, the detection device detects the shift in order to detect the shift of the mounting position of the image pickup unit mounted on the vehicle with the reference size to be compared being the width of the solid line that is the lane marking. It can be done without error.

  According to a sixth aspect of the present invention, when the vehicle travels on a sloped road, the detection device invalidates the deviation detection, so that the imaging unit is installed at an appropriate position. It is possible to avoid the possibility of detecting the deviation and to prevent the deviation detection accuracy from being lowered.

  According to the invention of claim 7, since the detection device has a displacement detection function, it is possible to appropriately function the function of determining the vehicle departure based on the vehicle outside image input from the imaging unit.

FIG. 1 is a diagram illustrating a vehicle. FIG. 2 is a diagram illustrating an in-vehicle device. FIG. 3 is a system block diagram of the in-vehicle device system. FIG. 4 is a diagram illustrating a method of mounting the imaging unit on the vehicle. FIG. 5 is a diagram illustrating a traveling state of the vehicle. FIG. 6 is a diagram illustrating a control flow of the in-vehicle device. FIG. 7 is a diagram illustrating a control flow of the in-vehicle device. FIG. 8 is a diagram showing an image outside the vehicle. FIG. 9 is a diagram illustrating a method for recognizing a sign shown on a road surface reflected in an image outside the vehicle. FIG. 10 is a diagram illustrating a map for determining vehicle departure. FIG. 11 is a diagram illustrating a control flow of the in-vehicle device. FIG. 12 is a diagram illustrating a control flow of the in-vehicle device. FIG. 13 is a diagram illustrating the size of the lane markings. FIG. 14 is a diagram illustrating the size of the lane markings. FIG. 15 is a diagram showing the size of the lane markings.

<Representative embodiment>
In the representative embodiment, the vehicle departure determination function determines whether a running vehicle has deviated from the road marking line based on the road marking line reflected in the image outside the vehicle acquired by the in-vehicle imaging unit. ”And a detection device having a“ deviation detection function ”for detecting deviation of the mounting position of the imaging unit mounted on the vehicle, that is, an in-vehicle device will be described as an example.

  The in-vehicle device will be described below in the order of “configuration of the in-vehicle device system”, “vehicle departure determination control”, and “deviation detection control” with reference to the accompanying drawings.

<Configuration of in-vehicle device system>
The configuration of the in-vehicle device system will be described based on FIG. 1, FIG. 2, and FIG. As shown in FIGS. 1 and 3, the in-vehicle device system includes an in-vehicle device 2 mounted on a vehicle 1, an imaging unit 6, an acceleration sensor 7, an operation unit 8, a vehicle speed sensor 9, a navigation device 10, and a gyro. A sensor 11 and the like are electrically connected via a signal line.

  The in-vehicle device 2 is mounted in a dashboard in front of the passenger seat of the vehicle 1.

  As shown in FIG. 2, the imaging unit 6 includes a lens 13 and an imaging device, and can acquire image data electronically. The imaging unit 6 includes an imaging unit main body 14 and a mounting device 16. The mounting device 16 forms an adhesive surface 12 for mounting the imaging unit 6 by being bonded to the windshield 17 of the vehicle 1. The mounting method is performed by sticking one side of the adhesive sheet coated on both sides with an adhesive to the adhesive surface 12 of the mounting device 16 and the other side to the windshield 17 of the vehicle 1.

  Furthermore, the mounting instrument 16 forms a support shaft 15 that combines the mounting instrument 16 and the imaging unit body 14 while allowing the imaging unit body 14 to be adjusted in the vertical direction. The adjustment method is performed by fastening the adjustment screw of the support shaft 15 and fixing the imaging unit main body 14 at a desired angle.

  Details of the mounting position of the imaging unit 6 will be described with reference to FIG. FIG. 4 is an enlarged view of region W of vehicle 1 shown in FIG. As shown in FIG. 4, the mounting position of the imaging unit 6 in the vehicle 1 is such that the imaging direction faces the outside of the vehicle in the vicinity of the rearview mirror 18 mounted on the windshield 17 inside the windshield 17 of the vehicle 1. Installed. Specifically, the vehicle 1 is mounted at the center of the left and right of the windshield 17 and below the upper end of the windshield 17 by a predetermined distance (for example, 10 cm), and the imaging direction of the imaging unit 6 is mounted at a predetermined angle (for example, 15 degrees). Is done.

  As a result, the imaging unit 6 acquires an image outside the vehicle that shows the front side of the vehicle 1. The mounting angle of the lens 13 included in the imaging unit 6 is such that the end portion of the bonnet of the vehicle 1 in the vehicle outside image captured by the imaging unit 6 is as shown in FIG. It is set so as to be reflected, and a lane marking that divides the traveling lane on the road surface in the traveling direction of the vehicle 1 is set to be reflected in a predetermined area.

  Note that the imaging unit 6 is mounted on the vehicle 1 at the manufacturing factory of the vehicle 1 or a vehicle sales company by the mounting method and the adjustment method as described above. Is called. However, the mounting position may become inappropriate due to a decrease in the adhesive force of the adhesive sheet with the passage of time, a decrease in the fixing force of the screw on the support shaft 15, or a collision due to an accident of the vehicle 1 or the like.

  The acceleration sensor 7 outputs acceleration indicating the magnitude of impact applied to the vehicle 1 to the in-vehicle device 2 as an acceleration signal. The acceleration sensor 7 is installed in the engine room of the vehicle 1.

  The operation unit 8 receives a user operation and outputs an operation signal indicating the operation content. The operation unit 8 is mounted in a dashboard in front of the passenger seat of the vehicle 1.

  The vehicle speed sensor 9 outputs the speed at which the wheel rotates when the vehicle 1 moves to the in-vehicle device 2 as a vehicle speed signal. The vehicle speed sensor 9 is installed in the engine room of the vehicle 1.

  The navigation device 10 includes a GPS signal receiving antenna, map data, a liquid crystal display, and the like. The navigation device 10 is provided with a display screen provided on the navigation device 10 on the upper part of the center console between the driver seat and the passenger seat of the vehicle 1.

  The gyro sensor 11 outputs a signal indicating an angle with respect to the horizontal direction applied to the vehicle 1 to the in-vehicle device 2. The gyro sensor 11 is installed in the engine room of the vehicle 1.

  The in-vehicle device 2 and the imaging unit 6 are electrically connected by a communication cable as shown in FIG. In the in-vehicle device 2, an in-vehicle image in the traveling direction of the vehicle 1 captured by the imaging unit 6 can be input in the in-vehicle device 2 via a communication cable.

  The in-vehicle device 2 and the acceleration sensor 7 are electrically connected by a communication cable. In the in-vehicle device 2, the acceleration signal output from the acceleration sensor 7 can be input via a communication cable.

  The in-vehicle device 2 and the operation unit 8 are electrically connected by a communication cable. In the in-vehicle device 2, a signal indicating the start and end of the operation unit 8 that accepts an operation for starting and ending the function of the in-vehicle device 2 can be input via a communication cable.

  The in-vehicle device 2 and the vehicle speed sensor 9 are electrically connected by a communication cable. In the in-vehicle device 2, the vehicle speed signal output from the vehicle speed sensor 9 can be input via a communication cable.

  The in-vehicle device 2 and the navigation device 10 are electrically connected by a communication cable. In the in-vehicle device 2, navigation information included in the navigation device 10 can be input via a communication cable.

  The in-vehicle device 2 and the gyro sensor 11 are electrically connected by a communication cable. In the in-vehicle device 2, the angular velocity signal output from the gyro sensor 11 can be input via a communication cable.

  The in-vehicle device 2 includes a volatile storage unit 3, a control unit 4, and a sound output unit 5 connected by a signal line such as a bus. Moreover, the vehicle-mounted apparatus 2 inputs the vehicle exterior image received from the imaging part 6 via a communication cable to the control part 4 via a signal line. Furthermore, the in-vehicle device 2 receives an acceleration signal received from the acceleration sensor 7, an operation signal received from the operation unit 8, a vehicle speed signal received from the vehicle speed sensor 9, navigation information received from the navigation device 10 via a communication cable, and The angular velocity signal received from the gyro sensor 11 is input to the control unit 4 via the signal line.

  The control unit 4 is a microcomputer including a calculation unit 4a (for example, a CPU) and a nonvolatile storage unit 4b (for example, a ROM). The control unit 4 causes the vehicle departure determination function, the deviation detection function, and the like to be exhibited when the calculation unit 4a executes the control program stored in the nonvolatile storage unit 4b. The details of the vehicle departure determination function and the deviation detection function will be described later.

  The volatile storage unit 3 is a volatile memory such as a RAM, for example, and functions as a working area that temporarily stores parameters and the like when the control unit 4 executes a control program.

  The sound output unit 5 is a buzzer, for example, and exhibits a function of receiving a signal transmitted when the first storage unit makes a vehicle departure determination and outputting a sound.

  Next, a vehicle departure determination function and a deviation detection function realized by the control unit 4 will be described.

  First, the vehicle departure determination function in the control unit 4 will be described based on the control flow shown in FIG. 6 and FIG.

  Each flag shown in the subsequent control is turned off in the initial process and the end process in the control, and turned off in the reset process of the in-vehicle device 2.

<Vehicle departure determination control>
The vehicle departure determination control executed by the control unit 4 will be described based on the control flow of FIG.

  The control unit 4 executes the control flow shown in FIG. 6 at a predetermined cycle when a signal indicating that the function is started is received from the operation unit 8 by a user operation. Therefore, the control part 4 transfers to step S1 of the control flow shown in FIG. 6 for every predetermined period.

  In step S1, the control unit 4 exhibits a departure determination function. Details of the vehicle departure determination process executed by the control unit 4 will be described in detail based on the control flow shown in FIG. Next, the control part 4 transfers to step S4 in the control flow shown in FIG.

  In step S4, the control unit 4 inputs an image outside the vehicle including the road surface acquired by the imaging unit 6 as shown in FIG. 8, and a lane line that divides the traveling lane indicated on the road surface included in the input vehicle outside image, That is, it is determined whether or not the right lane marking RL and the left lane marking LL are recognized. When it determines with having recognized the lane marking, it transfers to step S5 (in step S4 YES). If it is not determined that the lane marking has been recognized, the process proceeds to return (NO in step S4). That is, the process proceeds to step S2 in the control flow shown in FIG.

  A method of recognizing the lane markings by the control unit 4 will be described in detail. When the outside image that is a color image is converted into a monochrome outside image, the control unit 4 detects the difference in brightness indicated by each of the plurality of pixels that constitute the outside image, and recognizes the partition line in the outside image. To do. That is, the control unit 4 detects an edge when the difference in brightness indicated by each of the plurality of pixels is high. Then, as shown in FIG. 9, when the width L between the edge H1 and the edge H2 that are close to each other matches the width that specifies the road marking line preset in the nonvolatile storage unit 4b, It is determined that it has been recognized.

  In step S <b> 5, the control unit 4 determines the lane markings that divide the road lane in the traveling direction of the vehicle 1, as shown in FIG. 8. Included in the region.

  The predetermined region shown in the vehicle outside image is a region including a position at least 2 m away from the center C on the front end of the front end of the bonnet B of the vehicle 1 as shown in FIG. Thereby, as shown in FIG. 8, in the image outside the vehicle, the left lane line LL and the right lane line RL indicating the boundary between the road lane on which the vehicle 1 travels and the adjacent travel lane are on both outer sides of the bonnet B. Will be shown.

  As shown in FIG. 8, the control unit 4 uses the outside image to partition the center point C at the end of the hood B of the vehicle 1 included in the outside image and the road lane included in the outside image. Of the left lane line LL and the right lane line RL that is closer to the vehicle and calculates the relative distance D from the point P of the lane line RL when virtually extending to the right front wheel of the vehicle 1 To do. Next, the process proceeds to step S6.

  In step S6, the control unit 4 calculates an acceleration G when the vehicle 1 moves in the lateral direction based on the relative distance D calculated this time and the relative distance D calculated in the past. Next, the process proceeds to step S7.

  In step S <b> 7, the control unit 4 determines whether or not the vehicle 1 deviates from a lane marking that divides the traveling lane on the road surface in the traveling direction, that is, a departure determination by the vehicle departure determination function. When the control unit 4 performs the departure determination, the process proceeds to step S8 (YES in step S7). If the departure determination is not performed, the process proceeds to return (NO in step S7). That is, the process proceeds to step S2 in the control flow shown in FIG.

  The details of the vehicle departure determination function executed by the control unit 4 will be described. The control unit 4 determines whether or not the vehicle 1 deviates from the lane marking based on the calculated relative distance D and the lateral movement speed G that is the lateral movement speed.

  The control unit 4 refers to a map as shown in FIG. 10 stored in advance in the nonvolatile storage unit 4b provided, and a position where the relative distance D and the lateral movement speed G that is the lateral movement speed intersect. Ask for. And the control part 4 performs vehicle departure determination based on whether the said position is contained in the vehicle departure warning area W shown in a figure.

  Here, the map shown in FIG. 10 will be described. When the relative distance D between the vehicle 1 and the right lane line RL and the lateral movement speed G are located in the vehicle departure warning area W in the map, the possibility that the vehicle 1 departs from the right lane line RL is high. Or it is a case where the vehicle 1 has deviated from the right lane marking RL. That is, this vehicle departure is a concept that includes both the case where the vehicle 1 is likely to deviate from the lane line and the case where the vehicle 1 actually deviates from the lane line.

  When the possibility that the vehicle 1 departs from the right lane marking RL is high, the lateral movement speed G is relatively small and the relative distance D is relatively small, or the lateral movement speed G is relatively large and the relative distance D is large. Such a case is relatively large, and such a case is set in the vehicle departure warning area W.

  On the other hand, when the relative distance D is relatively large and the lateral movement speed G is relatively small, or when the relative distance D is relatively small and the lateral movement speed G is relatively large, the lane marking may be exceeded unexpectedly. Therefore, such an area in the map is not set as the vehicle departure warning area W.

  That is, the setting of the departure warning area W is set based on a human sense that the vehicle 1 seems to unexpectedly exceed or exceed the lane marking. Next, the process proceeds to step S8.

  In step S <b> 8, the control unit 4 sets the warning flag to ON in the volatile storage unit 3. Next, the process proceeds to return. That is, the process proceeds to step S2 in the control flow shown in FIG.

  In step S <b> 2, the control unit 4 determines whether or not the alarm flag is turned on in the volatile storage unit 3. If it is determined that the alarm flag is turned on, the process proceeds to step S3 (YES in step S2). If it is not determined that the alarm flag is turned on, the process proceeds to return (NO in step S2).

  In step S3, the control unit 4 controls the sound output unit 5 to output a sound. Thereby, the control part 4 can notify a driver | operator etc. that the vehicle 1 may deviate from the lane line, or deviated from the lane line.

  Next, the displacement detection function in the in-vehicle device 2 will be described based on the control flow of FIG. 11 and FIG.

<Displacement detection control>
The deviation detection control executed by the control unit 4 will be described based on the control flow of FIG.

  When receiving a signal indicating that the function is to be started from the operation unit 8 by a user operation, the control unit 4 executes the control flow shown in FIG. Therefore, the control part 4 transfers to step S9 of the control flow shown in FIG. 11 for every predetermined period.

  The control unit 4 processes the control shown in FIG. 11 and the control shown in FIG. 6 in parallel by multitask control.

  In step S9, the control unit 4 exhibits a deviation detection function. Details of the deviation detection processing executed by the control unit 4 will be described in detail based on the control flow shown in FIG. Next, the control part 4 transfers to step S13 in the control flow shown in FIG.

  In step S <b> 13, the control unit 4 determines whether or not the travel path on which the vehicle 1 travels is a gradient based on the navigation information received from the navigation device 10 or the angle signal received from the gyro sensor 11. That is, it is determined whether or not the travel path on which the vehicle 1 travels is a slope. If it is determined that the traveling road is a slope, the process proceeds to return (YES in step S13). As a result, the subsequent processing is avoided, and the parcel line size comparison processing is invalidated. And it transfers to step S10 in the control flow shown in FIG. If it is not determined that the gradient is present, the process proceeds to step S14 (NO in step S13).

  The case where the control unit 4 determines whether or not the travel path on which the vehicle 1 travels is a slope is a case where navigation information indicating that the travel path is a slope is received from the navigation device 10. Or it is a case where based on the angle signal received from the gyro sensor 11, the vehicle 1 determines more than a predetermined angle (for example, 5 degree | times) with respect to a horizontal direction.

  In step S14, as described above, the control unit 4 inputs an image outside the vehicle including the road surface acquired by the imaging unit 6 as shown in FIG. 8, and the road surface included in the predetermined region X of the input image outside the vehicle. It is determined whether or not the size of the lane marking is recognized based on the image. When it determines with having recognized the lane marking, it transfers to step S15 (in step S14 YES). If it is not determined that the lane marking has been recognized, the process proceeds to return (NO in step S14). That is, the process proceeds to step S10 in the control flow shown in FIG.

  As shown in FIG. 8, the predetermined region X of the vehicle outside image is set to a certain range on the vehicle side of the vehicle outside image when the vehicle outside is the vehicle side and the vehicle outside image is the non-vehicle side. Area. In addition, the horizontal width of the predetermined region X substantially matches the horizontal width of the outside image, and the vertical width of the predetermined region X is a width within 1/5 to 1/3 of the vertical width of the outside image.

  That is, the predetermined region X is a region in the vicinity of the hood of the vehicle 1 shown in the vehicle outside image, which is substantially the same as the width of the vehicle outside image and the vertical width of the vehicle outside image is reduced by 1/3 or more.

  In step S15, the control unit 4 derives the size of the lane marking included in the predetermined region X in the outside image. Next, the process proceeds to step S16.

  In step S <b> 16, the control unit 4 compares the derived lane line size with the reference size of the lane line stored in advance in the nonvolatile storage unit 4 b included in the control unit 4. Next, the process proceeds to step S17.

  In step S <b> 17, the control unit 4 determines whether or not the size of the lane marking included in the predetermined area X matches the reference size of the lane marking, or the size and the section of the lane marking included in the predetermined area X. It is determined whether or not the reference size of the line substantially matches. When it is not determined that they match, and when it is not determined that they substantially match, the process proceeds to step S18 (NO in step S17). If it is determined that they match, or if it is determined that they substantially match, the process proceeds to return (YES in step S17). That is, the process proceeds to step S10 in the control flow shown in FIG.

  Note that, as described above, when the control unit 4 determines that the size of the lane markings included in the predetermined region X and the reference size of the lane markings are substantially the same, the difference when comparing them is a predetermined value. The case where it falls within the range (for example, within 3 cm).

  Here, the size of the lane line included in the predetermined area X in the vehicle outside image to be compared by the control unit 4 is the size of the lane line included in the predetermined area X, and the imaging unit 6 is appropriate for the vehicle 1. The actual lane line size (for example, the actual lane line length, etc.) derived based on the mounting position on the windshield 17 of the vehicle 1 and the mounting angle when mounted on the vehicle.

  Further, the reference size of the lane line stored in advance in the non-volatile storage unit 4b is, for example, a broken line among the lane lines that divide the travel lane shown on the road surface in the travel path as shown in FIG. The length of the line (5 m for a general road), the length of the interval (5 m for a general road), or the combined length of the line and the distance (10 m for a general road).

  In addition, the reference size of the lane line stored in advance in the nonvolatile storage unit 4b is, for example, among the lane lines that divide the travel lane indicated on the road surface in the travel path, as illustrated in FIG. The length of the line on the broken line (8 m for the expressway), the length of the interval (12 m for the expressway), or the combined length of the line and the interval (20 m for the expressway) .

  In addition, the reference size of the lane line stored in advance in the nonvolatile storage unit 4b is, for example, among the lane lines that divide the travel lane shown on the road surface in the travel path, as shown in FIG. For example, the width (30 cm) in the solid line.

  In addition, as the reference size of the lane markings stored in advance in the nonvolatile storage unit 4b, various lane marking sizes that divide a traveling lane defined by law, not shown, can be considered.

  Moreover, the vehicle-mounted apparatus 2 extracts the reference size of the lane marking according to the travel path determined based on the navigation information received from the car navigation apparatus 10 from the nonvolatile storage unit 4b. Alternatively, the in-vehicle device 2 extracts the reference size of the lane marking according to the recognition result such as the traffic regulation sign included in the outside image from the nonvolatile storage unit 4b.

In step S <b> 18, the control unit 4 determines whether or not the left and right center point C of the front end portion of the vehicle bonnet B shown in the vehicle outside image matches the left and right center of the vehicle outside image. If it is determined that the left and right center point C matches the left and right center of the outside image, the process proceeds to step S19 ( YES in step S18). If it is not determined that the left-right center point C matches the left-right center of the outside image, the process proceeds to return ( NO in step S18). That is, FIG.
The process proceeds to step S10 in the control flow shown in FIG. Thereby, the process of step S19 is avoided and the comparison result is invalidated.

  In step S <b> 19, the control unit 4 counts one counter set in the volatile storage unit 3. This counter is used to count the number of times that the comparison results of the lane marking sizes do not match. The counter is reset in the initial process and the end process in the control executed by the control unit 4 and is reset in the reset process of the in-vehicle device 2. Next, the process proceeds to return. That is, the process proceeds to step S10 in the control flow shown in FIG.

  In step S <b> 10, the control unit 4 determines whether or not the counter set in the volatile storage unit 3 is a predetermined value (for example, 5) or more. If it is determined that the counter is equal to or greater than the predetermined value, it is determined that the mounting position of the imaging unit 6 in the vehicle has shifted, and the process proceeds to step S11. If it is not determined that the counter is greater than or equal to the predetermined value, the process proceeds to return.

  In step S11, the control unit 4 controls the sound output unit 5 to notify the user accordingly. That is, the user is notified that the mounting position of the imaging unit 6 mounted on the vehicle is shifted.

  That is, the in-vehicle device 2 derives the size of the lane marking on the road surface based on the image of the road surface on the vehicle side that is captured in the image outside the vehicle including the road surface around the vehicle acquired by the imaging unit 6, and the derived lane marking And the reference size of the lane markings on the road surface stored in advance in the non-volatile storage unit 4b, the displacement of the mounting position of the imaging unit 6 mounted on the vehicle was detected, and the displacement was detected In that case, the user is notified of this.

  For this reason, the user can know that the mounting position of the imaging unit 6 is misaligned, and can have the mounting position of the imaging unit 6 made appropriate by the sales company that purchased the vehicle 1 or the in-vehicle device system. . That is, the user can prevent erroneous vehicle departure determination in the in-vehicle device 2 due to an inappropriate mounting position of the imaging unit 6.

  The in-vehicle device 2 counts that the comparison results are inconsistent using a counter, and performs deviation detection when the inconsistency is counted more than a predetermined value. In order to notify the user of the fact, it is possible to notify the user of the highly accurate displacement detection without notifying the user of erroneous displacement detection.

  The in-vehicle device 2 disables the function of comparing the size of the lane markings so that the deviation is not detected when the traveling path on which the vehicle 1 travels is a slope. For this reason, it is possible to notify the user of accurate displacement detection without notifying the user of erroneous displacement detection.

  In addition, the in-vehicle device 2 can prevent the detection of deviation based on an erroneous comparison result in order to invalidate the comparison result when the left-right center point C and the left-right center point of the image outside the vehicle do not match.

  In addition, since the in-vehicle device 2 detects the deviation based on the lane line reflected in the predetermined area X of the vehicle outside image, it can prevent erroneous displacement detection based on the lane line at a position farther than the vehicle 1 in the vehicle outside image. it can.

  Here, the reason why the misalignment is detected is based on the following reason. When a specific traveling road, such as a sharp curve, an uphill road or a downhill road with a severe inclination angle, is shown at a position farther than the vehicle 1 in the outside image, the shape of the road surface reflected in the outside image is not a linear shape. In addition, the way the lane markings shown on the road surface are also deformed according to the road surface.

  If the in-vehicle device 2 detects the deviation based on the lane markings that are deformed according to the road surface reflected in the image outside the vehicle, even if the imaging unit 6 is mounted at an appropriate position, it is not deformed. Since the shift detection is performed based on the reference size of the lane markings, there is a possibility that erroneous shift detection may be performed.

<Modification>
The representative embodiment of the present invention has been described above. However, the present invention is not limited to the above-described representative embodiment, and various modifications can be made. Below, the modification which is other embodiment is demonstrated. Of course, you may combine the form demonstrated below suitably.

<Modification 1>
The mounting method of the imaging unit 6 in the representative embodiment is as follows: “Mounting of the imaging unit 6 to the vehicle 1 is inside the windshield of the vehicle 1 and the imaging direction is in the vicinity of the rearview mirror in the windshield. Although described as “mounted toward the outside of the vehicle”, the vehicle 1 may be mounted at the left and right center at the upper end inside the rear glass.

  In this case, the imaging unit 6 is mounted so that the imaging direction is directed to the outside of the vehicle, and the mounting angle of the lens included in the imaging unit 6 is determined when the vehicle 1 performs normal travel in the vehicle outside image captured by the imaging unit 6. The end of the trunk of the vehicle 1 is set to be reflected, and the lane marking that divides the traveling lane on the road surface in the direction opposite to the traveling direction of the vehicle 1 is set to be reflected in a predetermined area.

  The predetermined area shown in the vehicle outside image is an area including a position at least 2 m away from the center of the left and right of the front end portion of the trunk of the vehicle 1 in the left-right direction. As a result, the left lane line LL and the right lane line RL indicating the boundary between the road lane on which the vehicle 1 travels and the adjacent lane are displayed in the outside image.

  In order to mount the imaging unit 6 in this way, the control unit 4 includes a left side dividing line LL that divides the center point of the end of the trunk of the vehicle 1 included in the vehicle exterior image and the road lane included in the vehicle exterior image. And the relative distance D between the right lane line RL and the right lane line RL when it is virtually extended to the right rear wheel of the vehicle 1 can be calculated.

<Modification 2>
In the above representative embodiment, “the control unit 4 controls the sound output unit 5 to output a sound. Thereby, the control unit 4 may cause the vehicle 1 to deviate from the lane marking. Alternatively, the controller 4 of the in-vehicle device 2 transmits a signal indicating an alarm based on the vehicle departure determination to the navigation device 10 via the communication cable, and the signal is notified. The vehicle-mounted device 2 that receives the signal may output a sound or video indicating an alarm.

<Modification 3>
In the above representative embodiment, it has been described that “the acceleration sensor 7 is installed in the engine room of the vehicle 1”. However, the in-vehicle device 2 may be included in the acceleration sensor 7.

<Modification 4>
In the control flow shown in FIG. 12 in the above-described representative embodiment, “if the traveling road is determined to be a slope in step S13, the subsequent displacement detection processing of the mounting position of the imaging unit 6 is not executed”. However, based on the navigation information received from the navigation device 10 or the angle signal received from the gyro sensor 11, the control unit 4 included in the in-vehicle device 2 determines that the traveling path on which the vehicle 1 travels is a slope ( This determination is referred to as a first determination), and after a predetermined time has elapsed from the first determination, when it is determined that the travel path on which the vehicle 1 travels is a slope (this determination is referred to as a second determination), the first determination is made. If the inclination angle at the time of the second determination is the same as the inclination angle at the time of the second determination, the subsequent displacement detection processing of the mounting position of the imaging unit 6 is executed, and the inclination angle at the time of the first determination and the second Judgment If the inclination angle at the are not the same, it may not perform the deviation detection process of mounting position of the imaging unit 6 and later.

  That is, when the vehicle 1 continues to travel on a slope with the same inclination angle, the in-vehicle device 2 performs deviation detection, and the vehicle 1 travels on a slope with the inclination angle changing for a predetermined time. Does not detect misalignment.

  Thereby, when the vehicle 1 is traveling on a slope whose inclination angle changes in a predetermined time, that is, the lane markings included in the road surface reflected in the image outside the vehicle acquired by the imaging unit 6 included in the vehicle 1 have various shapes. Even when the imaging unit 6 is mounted at an appropriate position, the shift detection is performed based on the reference size of the lane marking that is not deformed, so that erroneous shift detection is prevented. be able to.

<Modification 5>
In the representative embodiment, “in step S16, the control unit 4 compares the derived lane line size with the reference size of the lane line stored in advance in the nonvolatile storage unit 4b included in the control unit 4. However, when the vehicle 1 travels on various travel routes, the reference size of the lane marking is learned and stored in the nonvolatile storage unit 4b provided in the control unit 4, for example, a flash memory. May be.

  When the control unit 4 learns the reference size of the lane marking, since it is a condition that the imaging unit 6 is mounted at an appropriate position, it is determined that no deviation detection has been performed before a predetermined time of learning. To run after.

  In addition, the control unit 4 stores the reference size learned together with the GPS signal received from the navigation device 10 in the flash memory. Then, when detecting the deviation, the control unit 4 reads the learning reference size related to the GPS signal corresponding to the received GPS signal from the flash memory, and performs the above-described deviation detection based on the read learning reference size. You may do it.

  Thereby, the in-vehicle device 2 can store the reference sizes of various lane markings in the flash memory, and can perform flexible misalignment detection that can handle various lane markings.

<Modification 6>
Each process in the control flow for explaining the control of each embodiment is shown as one series for convenience. However, the control unit 4 processes the subdivided processes in parallel by the multitask control function. May be.

2 On-vehicle device 4 Control unit 5 Sound output unit 6 Imaging unit 7 Acceleration sensor 9 Vehicle speed sensor 11 Gyro sensor

Claims (8)

A detection device that detects a shift in a mounting position of an imaging unit mounted on a vehicle,
Storage means for storing a reference size of a lane marking on the road surface;
Input means for inputting an image outside the vehicle including a road surface around the vehicle acquired by the imaging unit;
Derivation means for deriving the size of the lane markings on the road surface based on the image of the road surface on the vehicle side reflected in the image outside the vehicle;
Comparison means for comparing the size derived by the deriving means with the reference size stored in the storage means;
Detecting means for detecting the deviation based on a comparison result of the comparing means;
An informing means for informing the user when the deviation is detected;
A detection device comprising: The detection device according to claim 1,
A detection apparatus, further comprising: a first invalidation unit that invalidates the comparison result when a left-right center of the input image outside the vehicle does not match a left-right center of the vehicle reflected in the image outside the vehicle. In the detection device according to claim 1 or 2,
The detection device is characterized in that the shift is detected while the vehicle is running. In the detection apparatus in any one of Claims 1-3,
The partition line is a broken line, and the reference size includes any one of a length of a line, a length of an interval, and a length of the line and the interval in the broken line. In the detection apparatus in any one of Claims 1-3,
The detection device, wherein the partition line is a solid line, and the reference size is a width of the solid line. In the detection apparatus in any one of Claims 1-5,
Detecting means for detecting an angle of the vehicle with respect to a horizontal direction;
Further comprising a gradient determining means for determining whether or not the traveling path of the vehicle is a gradient based on the angle;
And a second invalidating unit that invalidates the comparing unit when the gradient is determined by the gradient determination. In the detection apparatus in any one of Claims 1-6,
A departure determination means for determining whether or not the vehicle departs from the lane marking, further comprising:
The said alerting | reporting means alert | reports that to a user, when determining the said vehicle deviation. A detection method for detecting a shift in a mounting position of an imaging unit mounted on a vehicle,
(A) inputting an image outside the vehicle including a road surface around the vehicle acquired by the imaging unit;
(B) deriving the size of the lane markings on the road surface based on the image of the road surface on the vehicle side reflected in the outside image;
(C) comparing the size derived in step (b) with the reference size of the lane marking on the road surface stored in advance in the storage means;
(D) a step of detecting the deviation based on a comparison result of the step (c);
(E) a step of notifying the user when the deviation is detected;
A detection method characterized by comprising:

Images