JP2009025198A - Apparatus and method for detecting road surface condition - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus for detecting road surface condition capable of reducing the effect of environmental light and detecting road surface condition with improved accuracy. <P>SOLUTION: The apparatus 1 for detecting road surface condition comprises a headlight 10 for lighting road surface, a camera 12 for imaging the images of S-polarization and P-polarization of the road surface, and a controlling section 18 for determining the road surface condition based on the images of S-polarization and P-polarization. The controlling section 18 calculates S-polarization component and P-polarization component of the multiple pixels in the multiple frames based on the multiple frames of the images of S-polarization and P-polarization that are taken over the predetermined period while driving a vehicle, calculates the difference between S-polarization component and P-polarization component for the multiple pixels, calculates the median among the resulted multiple differences, and determines the road surface condition based on the median. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a road surface state detection device that detects a road surface state.

  2. Description of the Related Art Conventionally, as a technique for detecting a road surface state, a technique for embedding a sensor on a road surface and a technique for analyzing a road surface image captured by a camera are known. The former is not practical because it requires a large number of sensors and increases costs. The latter method utilizes the property that the P-polarized light component has a higher transmittance from water to air than the S-polarized light component, and analyzes the P-polarized light component and the S-polarized light component reflected by the road surface to analyze the road surface condition. Techniques for detecting are known.

In the method of detecting the road surface state by image processing, ambient light such as street lights and headlights of oncoming vehicles, reflected light from road paint, etc. affect the detection of the road surface state. Patent Document 1 discloses a road surface state detection apparatus that improves calculation accuracy by removing noise generated due to an external light state that changes during traveling and an installation angle of an in-vehicle camera. This road surface state detection device compares the polarization ratio intensity acquired last time with the polarization ratio intensity acquired this time, and when the difference is larger than a predetermined threshold, determines that the polarization ratio intensity acquired this time is noise, Replace with the previous polarization specific intensity.
JP 2004-299443 A

  However, in the road surface state detection device described in Patent Document 1, it is possible to remove instantaneous noise, but when ambient light such as street lights and headlights of oncoming vehicles is continuously incident. The incident light cannot be processed as noise. Therefore, the road surface state detection device described in Patent Document 1 cannot remove the influence of ambient light such as street lights and headlights of oncoming vehicles.

  SUMMARY OF THE INVENTION In view of the above background, an object of the present invention is to provide a road surface state detection device capable of accurately detecting a road surface state.

  A road surface state detection device according to the present invention includes a polarization component acquisition unit that acquires an S-polarized component and a P-polarized component of light reflected from a road surface, and a road surface state that determines a road surface state based on the S-polarized component and the P-polarized component. A determination unit, and the road surface state determination unit obtains comparison data of each of the plurality of S polarization components and the P polarization component acquired by the polarization component acquisition unit, and obtains representative values of the obtained plurality of comparison data. The road surface state is determined based on the representative value.

  As described above, the comparison data of the plurality of S polarization components and the P polarization components of the reflected light from the road surface is obtained, and the representative value of the obtained comparison data is obtained, thereby reducing the influence of the comparison data including the ambient light. The value can be determined. By determining the road surface state using this representative value, the influence of ambient light can be reduced and the road surface state can be detected with high accuracy. The “comparison data” may be a difference between the S-polarized component and the P-polarized component or a ratio.

  The road surface state detection apparatus of the present invention includes a road surface state determination permission unit that determines whether or not to permit determination of a road surface state using data of the S polarization component and the P polarization component acquired by the polarization component acquisition unit. When it is determined that the determination of the road surface state is permitted, the road surface state determination unit performs the determination of the road surface state, and when it is determined that the determination of the road surface state is not permitted, the polarization component acquisition unit performs S A polarization component and a P polarization component may be acquired.

  As described above, the road surface state determination permission unit determines whether or not to permit the determination of the road surface state. When the road surface state determination is not permitted, the S surface polarization component and the P polarization component are further acquired and determined. Since the data used as material is increased, the accuracy of road surface condition determination can be improved.

  In the road surface state detection device of the present invention, the road surface state determination permission unit may determine whether or not to permit the determination of the road surface state based on the acquired number of data of the S polarization component and the P polarization component.

  By determining whether to permit the determination of the road surface state based on the number of data of the S-polarized component and the P-polarized component acquired in this way, the number of data is sufficient to reduce the influence of environmental light. Based on this, it is possible to appropriately determine the road surface condition.

  In the road surface state detection device, the road surface state determination permission unit obtains the comparison data of the newly acquired S polarization component and P polarization component, and the newly obtained comparison data and a plurality of comparison data obtained up to the previous time It is also possible to repeatedly perform the process of obtaining the representative value from the comparison data combined with the above and permit the determination of the road surface state when the variation of the representative value repeatedly calculated by the process converges.

  As the comparison data for obtaining the representative value is increased, the data on the road surface with little ambient light gradually increases, even when data on the road surface with a lot of ambient light is temporarily acquired. The representative value of data converges to a representative value that is less affected by ambient light. Therefore, the representative value in which the influence of the ambient light is reduced can be appropriately obtained by the configuration that permits the determination of the road surface state by the convergence of the fluctuation of the representative value.

  In the road surface state detection device, the road surface state determination unit obtains and stores the median value or average value of the comparison data for each predetermined unit determined in advance, and uses the stored median value or average value to store the median value or average value. A representative value may be obtained.

  As described above, the average value or the median value is obtained and stored for each predetermined unit, and the representative value is obtained using the stored average value or the median value. It becomes possible to delete the data of the polarization component and the P polarization component. This makes it possible to obtain representative values of a large number of comparison data using a memory with a small capacity.

  In the road surface state detection apparatus, the polarization component acquisition unit includes a polarization image capturing unit that captures an S-polarized image and a P-polarized image of the road surface, and is positioned from among the captured S-polarized image and the P-polarized image itself. The S polarization component and the P polarization component of a plurality of pixels selected without bias may be obtained.

  With this configuration, a plurality of S-polarized components and P-polarized components can be efficiently acquired from an S-polarized image and a P-polarized image obtained by one imaging. Further, by selecting a plurality of pixels without positional deviation, the possibility of erroneous determination due to partial factors can be reduced. For example, even when there is an obstacle such as a motorcycle in the imaging range and reflected light from the road surface cannot be obtained from that part, the road surface without an obstacle can be obtained by acquiring the S-polarized component and the P-polarized component without positional deviation. You can get data at.

  In the road surface state detection device, the polarization component acquisition unit is configured to determine whether the polarization component acquisition unit includes all the pixels of the plurality of frames based on the S-polarization images and the P-polarization images of the plurality of frames captured over a predetermined time while the vehicle is traveling. The S-polarized component and the P-polarized component of a plurality of pixels selected without positional deviation may be obtained.

  The inventors have found through experiments that it is important to acquire road surface images between streetlights in order to reduce the influence of streetlights that are periodic ambient light. Furthermore, in general, the distance between street lamps is 20 to 30 m, but the distance greatly affected by street lamps is found to be about 8 m. Therefore, the distance of the road surface where the influence of the streetlight is small is 12 m or more. By acquiring many road surface images in this section, it is possible to reduce the influence of street lamps and increase the discrimination accuracy. According to the configuration of the present invention, the reflected light for detecting the road surface condition is acquired from a wide range by imaging the road surface over a predetermined time while the vehicle is traveling. To obtain comparison data between the S-polarization component and P-polarization component of a plurality of pixels selected without positional deviation from all the pixels of a plurality of captured frames, and to obtain a representative value from all the obtained comparison data Thus, a representative value in which the influence of the comparison data including ambient light is reduced can be obtained. By determining the road surface state using this representative value, the road surface state can be detected with high accuracy.

  In the road surface state detection device, the road surface state determination unit determines that the road surface is in a wet state when the P-polarized light component is larger than the S-polarized light component based on the representative value. May be determined to be in a dry state.

  Since the road surface is covered with water in a wet state, the reflected light on the road surface is transmitted from the water into the air. Since the P-polarized light component has a higher transmittance from water to air than the S-polarized light component, it can be determined that the wet state when the P-polarized light component is large.

  In the road surface state detection device, the road surface state determination unit may exclude the frame from the road surface determination when the frame captured by the polarization image capturing unit includes an image of road paint.

  Since the road paint has a high reflectance, the amount of reflected ambient light increases. According to the configuration of the present invention, the road surface state can be appropriately detected by excluding the frame including the road paint image from the road surface determination.

  The road surface state detection apparatus includes a reference image storage unit that stores a frame that does not include an on-road paint image as a reference image, and the road surface state determination unit includes all the pixels of the frame captured by the polarization image capturing unit. Histogram of comparison data of S-polarization component and P-polarization component of pixel selected without positional deviation from among them, and S-polarization component and P-polarization component of pixel selected without positional deviation from all pixels of the reference image It may be determined whether or not road paint is included in the frame by comparing with a histogram of the comparison data.

  Thus, the presence or absence of road paint can be determined by comparing the histograms of the comparison data of the reference image and the captured image. The reference image storage unit may store an image that does not include road paint that has been captured in advance. In addition, the image stored in the reference image storage unit may be updated with an image that is determined not to include road paint by determining whether road paint is included.

  In the road surface state detection device, the road surface state determination unit acquires a road environment in which the vehicle is traveling based on map information and information on a current position of the vehicle, and the vehicle is a place where there is a lot of road paint based on the road environment. The polarization component acquisition unit may be controlled so as to stop acquisition of the S-polarized component and the P-polarized component while traveling in a place with a lot of road paint.

  With this configuration, it is possible to exclude the S-polarized component and the P-polarized component obtained from a place where the road paint is likely to be included from the road surface determination. The “road environment” is information on roads and facilities around a running vehicle. The road environment may be determined by the road surface state detection device, or may be acquired from a navigation device that cooperates with the road surface state detection device.

  In the road surface state detection device, the road surface state determination unit may obtain a median value of a plurality of comparison data as the representative value.

  The median is a value located at the center when a finite number of data is sorted. Even if comparison data with a large influence of ambient light is included, if there are more comparison data with a small influence of ambient light than comparison data with a large influence of ambient light, comparative data with a small influence of ambient light is in the center. To position. Therefore, by using the median as the representative value, it is possible to appropriately determine the road surface state using the comparison data that is less influenced by the ambient light.

  The road surface state detection method of the present invention includes a step of acquiring a plurality of S-polarized components and a P-polarized component included in reflected light from the road surface, and a step of obtaining comparison data of the plurality of S-polarized components and the P-polarized components. And obtaining a representative value of the obtained plurality of comparison data, and determining a road surface state based on the representative value.

  With this configuration, as in the road surface condition detection device of the present invention, the influence of environmental light is reduced by obtaining a representative value of comparison data of a plurality of S-polarized components and P-polarized components of reflected light from the road surface, The road surface condition can be detected with high accuracy. In addition, various configurations of the road surface state detection device of the present invention can be applied to the road surface state detection method of the present invention.

  A program of the present invention is a program for detecting a road surface state by a computer including a polarization component acquisition unit that acquires an S-polarized component and a P-polarized component of reflected light from the road surface. Obtaining an S-polarized component and a P-polarized component; obtaining comparison data of each of the plurality of S-polarized components and P-polarized component; obtaining a representative value of the obtained plurality of comparison data; And a step of determining a road surface state on the basis thereof.

  With this configuration, as in the road surface condition detection device of the present invention, the influence of environmental light is reduced by obtaining a representative value of comparison data of a plurality of S-polarized components and P-polarized components of reflected light from the road surface, The road surface condition can be detected with high accuracy. Various configurations of the road surface condition detection device of the present invention can be applied to the program of the present invention.

  According to the present invention, by obtaining a representative value of comparison data of a plurality of S-polarized components and P-polarized components of reflected light from a road surface, the influence of ambient light can be reduced and the road surface state can be detected with high accuracy.

Hereinafter, a road surface state detection device according to an embodiment of the present invention will be described with reference to the drawings.
(First embodiment)
FIG. 1 is a diagram illustrating a configuration of a road surface state detection device 1 according to the first embodiment. The road surface state detection apparatus 1 uses a headlight 10 that irradiates a road surface, a camera 12 that captures a P-polarized image and an S-polarized image of the road surface, and a road surface state using the P-polarized image and the S-polarized image captured by the camera 12. And a control unit 18 that performs a process of detecting.

  It is preferable that the camera 12 captures an area where the reflected light from the headlight 10 is strong. In the present specification, “ambient light” is light based on a light source other than the headlight of the host vehicle, and is, for example, light from a streetlight, a headlight of an oncoming vehicle, or the like. “Reflected light” refers to all light from the road surface incident on the camera 12 and does not refer to only light that has been totally reflected (specularly reflected). The light of the headlight 10 and the ambient light are, for example, mirror-reflected by water accumulated on the road surface and scattered by unevenness of the road surface, but both “reflected light” are included. The camera 12 includes a P-polarized image acquisition unit 14 and an S-polarized image acquisition unit 16. The light from the road surface incident on the camera 12 is divided into two by a half mirror or the like, and is incident on the P-polarized image acquisition unit 14 and the S-polarized image acquisition unit 16. Therefore, the P-polarized image acquisition unit 14 and the S-polarized image acquisition unit 16 acquire a P-polarized image and an S-polarized image for the same image. The P-polarized image acquisition unit 14 includes a P-polarized filter that transmits only P-polarized light and an imaging device such as a CCD. Similarly, the S-polarized image acquisition unit 16 includes an S-polarized filter that transmits only S-polarized light and an imaging device such as a CCD.

  FIG. 2 is a diagram for explaining the P-polarized component and the S-polarized component. As shown in FIG. 2, the P-polarized light component is a wave that vibrates in a plane perpendicular to the reflecting surface (a surface including a normal line standing on the reflecting surface and the traveling direction of light). The S-polarized component is a wave that oscillates in a plane perpendicular to the plane of vibration of P-polarized light.

  The configuration of the road surface condition detection device 1 will be described with reference to FIG. 1 again. The control unit 18 includes a camera image acquisition control unit 20, a noise removal unit 22, a polarization difference data calculation unit 24, and a road surface state determination unit 30. Further, the road surface state determination unit 30 includes a road surface state determination permission unit 25, and the road surface state determination permission unit 25 includes a median value calculation unit 26 and a convergence determination unit 28. In FIG. 1, the functions of the control unit 18 are shown as functional blocks. However, the functions of the control unit 18 can be realized, for example, by executing a program stored in a memory by the CPU.

  FIG. 3 is a diagram illustrating an example of a road surface state detection program 50 for realizing the function of the control unit 18 of the road surface state detection device 1. The road surface state detection program 50 includes a camera image acquisition control subprogram 52, a noise removal subprogram 54, a polarization difference data calculation subprogram 56, and a road surface state determination subprogram 62. The road surface state determination subprogram 62 includes a road surface state determination permission subprogram 57, and the road surface state determination permission subprogram 57 further includes a median value calculation subprogram 58 and a convergence determination subprogram 60. . By executing the camera image acquisition control subprogram 52 by the CPU, the function of the camera image acquisition control unit 20 of the control unit 18 is realized. Thereby, the road surface condition detection apparatus 1 performs acquisition control of the camera image captured by the camera 12. The function of the noise removal unit 22 is realized by executing the noise removal subprogram 54 by the CPU. Similarly, the polarization difference data calculation subprogram 56, the road surface state determination permission subprogram 57, the median value calculation subprogram 58, the convergence determination subprogram 60, and the road surface state determination subprogram 62 are each executed by the CPU, whereby the polarization difference data is calculated. The functions of the data calculation unit 24, the road surface state determination permission unit 25, the median value calculation unit 26, the convergence determination unit 28, and the road surface state determination unit 30 are realized.

  Next, the configuration of the control unit 18 will be described. The camera image acquisition control unit 20 controls timing for acquiring an image captured by the camera 12. The camera 12 always captures the road surface, and the camera image acquisition control unit 20 acquires a camera image from the constantly captured camera images. The camera image acquisition control unit 20 acquires a camera image for a predetermined time while the vehicle is traveling, and stops acquiring the camera image while the vehicle is stopped. Even if a camera image is acquired while the vehicle is stopped, only the same part of the road surface can be obtained, so if the road surface happens to be heavily affected by ambient light, the representative value will be affected by the ambient light. Because. The camera image acquisition control unit 20 may change the timing of camera image acquisition according to the traveling speed of the vehicle. For example, when the vehicle is traveling at a low speed, the time for acquiring the camera image may be longer than when the vehicle is traveling at a high speed. Thereby, the size of the road surface to be imaged can be kept substantially constant in the case of high speed traveling and the case of low speed traveling.

  The noise removing unit 22 performs processing for removing noise that is not caused by ambient light from the P-polarized image and the S-polarized image captured by the camera 12. For example, since the intensity of incident light may not be expressed as a luminance value depending on the width of the light receiving sensitivity of the camera, the noise removing unit 22 determines that each pixel value is 0 to 255. A pixel having a value of 0 and a pixel value of 255 is determined to be noise, and the pixel data is removed so as not to be used in the following calculation. When the actual brightness is brighter than the brightness corresponding to the pixel value 255, the pixel value is 255. Therefore, the pixel value 255 may not represent a correct value. Pixel value 255 is removed as noise. The pixel value 0 is also processed as noise for the same reason. Note that an appropriate range may be set by the camera 12 as to which range of pixel values is treated as noise.

  The polarization difference data calculation unit 24 obtains the P polarization intensity and the S polarization intensity of each pixel in the image based on the P polarization image and the S polarization image captured by the camera 12, and the obtained P polarization intensity and S polarization intensity. The difference is calculated. Since the camera 12 obtains a P-polarized image and an S-polarized image by dividing the light from the subject by a half mirror or the like, the camera 12 obtains the P-polarized light intensity and the S-polarized light intensity of the pixel showing the same point on the road surface. Can do. Note that the P-polarized light intensity and the S-polarized light intensity of each pixel are the P-polarized light component and the S-polarized light component at a predetermined point. Therefore, the configuration of the camera 12 and the polarization difference data calculation unit 24 for obtaining the P-polarization intensity and the S-polarization intensity of each pixel corresponds to the “polarization component acquisition unit” of the present invention.

  The polarization difference data calculation unit 24 obtains a pixel value difference by subtracting the pixel value of the P-polarized image from the pixel value of the S-polarized image for effective pixels that are not noise. Hereinafter, the difference data is referred to as “polarization difference data”. The polarization difference data calculation unit 24 stores the calculated polarization difference data in the polarization difference data storage unit 32. The polarization difference data storage unit 32 stores the polarization difference data obtained by the polarization difference data calculation unit 24 until the detection of the road surface state is completed.

The polarization difference data calculation unit 24 obtains polarization difference data not only for pixels within one frame but also for pixels within a plurality of frames.
4A to 4C are diagrams illustrating examples of histograms of polarization difference data of all effective pixels included in an image. 4A is a histogram of polarization difference data included in an image of a plurality of frames, FIG. 4B is a histogram of polarization difference data included in an image of a frame having a small influence of ambient light, and FIG. 4C is an environment. It is a histogram of polarization difference data included in an image of a frame having a large influence of light. As shown in FIG. 4B, in the polarization difference data histogram of the frame where the influence of the ambient light is small, the frequency data is distributed more on the minus side. As shown in FIG. 4C, in the histogram of the polarization difference data of the frame where the influence of the ambient light is large, the frequency data is largely distributed on the plus side. Thus, the histogram of the polarization difference data for each frame differs depending on the magnitude of the influence of the ambient light. In the histogram of polarization difference data of a plurality of frames, as shown in FIG. 4A, the frequency data is distributed on both the minus side and the plus side, but tends to increase on the minus side. Since a plurality of frames imaged during traveling have different portions of the road surface imaged according to the traveling of the vehicle, even if the ambient light is included in a part of the imaging range, a portion where the influence of the environmental light is small This is because the polarization difference data of the reflected light obtained from a wide range including can be obtained.

  The road surface state determination unit 30 includes a road surface state determination permission unit 25, determines whether to permit the determination of the road surface state, and permits the road surface state using the polarization difference data when permitting the determination of the road surface state. judge. The road surface state determination permission unit 25 includes a median value calculation unit 26 and a convergence determination unit 28. The median value calculation unit 26 obtains the median value of the plurality of polarization difference data, and the convergence determination unit 28 determines whether or not the median value has converged. Permits the determination of the road surface condition. Then, the road surface state determination unit 30 determines the road surface state based on the converged median value. Hereinafter, the median value calculation unit 26 and the convergence determination unit 28 will be described.

  The median value calculation unit 26 obtains the median value of the polarization difference data of all effective pixels included in the captured image. The median is a value located at the center when the polarization difference data is sorted according to the size. The median value calculation unit 26 adds the polarization difference data obtained from the new frame every time a new frame is acquired from the camera 12, and calculates the median value of the polarization difference data. Specifically, since the polarization difference data of all effective pixels is stored in the polarization difference data storage unit 32, the median value calculation unit 26 uses the polarization difference data stored in the polarization difference data storage unit 32 to perform central processing. Calculate the value.

  The convergence determination unit 28 determines whether or not the variation in the median value of the polarization difference data calculated every time a new frame is acquired has converged.

  FIG. 5 is a diagram showing fluctuations in the median value calculated by the median value calculation unit 26. In the example shown in FIG. 5, when the median value of the polarization difference data is obtained using the captured image of 10 frames, it is −35, −30 for 20 frames, and −25 for 40 frames. In this example, as the number of frames increases, the median value of the polarization difference data gradually increases, and fluctuates in the range of −25 to −23 for 40 to 80 frames. When the number of frames used for calculation increases in this way, the influence of ambient light can be reduced, and the variation in the median value of polarization difference data converges. The median is a value that varies depending on the degree of road surface wetting, the brightness of the surroundings, or the light receiving sensitivity of the camera, and the median of the polarization difference data shown in FIG. 5 is merely an example.

  The convergence determination unit 28 determines that the fluctuation has converged, for example, when the state where the median value of the polarization difference data fluctuates within a predetermined threshold value continues for a predetermined number of frames. When the convergence determination unit 28 determines that the fluctuation has converged, the road surface state permission unit 25 permits the road surface state determination and notifies the camera image acquisition control unit 20 of the determination result that permits the road surface state determination. . Upon receiving this notification, the camera image acquisition control unit 20 controls the camera 12 to stop the acquisition of the camera image.

  The road surface state determination unit 30 determines whether the road surface is in a dry state or a wet state based on the median value of the polarization difference data obtained by the median value calculation unit 26 and the convergence determination unit 28. When the road surface is dry, the light scattered on the road surface enters the camera 12, so that the S-polarized light intensity and the P-polarized light intensity are substantially equal. When the road surface is wet, light scattered on the road surface passes through moisture on the road surface and enters the camera 12. Since the transmittance of light from water into the air is greater for P-polarized light than for S-polarized light, the light received by the camera 12 has greater P-polarized light intensity than S-polarized light intensity.

  FIG. 6 is a diagram illustrating the relationship between the road surface state and the polarization difference data (SP). As shown in FIG. 6, when the road surface is in a dry state, the polarization difference data is almost zero. Since there is a slight error, the polarization difference data at the time of drying varies in the vicinity of 0 in FIG. When the road surface is wet, the polarization difference data is a negative value. Although the value indicating the intensity of the reflected light changes depending on the light reception sensitivity of the camera, the tendency that the polarization difference data takes a value near 0 in the dry state and the polarization difference data takes a negative value in the wet state. The same is true regardless of the light receiving sensitivity of the camera. Therefore, the road surface state determination unit 30 determines the dry state when the polarization difference data is −10 or more, and determines the wet state when the polarization difference data is smaller than −10 in consideration of the error. Note that the threshold for determining the road surface condition can be appropriately determined according to the characteristics of the camera 12 and the like.

  FIG. 7 is a flowchart illustrating the operation of the road surface condition detection device 1 according to the first embodiment. The road surface state detection device 1 captures an S-polarized image and a P-polarized image of the road surface with the camera 12 while the vehicle is traveling (S10). When capturing the S-polarized image and the P-polarized image of one frame, the control unit 18 removes noise that is not caused by the ambient light from the captured S-polarized image and P-polarized image (S12). In this embodiment, an example in which image processing is performed every time one frame is imaged will be described. However, the processing may be shifted to step S12 and subsequent steps when a plurality of frames are imaged. Subsequently, the control unit 18 of the road surface state detection apparatus 1 obtains the polarization difference data of valid pixels in the frame (S14), and stores the obtained polarization difference data in the polarization difference data storage unit 32 (S16).

  Next, the control unit 18 reads all the polarization difference data stored in the polarization difference data storage unit 32, obtains the median value of the polarization difference data (S18), and determines whether or not the variation of the median value has converged. (S20). That is, the control unit 18 determines whether or not the determination of the road surface state is permitted. When it is determined that the variation in the median value has not converged (NO in S20), that is, when it is determined that the determination of the road surface state is not permitted, the road surface state detection device 1 displays the S-polarized image and the P-polarized image. The process proceeds to the image capturing process (S10). The road surface state detection apparatus 1 repeats the above processing until the fluctuation of the median value of the polarization difference data converges.

  At the stage of processing the first frame, the polarization difference data storage unit 32 stores the polarization difference data of effective pixels in one frame, and the median value of the polarization difference data is obtained for the first time at this stage. Has not converged. Therefore, the road surface state detection apparatus 1 shifts to a process for imaging the road surface (S10). The road surface state detection apparatus 1 repeats the same operation as described above, obtains the polarization difference data of the pixel included in the second frame (S14), and stores it in the polarization difference data storage unit 32 (S16). At this stage, the polarization difference data storage unit 32 stores the polarization difference data of pixels in one frame and two frames. The control unit 18 reads out the polarization difference data stored in the polarization difference data storage unit 32, and obtains the median value of the polarization difference data of the pixels in one frame and two frames (S18). Hereinafter, similarly, a new frame is imaged (S10), and the median value of the polarization difference data of the pixels of all imaging frames including the polarization difference data of the newly imaged frame is obtained (S18), and the variation of the median value has converged. It is determined whether or not (S20).

  When it is determined that the median has converged (YES in S20), that is, when it is determined that the determination of the road surface state is permitted, the control unit 18 proceeds to a process of determining the road surface state (S22). The control unit 18 may stop the imaging of the road surface by the camera 12. The control unit 18 compares the median value of the polarization difference data with a predetermined threshold value, determines that the road surface is wet when the median value is smaller than the predetermined threshold value, and when the median value is equal to or greater than the predetermined threshold value. Judged to be dry.

Next, the road surface state determination apparatus 1 notifies the vehicle of the determination result of the road surface state, and the vehicle controls the headlight 10 based on the road surface state (S24). Specifically, the vehicle controls the optical axis of the headlight 10 up and down based on the road surface condition. When the road surface state changes from dry to wet, the optical axis of the headlight 10 is raised, and when the road surface state changes from wet to dry, the optical axis of the headlight 10 is lowered.
The configuration and operation of the road surface state detection device 1 according to the first embodiment have been described above.

  The road surface state detection apparatus 1 according to the present embodiment determines the road surface state using the median value of polarization difference data of a plurality of pixels obtained from an image captured over a predetermined time while the vehicle is traveling. Even if an image of a road surface with ambient light is included in the part, the influence of the polarization difference data caused by the image with ambient light can be reduced. Can be determined.

  Note that the road surface condition detection device 1 according to the present embodiment is based on the premise that the area where the influence of the environmental light is small is wider than the area where the influence of the environmental light is large on the road surface. For example, streetlights that cause periodic ambient light are stipulated by law to be installed every 20-30 m. However, the installation interval is short at intersections and roads with poor visibility.

  FIG. 8 is a diagram showing a region where the influence of the environmental light is large and a region where the influence of the environmental light is small, taking as an example a case where the interval between the street lights is 30 m. As shown in FIG. 8, since the area (distance) where the influence of the streetlight is strong is shorter than between the streetlights, the area between the streetlights where the influence of environmental light is small is smaller than the area where the influence of environmental light is great. In a predetermined road space (distance), the road surface state is observed, and more evaluation images that are less affected by ambient light (for example, street lights) are obtained than evaluation images that are more affected by ambient light. Polarization difference data of the obtained evaluation images is obtained. When sorted according to size, polarization difference data that is less affected by ambient light becomes the median value. In the present embodiment, the road surface condition can be appropriately determined using this median value. In addition, regarding the headlight of the oncoming vehicle, when the distance from the oncoming vehicle is reduced, the road surface irradiated by the headlight of the oncoming vehicle becomes sideways, so that it cannot be seen from the camera of the own vehicle and the influence is reduced. Therefore, by observing the road surface state in a predetermined road space and determining the road surface state using the median value, the road surface state can be appropriately determined.

  Further, in the road surface state detection device 1 according to the present embodiment, the control unit 18 determines whether or not the fluctuation of the median value has converged, and permits the determination of the road surface state when the fluctuation has converged, and the converged center. Since the road surface state is obtained using the value, the polarization difference data of the pixel showing the point where the influence of the environmental light is small can be used for the determination of the road surface state.

  Further, in the road surface condition detection apparatus 1 of the present embodiment, the headlight 10 can use existing equipment, and the camera 12 is hardware that can be used for other purposes such as monitoring the situation ahead. That is, the road surface state detection device 1 can be realized without using additional hardware necessary only for detecting the road surface state.

(Second Embodiment)
FIG. 9 is a diagram illustrating a configuration of the road surface state detection device 2 according to the second embodiment. The basic configuration of the road surface state detection device 2 of the second embodiment is the same as that of the road surface state detection device 1 of the first embodiment, but the median value of the polarization difference data by the median value calculation unit 26 is the same. The way to ask is different.

  FIG. 10A and FIG. 10B are diagrams for explaining the principle of how the median value is calculated by the median value calculation unit 26. As shown in FIG. 10A, the median value calculation unit 26 performs processing for each frame, and obtains a representative value for each frame of the polarization difference data. As the representative value, a median value of polarization difference data of pixels included in the frame may be used, or an average value may be used. The median value calculation unit 26 stores the obtained representative value in the representative value storage unit 34. The median value calculation unit 26 reads the representative value from the representative value storage unit 34 and obtains the median value of the representative value as shown in FIG.

  FIG. 11 is a flowchart illustrating the operation of the road surface condition detection device 2 according to the second embodiment. The road surface state detection device 2 captures an S-polarized image and a P-polarized image of the road surface with the camera 12 while the vehicle is traveling (S30). The control unit 18 of the road surface state detection device 2 removes noise that is not caused by the influence of ambient light from the captured S-polarized image and P-polarized image (S32), and obtains polarization difference data of effective pixels in the frame. (S34) The obtained polarization difference data is stored in the polarization difference data storage unit 32 (S36). Up to this point, the operation is the same as the operation of the road surface condition detection device 1 according to the first embodiment.

  After storing the polarization difference data in the polarization difference data storage unit 32 (S36), the control unit 18 calculates a representative value of the frame of the polarization difference data using the stored polarization difference data (S38). The control unit 18 stores the representative value in the representative value storage unit 34 (S40), and deletes the data stored in the polarization difference data storage unit 32 (S42). Thereby, the representative value storage unit 34 stores the representative value of the polarization difference data of each frame. Subsequently, the control unit 18 obtains the median value of the polarization difference data using the representative value of the polarization difference data frame stored in the representative value storage unit 34 (S44).

  After obtaining the median value, it is determined whether the median value has converged (S46) as in the road surface condition detection device 1 of the first embodiment, and it is determined that the median value has not converged. In this case (NO in S46), a new S-polarized image and P-polarized image are captured by the camera 12 (S30), and the above-described processing is repeated. If it is determined that the median has converged (YES in S46), the road surface condition is determined using the median (S48). After determining the road surface state, the road surface state determination device 2 notifies the vehicle of the determination result of the road surface state, and the vehicle controls the headlight 10 based on the road surface state (S49).

  The road surface condition detection apparatus 2 according to the second embodiment obtains a representative value of the polarization difference data for each frame, stores it in the representative value storage unit 34, and deletes the polarization difference data of the frame for which the calculation of the representative value is completed. Therefore, it is not necessary to store the polarization difference data of all the frames until the detection of the road surface state is completed. Therefore, a wide range of imaging data can be processed with a small memory.

  Moreover, the road surface state detection apparatus 2 of this Embodiment reduces the influence of environmental light and can determine a road surface state appropriately similarly to the road surface state detection apparatus 1 of 1st Embodiment. Have

  In this embodiment, an example in which the representative value of a frame is obtained and the median of the representative values of a plurality of frames is obtained has been described. However, the median of a frame is obtained and the median of each of the plurality of frames is obtained. Even when the representative value is obtained, the road surface state can be appropriately determined as in the present embodiment.

(Third embodiment)
FIG. 12 is a diagram illustrating a configuration of the road surface state detection device 3 according to the third embodiment. The basic configuration of the road surface state detection device 3 of the third embodiment is the same as that of the road surface state detection device 1 of the first embodiment, except that the control unit 18 includes a road paint determination unit 38. Different. The road paint determination unit 38 determines whether the captured image includes road paint. The road paint is, for example, paint for lane guidance for turning right or left in the vicinity of an intersection or paint such as a pedestrian crossing. The road paint has a higher reflectance than the part without the road paint. In particular, white road paint contains glass beads and has a very high reflectance. For this reason, road paint may cause erroneous detection of road surface conditions.

  The road surface state detection apparatus 3 includes a reference image storage unit 36 that stores a P-polarized image and an S-polarized image that do not include road paint as reference images. The reference image storage unit 36 captures and stores in advance a typical image that does not include road paint.

  The road paint determination unit 38 compares the polarization difference data histogram of the reference image read from the reference image storage unit 36 with the polarization difference data histogram of the captured image, so that the captured image is subjected to road paint. It is determined whether the image is included. When the control unit 18 determines that the captured image includes road paint, the control unit 18 deletes the polarization difference data obtained from the image from the polarization difference data storage unit 32. Thereby, the road surface state detection apparatus 3 determines the road surface state using the polarization difference data excluding the polarization difference data obtained from the image including the road paint.

  FIG. 13 is a flowchart illustrating the operation of the road surface condition detection device 3 according to the third embodiment. The road surface state detection device 3 captures an S-polarized image and a P-polarized image of the road surface with the camera 12 while the vehicle is traveling (S50). The control unit 18 of the road surface state detection device 3 removes noise that is not caused by the influence of ambient light from the captured S-polarized image and P-polarized image (S52), and obtains polarization difference data of effective pixels in the frame. (S54) The obtained polarization difference data is stored in the polarization difference data storage unit 32 (S56). Up to this point, the operation is the same as the operation of the road surface condition detection device 1 according to the first embodiment.

  Next, the control unit 18 performs road surface paint determination processing for determining whether or not the captured image includes road paint (S58). The control unit 18 compares the histogram of the polarization difference data stored in the polarization difference data storage unit 32 with the histogram of the polarization difference data of the reference image stored in the reference image storage unit 36 to determine the presence or absence of road paint. judge. An example of the road paint determination process will be described later.

  Based on the result of the road paint determination process (S58), the control unit 18 determines whether the captured image includes road paint (S60). When it is determined that the captured image includes road paint (YES in S60), the control unit 18 performs processing to delete the polarization difference data corresponding to the image including road paint from the polarization difference data storage unit 32 (S62). ). After deleting the polarization difference data of the image including the road paint, or when it is determined that the road paint is not included in the road paint determination process (NO in S60), the polarization difference data stored in the polarization difference data storage unit 32 And the median of the polarization difference data is calculated (S64).

  After obtaining the median value, it is determined whether the median value has converged (S66) as in the road surface condition detection device 1 of the first embodiment, and it is determined that the median value has not converged. In this case (NO in S66), a new S-polarized image and P-polarized image are captured by the camera 12 (S50), and the above-described processing is repeated. If it is determined that the median has converged (YES in S66), the road surface condition is determined using the median (S68). After determining the road surface state, the road surface state determination device 2 notifies the vehicle of the determination result of the road surface state, and the vehicle controls the headlight 10 based on the road surface state (S49).

  Next, an example of the road paint determination process will be described. However, the road paint determination process of the present invention is not limited to the process described below.

  FIG. 14 is a flowchart showing a detailed operation of the road paint determination process. The control unit 18 reads the reference image stored in the reference image storage unit 36 (S70), and calculates the polarization difference data of the pixels included in the read reference image (S72). Next, the control unit 18 generates a histogram of the polarization difference data of the reference image (S74). In the present embodiment, an example of obtaining a histogram of polarization difference data using a reference image read from the reference image storage unit 36 when determining whether or not the image includes a road paint image will be described. However, a histogram of the polarization difference data of the reference image may be obtained in advance, and the histogram may be stored in the reference image storage unit 36.

  Next, the control unit 18 uses the data stored in the polarization difference data storage unit 32 to generate a histogram of the polarization difference data of the pixels in the determination target frame (S76). In this example, the captured image to be determined is described as an image including road paint. The control unit 18 removes the component having the polarization difference data “+” from both the polarization difference data histogram of the reference image and the polarization difference data histogram of the captured image (S78). This is because an image including road paint is characterized by the “−” component of the polarization component data, and therefore the presence or absence of road paint can be determined by comparing only the “−” component.

  FIG. 15A is a diagram illustrating a histogram of polarization difference data between a reference image and a captured image. In FIG. 15A, only the “−” component of the polarization difference data is shown. As shown in FIG. 15A, in the captured image to be determined, the polarization difference data has a peak in the vicinity of −50. That is, having a peak in the vicinity is a feature of an image including road paint.

  The control unit 18 obtains the absolute value of the difference between the histogram of the reference image and the histogram of the captured image (S80). FIG. 15A shows a graph of the absolute value of the difference with a bold line. Although two peaks appear in the difference graph, the distribution having the larger peak of the “−” component (the left peak in FIG. 15A) is the distribution due to the reflected light from the road paint. Therefore, whether or not the captured image includes a road paint image is determined by determining whether or not the histogram distribution of the captured image has a distribution having a larger peak of the “−” component of the histogram of the captured image. Can be determined. The control unit 18 shifts the difference graph so that the valley portions of the two peaks in the difference graph become 0 (S82).

  FIG. 15B is a diagram illustrating a difference graph after shift correction. Next, the control unit 18 calculates the average value of the “−” components of the difference graph shown in FIG. 15B (S84), and compares the average value with a predetermined threshold value (S86). When it is determined that the average value is larger than the predetermined threshold value (YES in S86), the control unit 18 determines that the determination target image includes a road paint image (S88), and the average value is equal to or less than the predetermined threshold value. If it is determined (NO in S86), it is determined that the determination target image does not include a road paint image (S90). In the present embodiment, the example in which it is determined whether or not the captured image includes the road paint image based on the average value of the “−” components of the graph shown in FIG. 15B has been described. Is not limited to the method using the average value, and for example, the distribution forms of the “−” components may be compared. The operation of the road paint determination process has been described in detail above.

  The road surface state detection device 3 according to the third embodiment determines whether or not the captured image includes a road paint image. If the road paint image is included, the polarization difference corresponding to the image is determined. Delete the data. As a result, the road surface state can be detected using only the image that does not include the road paint image, so that the road surface state can be detected quickly. Even if there is an image including road paint, the road surface state can be detected appropriately as in the first embodiment or the second embodiment described above, but by excluding the image including road paint from the calculation target. The median of the polarization difference data converges faster.

  The road surface state detection device 3 of the present embodiment also has the effect of reducing the influence of ambient light and appropriately determining the road surface state, as with the road surface state detection device 1 of the first embodiment. Have

(Fourth embodiment)
FIG. 16 is a diagram illustrating a configuration of the road surface state detection device 4 according to the fourth embodiment. The road surface state detection device 4 is connected to a navigation device 46. The basic configuration of the road surface state detection device 4 according to the fourth embodiment is the same as that of the road surface state detection device 1 according to the first embodiment. However, the camera image acquisition control unit 20 performs a camera image acquisition control method. Is different.

  The navigation device 46 includes a vehicle position detection unit 40 that detects the current position of the vehicle, a map information storage unit 42 that stores map information including information such as roads and facilities, and a control unit 44 that controls the navigation device 46. Have. The vehicle position detection unit 40 has a function of detecting the current position of the vehicle. As the vehicle position detection unit 40, for example, a geomagnetic sensor, a gyroscope, a distance sensor, a GPS receiver, a steering rotation sensor, a vehicle sensor for each rolling wheel, or the like can be used. The vehicle position detection unit 40 may be configured using some of these sensors. Since these sensors have errors of different properties, the current position of the vehicle can be detected with high accuracy by complementing with a plurality of sensors. The control unit 44 has a function of acquiring a road environment in which the vehicle is traveling in accordance with a request from the road surface state detection device 4 in addition to the function of a normal navigation device. The road environment is information on roads and facilities around a running vehicle.

  The camera image acquisition control unit 20 acquires the road environment in which the vehicle is traveling from the navigation device 46. The camera image acquisition control unit 20 determines whether or not the vehicle is in the vicinity of the intersection based on the acquired road environment, and stops the acquisition of the camera image when the vehicle is in the vicinity of the intersection. The camera image acquisition control unit 20 sets a predetermined threshold (for example, 10 m) in advance, determines that the vehicle is near the intersection when the vehicle is within the threshold from the intersection, and when the vehicle is more than the threshold from the intersection. It can be determined that it is not near the intersection.

  FIG. 17 is a flowchart illustrating the operation of the road surface condition detection device 4 according to the fourth embodiment. The road surface condition detection device 4 first acquires the road environment in which the vehicle is traveling from the navigation device 46 (S100). Specifically, in this process, the navigation device 46 detects the current position of the vehicle in response to a request from the road surface state detection device 4, and the detected current position of the vehicle and the map stored in the map information storage unit 42. Based on the information, information on the road environment is obtained. Then, the navigation device 46 transmits the obtained road environment information to the road surface state detection device 4. The road surface state detection device 4 determines whether or not the vehicle is near the intersection based on the road environment information (S102).

  If it is determined that the vehicle is near the intersection (YES in S102), the process returns to the process of detecting the current position of the vehicle (S100), and the above process is repeated until it is determined that the vehicle is not near the intersection. When it is determined that the vehicle is not near the intersection (NO in S102), the road surface state is detected by the same process as the road surface state detection device 1 of the first embodiment.

  Since the road surface state detection device 4 according to the fourth embodiment stops the imaging of the road surface when the vehicle is near the intersection, the imaging of the image including a lot of road paint near the intersection is avoided, and the road paint is not included. The road surface condition can be detected using the image.

  The road surface state detection device 4 of the present embodiment also has the effect of reducing the influence of ambient light and appropriately determining the road surface state, like the road surface state detection device 1 of the first embodiment. Have

  In this embodiment, an example has been described in which an intersection is assumed as a place where there is a lot of road paint, and imaging is stopped near the intersection. If stored, imaging may be stopped at a point where there is a pedestrian crossing or street lamp.

  As mentioned above, although the road surface condition detection apparatus of the present invention has been described in detail with reference to the embodiment, the present invention is not limited to the above-described embodiment.

  In any of the above-described embodiments, the difference between the S change component and the P polarization component is used as the comparison data between the S polarization component and the P polarization component, but the difference between the S change component and the P polarization component is used as this comparison data. Even if the ratio between the S-polarized light component and the P-polarized light component is used instead of, the road surface state can be detected with high accuracy as in the above-described embodiment.

  In the above-described embodiment, the road surface state determination permitting unit 25 determines whether or not to permit the determination of the road surface state depending on whether or not the variation in the median value of the polarization difference data has converged. The determination method by the determination permission unit 25 is not limited to the method described in the above embodiment. For example, the road surface state determination permission unit 25 determines whether or not the number of data of the acquired S-polarized component and P-polarized component exceeds a predetermined threshold, and determines the road surface state when the number of data exceeds the predetermined threshold. May be allowed. As the threshold, for example, a sufficiently large number of frames (for example, 100 frames) or an imaging time (for example, 10 seconds) is set. When the number of data exceeds a predetermined threshold value, the S-polarized component and the P-polarized component are acquired from a wide range. Therefore, the influence of environmental light can be reduced and the road surface condition can be appropriately determined. it can. In addition, since it is not necessary to calculate the median value every time data on the S polarization component and the P polarization component is newly acquired, the program can be simplified. Further, by adopting such a configuration, when a wide range of images can be obtained by one imaging, it is possible to determine the road surface state using a polarization image obtained by one imaging.

  Further, the number of frames and the predetermined time for detecting the road surface condition may be changed according to the traveling speed of the vehicle. The “predetermined time” here does not mean the shutter time. As described above, since it is necessary to observe a road surface of a predetermined length in order to acquire an image with little influence of environmental light such as a streetlight, a camera image is acquired according to the traveling speed of the vehicle. You need to change the time. When the traveling speed of the vehicle is slow, even if camera images for the same time as when the traveling speed is fast are acquired, only a small range of images can be obtained compared to when the traveling speed is fast. For example, in the case of a vehicle traveling at a speed of 60 km / h, an image for 100 m can be obtained by acquiring a camera image for 6 seconds. In a vehicle traveling at a speed of 40 km / h, in order to obtain an image for 100 m, a camera image for 9 seconds is required. Therefore, when the traveling speed is low, it is preferable to detect the road surface state using an image captured for a long time.

  In the above-described embodiment, the example in which the road surface state is detected using all the pixels determined not to be noise of the captured frame has been described, but it is not always necessary to use all the pixels. For example, the captured frame may be divided into a plurality of blocks (for example, 5 × 5 blocks), one pixel is selected from each of the divided blocks, and the road surface state is detected using the selected pixels. Good. As a result, the road surface state can be detected with a small amount of data, so that the polarization difference data storage unit 32 for storing the polarization difference data can be configured by a small capacity memory. In addition, when selecting the pixel used for calculation of road surface state detection, it is preferable to select the pixel without positional deviation so as not to acquire only the pixel including the ambient light.

  In the second embodiment described above, an example has been described in which a representative value is obtained for each frame and a median value of each representative value of a plurality of frames is obtained. However, instead of obtaining a representative value for each frame, a plurality of representative values are obtained. The representative value may be obtained for each frame. At this time, a part may be overlapped between a plurality of frames for which the representative value is obtained. For example, the representative value of frame 1 to frame 5, the representative value of frame 4 to frame 8, the representative value of frame 7 to frame 11, etc. Also good. As a result, the range for which the representative value is obtained can be widened without reducing the frequency of obtaining the representative value, so that the road surface condition can be quickly determined.

  In the third embodiment described above, whether or not road paint is included is determined by comparing the captured image with the reference image, but the presence or absence of road paint may be determined by another method. For example, the color included in the captured image may be analyzed, and if there are more white pixels than a predetermined threshold, it may be determined that road paint is included.

  As described above, the present invention has an excellent effect that a road surface state can be detected with high accuracy, and can be applied to a driving support device that supports driving by detecting a road surface state. For example, by applying the present invention to a headlight control system (AFS), visibility can be maintained regardless of the road surface condition. When the road surface is dry, the light from the headlight is scattered on the road surface to obtain almost uniform visibility at both long and short distances. However, when the road surface is wet, the headlight light is reflected at a short distance. Becomes stronger and is dazzling for the driver of the host vehicle, so visibility is reduced. By detecting the road surface state by the road surface state detection device of the present invention, the headlight is slightly raised when the road surface is wet. Thereby, the reflection of the light of the headlight from the road surface is substantially uniform at both a long distance and a short distance, and the visibility of the driver of the host vehicle can be prevented from deteriorating. As another example, when the road surface is wet, the friction between the tire and the road surface becomes small and slippery, so that the vehicle is controlled based on the detection result of the road surface condition, so that slip is less likely to occur. be able to.

It is a figure which shows the structure of the road surface state detection apparatus of 1st Embodiment. It is a figure explaining a P polarization component and a S polarization component. It is a figure which shows the example of a road surface state detection program. (A) A histogram of polarization difference data included in an image of a plurality of frames. (B) A histogram of polarization difference data included in an image of a frame that is less affected by ambient light. (C) A histogram of polarization difference data included in an image of a frame having a large influence of ambient light. It is a figure which shows the relationship between the median of polarization | polarized-light difference data, and the number of frames. It is a figure which shows the relationship between polarization difference data and a road surface state. It is a figure which shows operation | movement of the road surface state detection apparatus of 1st Embodiment. It is a figure which shows the example of the area | region where the influence of environmental light is large, and the area | region where the influence of environmental light is small. It is a figure which shows the structure of the road surface state detection apparatus of 2nd Embodiment. (A) It is a figure which shows the example which calculated the representative value for every flame | frame. (B) It is a figure which shows the example which calculated | required the median value of the representative value. It is a figure which shows operation | movement of the road surface state detection apparatus of 2nd Embodiment. It is a figure which shows the structure of the road surface state detection apparatus of 3rd Embodiment. It is a figure which shows operation | movement of the road surface state detection apparatus of 3rd Embodiment. It is a figure which shows the operation | movement of a road paint determination process. (A) It is a figure which shows the histogram of a captured image and a reference | standard image, and its difference. (B) It is a figure which shows the example which shifted the difference. It is a figure which shows the structure of the road surface state detection apparatus of 4th Embodiment. It is a figure which shows operation | movement of the road surface state detection apparatus of 4th Embodiment.

Explanation of symbols

1-4 Road surface state detection device 10 Headlight 12 Camera 14 P-polarized image acquisition unit 16 S-polarization image acquisition unit 18 Control unit 20 Camera image acquisition control unit 22 Noise removal unit 24 Polarization difference data calculation unit 25 Road surface state determination permission unit 26 Median value calculation unit 28 Convergence determination unit 30 Road surface state determination unit 32 Polarization difference data storage unit 34 Representative value storage unit 36 Reference image storage unit 38 Road paint determination unit 40 Vehicle position detection unit 42 Map information storage unit 44 Control unit 46 Navigation device 50 Road surface state detection program 52 Camera image acquisition control subprogram 54 Noise removal subprogram 56 Polarization difference calculation subprogram 57 Road surface state determination permission subprogram 58 Median value calculation subprogram 60 Convergence determination subprogram 62 Road surface state determination subprogram

Claims (14)

A polarization component acquisition unit that acquires an S-polarized component and a P-polarized component of reflected light from the road surface;
A road surface state determination unit that determines a road surface state based on the S polarization component and the P polarization component;
With
The road surface state determination unit
Road surface for determining the comparison data of the plurality of S polarization components and the P polarization component acquired by the polarization component acquisition unit, obtaining representative values of the obtained plurality of comparison data, and determining the road surface state based on the representative values State detection device. A road surface state determination permission unit that determines whether or not to permit the determination of the road surface state using the data of the S polarization component and the P polarization component acquired by the polarization component acquisition unit,
When it is determined that the determination of the road surface state is permitted, the road surface state determination unit performs the determination of the road surface state, and when it is determined that the determination of the road surface state is not permitted, the polarization component acquisition unit performs S polarization The road surface state detection device according to claim 1, which acquires a component and a P-polarized component.   The road surface state detection device according to claim 2, wherein the road surface state determination permission unit determines whether or not to permit determination of a road surface state based on the acquired number of data of the S polarization component and the P polarization component.   The road surface condition determination permission unit obtains the comparison data of the newly acquired S-polarized component and P-polarized component, and from the comparison data obtained by combining the newly obtained comparison data and a plurality of comparison data obtained up to the previous time The road surface state detection device according to claim 2, wherein the process of obtaining the representative value is repeatedly performed, and the determination of the road surface state is permitted when the variation of the representative value repeatedly calculated by the process converges.   The road surface state determination unit obtains and stores a median value or an average value of the comparison data for each predetermined unit determined in advance, and obtains the representative value using the stored median value or average value. The road surface condition detection apparatus in any one of -4. The polarization component acquisition unit
A polarization image capturing unit that captures an S-polarized image and a P-polarized image of the road surface;
The road surface state detection device according to any one of claims 1 to 5, wherein the S-polarized component and the P-polarized component of a plurality of pixels selected from the captured S-polarized image and P-polarized image themselves without positional deviation are obtained. The polarization component acquisition unit
S-polarized light of a plurality of pixels selected without positional deviation from all the pixels of the plurality of frames based on the S-polarized images and P-polarized images of the plurality of frames captured over a predetermined time during traveling of the vehicle The road surface state detection device according to claim 1, wherein the road surface state detection device calculates a component and a P-polarized component.   The road surface state determination unit determines that the road surface is wet when the P-polarized component is larger than the S-polarized component based on the representative value, and the road surface is determined when the S-polarized component and the P-polarized component are substantially equal. The road surface state detection device according to any one of claims 1 to 7, wherein the road surface state detection device determines that the state is dry.   The road surface state detection device according to any one of claims 6 to 8, wherein the road surface state determination unit excludes the frame from the road surface determination when the frame captured by the polarization image capturing unit includes an image of road paint. . A reference image storage unit that stores a frame that does not include an image of road paint as a reference image;
The road surface state determination unit includes a histogram of comparison data of S-polarized light components and P-polarized light components of pixels selected without positional deviation from all the pixels of the frame captured by the polarized image capturing unit, and the reference image 10. It is determined whether road paint is included in the frame by comparing a histogram of comparison data of S-polarized light components and P-polarized light components of pixels selected without any positional deviation among all the pixels. The road surface state detection device according to 1.   The road surface state determination unit acquires a road environment in which the vehicle is traveling based on the map information and the current position information of the vehicle, and whether the vehicle is traveling in a place with a lot of road paint based on the road environment. The road surface condition detection device according to any one of claims 1 to 10, wherein the polarization component acquisition unit is controlled to stop acquisition of the S-polarized component and the P-polarized component while traveling in a place with a lot of road paint. .   The road surface state detection device according to any one of claims 1 to 11, wherein the road surface state determination unit obtains a median value of a plurality of comparison data as the representative value. Obtaining a plurality of S-polarized light components and P-polarized light components included in the reflected light from the road surface;
Obtaining comparison data of each of the plurality of S-polarized components and P-polarized components;
Obtaining a representative value of a plurality of obtained comparison data;
Determining a road surface state based on the representative value;
A road surface condition detection method comprising: A program for detecting a road surface state by a computer including a polarization component acquisition unit that acquires an S-polarized component and a P-polarized component of reflected light from the road surface.
Obtaining a plurality of S and P polarization components;
Obtaining comparison data of each of the plurality of S-polarized components and P-polarized components;
Obtaining a representative value of a plurality of obtained comparison data;
Determining a road surface state based on the representative value;
A program that executes