JP2007208865A - System for detecting camera state - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To objectively and quantitatively detect the presence/absence of the fault of a camera (including an optical system) in a system for detecting a camera state. <P>SOLUTION: The system includes: a camera 10 loaded on the rear upper part of a vehicle 200 and provided with a focusing distance fixed optical system for photographing the scene (image) P of surroundings (for instance, the rear part of a vehicle); a BPF 30 for extracting the high frequency components SH of a high space frequency of the image P photographed by the camera 10; an area selection part 61 and an area selection circuit 62 for receiving the input of an instruction for selecting a partial area range p of the image P and selecting the partial area range p; an integration circuit 40 for calculating the integral value ∫<SB>p</SB>(SH) of the high frequency components SH for the area range p; and a stain decision circuit 50 for determining the presence of the stain (fault) of the camera 10 or the optical system provided in the camera 10 by comparing the integral value ∫<SB>p</SB>(SH) with a reference value K. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a camera state detection system, and more particularly to detection of a state of a camera that is mounted on a vehicle and photographs the periphery of the vehicle.

  2. Description of the Related Art In recent years, an in-vehicle camera system that supports a vehicle occupant's driving operation by mounting a camera and a monitor on a vehicle and displaying the image information obtained by photographing with the camera as a visible image on the monitor has become widespread.

For example, a parking assist system or a side monitor system that takes a picture of the vehicle periphery such as the rear or side of the vehicle and displays the taken image or video on an in-vehicle monitor to reduce the labor of the backward operation or the width adjustment operation. (Patent Document 1).
JP 2001-055099 A

  By the way, since the camera used for these vehicle-mounted camera systems is fixed to the vehicle, the distance to the shooting target range is unchanged.

  However, when an obstacle collides with the camera, the distance relationship between the camera and the periphery of the vehicle that is the subject of photographing may change.

  In such a case, an image or video obtained by photographing with a camera is not an image obtained in an appropriate in-focus state or the like, and thus is an out-of-focus image or the like.

  However, even if the camera captures such an out-of-focus image, it is not always necessary to determine whether the occupant is out of focus by observing the visible image displayed on the in-vehicle monitor. It may not be possible to properly recognize the presence or absence of a fault.

  That is, it cannot be determined quantitatively only by human eyes, and it cannot be distinguished whether the camera has a fault or the monitor has a fault.

  In addition, when the optical system is soiled or damaged, such as when oil adheres to the optical system provided in the camera, or when it becomes cloudy, the contrast of the image obtained by shooting decreases or the sharpness Therefore, a phenomenon similar to the shift of the camera installation position described above is exhibited.

  Conventionally, it has not been possible to detect objectively and quantitatively whether or not these cameras (including the optical system) are faulty.

  The present invention has been made in view of the above circumstances, and is a camera that can objectively and quantitatively detect the presence or absence of a failure in a camera (including an optical system) that is mounted on a vehicle and photographs the periphery of the vehicle. An object is to provide a state detection system.

  The camera state detection system according to the present invention calculates an integral value of a high-frequency component for image information obtained by a vehicle-mounted camera that captures the periphery of the vehicle, and compares the obtained integral value with a predetermined reference value. By doing so, it is possible to detect the presence or absence of a camera failure.

  That is, a camera state detection system according to the present invention includes a camera that is mounted on a vehicle and captures the periphery of the vehicle, and a high-frequency component extraction unit that extracts a high-frequency component from image information captured by the camera. An integral value calculating means for calculating an integral value of a part or all of the image information of the high frequency component extracted by the high frequency component extracting means; and the integral value calculated by the integral value calculating means. And camera state detection means for detecting a failure of the camera.

  Here, as the high-frequency component extracting means, in addition to means for performing high-frequency component extraction processing in the real image region such as a bandpass filter and a high-pass filter that allow high-frequency components to pass through, in the spatial frequency region using discrete Fourier transform and discrete cosine transform. It may be a means for performing a high-frequency component extraction process.

  The integral value of the high frequency component calculated by the integral value calculating means means an integrated value of the level of the high frequency component, and does not mean an integrated value of the frequency (frequency) of the high frequency component.

  At least two of the high-frequency component extraction unit, the integral value calculation unit, and the camera state detection unit may be integrally configured. For example, the camera state detection unit includes the high-frequency component extraction unit and the integral value calculation unit. These functions may also be used.

  A camera failure is caused by, for example, contamination of the optical system provided in the camera, damage, etc., displacement of the installation position of the camera relative to the vehicle on which the camera is mounted (especially displacement of the distance along the optical axis direction). ) Etc.

  Such contamination and damage of the optical system, and deviations in the installation position and angle of the camera are such that the image or video obtained by the camera is blurred or out of focus compared to the case without such obstacles. It becomes a blurred image or video, and the sharpness of the image or video decreases.

  In the camera state detection system of the present invention, the high frequency component extraction means extracts high frequency components from the image information obtained by photographing with the camera, and the integral value calculation means integrates the high frequency components extracted by the high frequency component extraction means. In order to calculate the value, the integrated value calculated when there is dirt or damage in the optical system, or when the installation position or installation angle of the vehicle with respect to the ground is shifted due to contact with the camera, etc. It is smaller than the integral value (reference value) calculated when there is no such obstacle.

  Therefore, according to the camera state detection system of the present invention, the camera state detection unit can objectively and quantitatively detect a camera failure based on the integral value calculated by the integral value calculation unit.

  In addition, a technique (described in Japanese Patent Application Laid-Open No. 2001-119614) for detecting contamination of the optical system according to the height of the spatial frequency of the image or the magnitude of the differential value has been proposed. Whereas the camera state detection system according to the present invention is determined according to the integral value of the high-frequency component, it is easily affected by the noise generated accidentally in a part of the image, and erroneous determination is likely to occur. Such a fear can be suppressed.

  In addition, a technique (described in Japanese Patent Application Laid-Open No. 2003-189294) for detecting contamination of an optical system in accordance with the ratio of a further high frequency component in the high frequency component of an image has been proposed.

  In the camera state detection system according to the present invention, it is preferable that the camera state detection unit determines whether or not there is a failure by comparing the integral value with a preset reference value.

  The calculated integral value becomes smaller as the sharpness of the image and video obtained by photographing with the camera decreases.

  Therefore, by setting an integral value of a high frequency component of an image or video obtained by photographing with the camera in a state where no trouble has occurred in the camera as a reference value, the calculated integral value is Thus, it is possible to easily detect the presence or absence of a camera failure by a simple process of simply comparing the size with a preset reference value.

  In the camera state detection system according to the present invention, the camera state detection unit preferably serves as at least one of the high-frequency component extraction unit and the integral value calculation unit.

  Since the camera state detection means serves as at least one of the high-frequency component extraction means and the integral value calculation means, the number of components can be reduced, and the size can be reduced according to the reduction in the number of parts. And manufacturing cost can be reduced.

  In the camera state detection system according to the present invention, the camera preferably includes an optical system with a fixed focal length.

  A camera equipped with an optical system that can change the focal length is configured so that the optical system can be displaced in the direction of the optical axis in accordance with the distance to the object to be photographed, and the in-focus state can be adjusted.

  A camera having such a variable focal length optical system includes an autofocus device such as a contrast evaluation method as represented by a general digital camera (Japanese Patent Laid-Open No. 2005-003933). etc).

  However, since the subject to be photographed is limited to a predetermined distance range such as the periphery of the vehicle, the vehicle-mounted camera for photographing the periphery of the vehicle that does not need to change the focal length includes an optical system with a fixed focal length. However, unlike a camera having an optical system with variable focal length, it does not have an autofocus device.

  In particular, when a camera having such an optical system with a fixed focal length is mounted on a vehicle, it has been considered that the installation position of the camera with respect to an object to be photographed such as the ground does not change.

  However, the installation position of the camera relative to the object to be photographed such as the ground may change due to the above-described contact with the camera or the like.

  Therefore, in a camera provided with such an optical system with a fixed focal length, the effects exhibited by the above-described camera state detection system of the present invention can be made more remarkable.

  In the camera state detection system according to the present invention, the camera state detection system further includes a travel detection unit that detects the travel of the vehicle, and the camera state detection unit detects a failure of the camera when the travel of the vehicle is detected by the travel detection unit. Is preferably repeatedly detected.

  When the vehicle is stopped, the vehicle is often parked near a monotone wall or stopped on a road without a road sign.

  In such a situation, the image information corresponding to the image and video around the vehicle photographed by the camera may not contain many high-frequency components in the first place.

  On the other hand, in the state where the vehicle is traveling, the images and videos taken by the camera change from moment to moment, so there is no situation where the obtained image information always contains only a few high-frequency components.

  For example, when the subject to be photographed by the camera includes a road surface in the vicinity of the vehicle, there is always an opportunity to photograph the sign drawn on the road surface while the vehicle is traveling. A high frequency component of spatial frequency is included in the contour.

  Therefore, when the vehicle travels, that is, when the travel detection unit detects the travel of the vehicle, the camera state detection unit repeatedly detects the failure state of the camera, thereby reliably and accurately compared to when the vehicle is stopped. It is possible to detect a failure state of the camera.

  In the camera state detection system according to the present invention, the camera state detection system further includes area selection means for selecting a part of the entire area of the image information obtained by the camera, and the high frequency component extraction means is selected by the area selection means. The high-frequency component is extracted only for the partial area that has been selected, or the integral value calculation means calculates the integral value only for the partial area selected by the area selection means. Is preferred.

  Even if the high-frequency component extraction means extracts high-frequency components only for the partial area selected by the area selection means, the integral value calculation means is for the entire area of the image information obtained by photographing with the camera. An integral value may be calculated, or an integral value may be calculated only for a selected partial region.

  Even if an integral value is calculated for the entire area of the image information, high-frequency components are extracted only for the selected partial area, so for other remaining areas excluding the partial area. This is because no high-frequency component is detected, and as a result, the integral value for the entire region of the image information and the integral value for the selected partial region are the same.

  Even if the integral value calculating means calculates the integral value only for the partial area selected by the area selecting means, the high frequency component extracting means is for the entire area of the image information obtained by photographing with the camera. A high frequency component may be extracted, or a high frequency component may be extracted only for a selected partial region.

  Even when high-frequency components are extracted for the entire area of the image information, the integral value is calculated only for the selected partial area. The remaining area is excluded from the target of integral value calculation, and the calculated integral value is the same when the high-frequency component is extracted for the entire area of the image information and when the high-frequency component is obtained for a part of the area. Because it becomes.

  In the camera state detection system according to the present invention, the camera is preferably a camera used in an in-vehicle camera system including at least an in-vehicle monitor that displays image information obtained by photographing as a visible image.

  Detects the fault status of cameras used in various in-vehicle camera systems, such as a parking assistance system for monitoring the rear of the vehicle during parking, and a side monitor system for monitoring the side of the vehicle while traveling. Because it can.

  According to the camera state detection system according to the present invention, since it is possible to detect a failure of the camera, without directly observing and inspecting the camera itself or directly measuring and inspecting the installation state of the camera, It is possible to detect whether or not a failure has occurred in the camera.

  In addition, an image or video obtained by the camera is displayed on an in-vehicle monitor, and the display is visually observed, which makes it objective and quantitative compared to the case where the presence or absence of a camera failure is monitored sensuously and indirectly. It is possible to detect whether there is a failure.

  In addition, according to the camera state detection system in which the camera state detection unit detects the failure state by comparing the integrated value with a preset reference value to determine the presence or absence of the failure. The presence or absence of a camera failure can be easily detected by a simple process of simply comparing the integral value of the high-frequency component of an image or video obtained by shooting with a preset reference value.

  In addition, according to the camera state detection system in which the camera state detection unit also serves as at least one of the high-frequency component extraction unit and the integral value calculation unit, the number of components can be reduced, and the number of parts can be reduced. According to the reduction, the size can be reduced and the manufacturing cost can be reduced.

  In addition, when a camera having a fixed focal length optical system is mounted on a vehicle, it is not considered at all that the installation position of the camera with respect to a shooting target such as the ground changes. According to the camera state detection system provided with the above, the effects exhibited by the above-described camera state detection system of the present invention can be made more remarkable.

  The camera further includes a travel detection means for detecting the travel of the vehicle, wherein the camera state detection means repeatedly detects a failure of the camera in a state in which the travel of the vehicle is detected by the travel detection means. According to the state detection system, it is possible to detect a camera failure more accurately and more accurately than when the vehicle is stopped.

  On the other hand, since the camera failure is not detected while the vehicle is stopped, it is possible to prevent erroneous detection of the camera failure.

  Further, the image processing apparatus further includes a region selecting unit that selects a partial region of the entire region of the image information obtained by photographing with the camera, and the high frequency component extracting unit is configured to select only the partial region selected by the region selecting unit. According to the camera state detection system, the high-frequency component is extracted, or the integral value calculating means calculates the integral value only for the partial area selected by the area selecting means. High-frequency components can be extracted only for a part of the entire information area, so that the extraction process can be speeded up. Alternatively, the integral value can be calculated only for a part of the area. be able to.

  In addition, by designating a partial area having a relatively large contrast or tone difference (luminance difference or density difference) from the entire area of the image information obtained by the camera by the area selection unit, the entire area is selected. Even if it is not possible to obtain an appropriate determination result by comparison with the target integral value, an appropriate determination is possible depending on the comparison with the integral value for a part of the region where contrast or tone difference is relatively large. The result can be obtained.

  In addition, according to the camera state detection system, which is a camera used in an in-vehicle camera system that includes at least an in-vehicle monitor that displays image information obtained by photographing as a visible image, the It is possible to detect the failure of cameras used in various in-vehicle camera systems such as a parking assistance system for monitoring the rear and a side monitor system for monitoring the side of the vehicle while traveling.

  Embodiments of a camera state detection system according to the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a camera state detection system 100 according to an embodiment of the present invention.

  The illustrated camera state detection system 100 is mounted on the rear upper part of a vehicle 200 and has a fixed focal length optical system 10 that captures a scene (image) P around the vehicle 200 (for example, behind the vehicle). And a bandpass filter (BPF) 30 (high frequency) for extracting a high frequency component SH having a high spatial frequency from the image P obtained by photographing with the camera 10 (image information corresponding to the image P is S). Component extraction means), an integration circuit 40 (for example, a capacitor; integral value calculation means) for calculating the integral value ∫ (SH) of the high-frequency component SH, and the calculated integral value ∫ (SH) are compared with the reference value K. Thus, the camera 10 or the optical system provided with the camera 10 is configured to include a dirt determination circuit 50 (camera state detection means) that determines the presence or absence of dirt (failure).

  Here, the camera 10 is mounted as an element constituting a parking support system or the like mounted on the vehicle 200, for example. This parking support system or the like is an image taken by the camera 10 in addition to the camera 10. A monitor that displays a video as a visible image, a controller that calculates a guidance trajectory line, a movement prediction trajectory line, and the like for guiding the vehicle to be properly parked at a predetermined position during a parking operation .

  The camera state detection system 100 further includes an absolute value circuit (for example, a rectifier) 30 that outputs an absolute value high frequency component | SH | obtained by inverting the negative component of the high frequency component SH extracted by the BPF 30 to the positive side; An area selection unit 61 that receives an instruction to select a partial area range p from the image P and outputs a horizontal synchronization signal / vertical synchronization signal corresponding to the specified partial area range p, and area selection An area selection circuit 62 for selecting a partial area range p (image information corresponding to the partial area range p is s) based on the vertical synchronization signal / horizontal synchronization signal output from the unit 61; I have.

  In addition, the area selection part 61 may be comprised by the touchscreen sensor etc. which can select the area | region range p on the monitor which comprises the parking assistance system mentioned above, for example, the area selection part 61, the area selection circuit 62, Constitutes area selection means.

Thereby, the integrating circuit 40 integrates the absolute value high-frequency component | SH | output from the absolute value circuit 30 with respect to a partial region range p to calculate an integrated value ∫ _p | SH |.

Further, the reference value K to be compared by the dirt determination circuit 50 is set in advance for each image type (motif) in the region range p, and these preset reference values K are a plurality of reference values. Are stored in the memory 72 as values K1, K2,..., And the dirt determination circuit 50 selects from the integral value ∫ _p | SH | and a plurality of reference values K1, K2,. Any of the reference values Ki (i = 1, 2,...) Is compared to determine whether the optical system is dirty.

Accordingly, the dirt determination circuit 50 determines whether the camera 10 or the optical system provided in the camera 10 is dirty by comparing the calculated integral value ∫ _p | SH | with the selected reference value Ki. To do.

  Therefore, a motif selection unit 71 for selecting a motif of the image p is provided, and any reference value Ki (i = 1, 2,...) Corresponding to the selection instruction input to the motif selection unit 71 is stored in the memory. 72, and is input to the dirt determination circuit 50.

  Next, the operation of the camera state detection system 100 of this embodiment will be specifically described.

  First, a landscape (image) P on the road surface behind the vehicle is photographed by the camera 10. An example of the image P obtained by photographing at this time is shown in FIG. Note that this image P is an image captured by the camera 10 that is horizontally reversed. This is because the camera 10 faces rearward and the monitor displaying this image P is arranged so that the occupant facing forward faces the occupant who sees the image P displayed on the monitor. This is because the left / right reversing process is performed by the controller of the parking assistance system so as not to give a sense of incongruity in the left / right direction.

  In the image information S as a signal corresponding to the obtained image P, only a high frequency component SH having a predetermined spatial frequency or higher is extracted by the BPF 20.

  For example, when the image information S shown in FIG. 3A is scanned in the horizontal direction, the distribution of luminance values (signal values) in the image information S is as shown in FIG. The high-frequency component SH, which is a portion with a large difference in luminance (a portion where the luminance change is steep), is as shown in FIG.

  This high-frequency component SH becomes a negative change when changing from a large luminance value to a small luminance value, and becomes a positive change when changing from a small luminance value to a large luminance value. Therefore, when calculating the integral value later, Therefore, it is necessary to prevent mutual cancellation between the integral value in the positive change portion and the integral value in the negative change portion.

  Therefore, the absolute value circuit 30 performs an absolute value process for inverting the negative change portion to the positive side with respect to the output SH of the BPF 20, and outputs an absolute value high frequency component | SH | shown in FIG. .

  On the other hand, the occupant uses the touch panel sensor (area selection unit 61) installed on the display surface of the monitor that constitutes the above-described parking assistance system, for example, to display the entire image P (the entire captured image P) displayed on the monitor. A partial region range p (indicated by a two-dot chain line in FIG. 2) is selected.

  This selected region range p is a region range in the visible region that the occupant recognizes as containing the high-frequency component SH.

  The area selection unit 61 outputs a horizontal synchronization signal / vertical synchronization signal corresponding to the region range p set as described above, and the area selection circuit 62 to which the horizontal synchronization signal / vertical synchronization signal is input receives the image information S. The region range s corresponding to the region range p is specified above, and the specified region range s is input to the integration circuit.

The integration circuit 40 integrates the absolute value high frequency component | SH | with respect to the set region range s, and inputs the calculated integration value ∫ _p | SH |

The dirt determination circuit 50 compares the reference value K stored in the memory 72 with the inputted integral value ∫ _p | SH | to determine whether the main body of the camera 10 or the optical system included in the camera 10 is dirty. Although it is determined that there is no contamination of the main body of the camera 10 or the optical system provided in the camera 10, depending on the motif of the image P to be compared, the integral value of the high frequency component may be small in the first place.

  Therefore, for each motif of the image P, the reference value Ki to be compared is stored in the memory 72, and the occupant inputs the motif of the selected region range p (image information s) to the motif selection unit 71, The motif selection unit 71 inputs the selection of the input motif to the memory 72, and the memory 72 uses the reference value Ki (i = 1, 2,...) Corresponding to the selection of the input motif as the stain determination circuit 50. To enter.

Then, the dirt determination circuit 50 compares the input integral value _p | SH | with the reference value Ki (i = 1, 2,...), And compares the size with the optical system provided in the main body of the camera 10 or the camera 10. The presence or absence of dirt is judged. Specifically, the integral value ∫ _p | When the> reference value Ki, dirt camera 10 determines that there is no integral value ∫ _p | | SH SH | When ≦ reference value Ki is dirt camera 10 It is determined that it has occurred.

  The determination result is notified to the occupant, for example, by being displayed on the monitor of the parking assistance system.

  Thereby, the passenger | crew can detect the presence or absence of the stain | pollution | contamination of the optical system with which the main body of the camera 10 or the camera 10 was equipped. Therefore, it is possible to detect whether or not the camera 10 is dirty without directly observing and inspecting the camera 10 visually.

  Further, the image P obtained by the camera 10 is displayed on a monitor, and the display is visually observed, so that it is more objective and quantitative than the case where the presence or absence of dirt on the camera 10 is monitored sensorially and indirectly. It is possible to detect whether there is a failure.

  In addition, although the above-mentioned description is the description for the purpose of detecting the presence or absence of dirt of the main body of the camera 10 or the optical system included in the camera 10, the camera 10 or the optical system is deteriorated or damaged, or the camera Without directly measuring the installation state of the camera 10, the displacement of the installation position of the camera 10 with respect to the vehicle 200 on which the camera 10 is mounted (particularly the displacement of the distance along the optical axis direction) can be measured. It can also be detected.

  Further, as so-called self-diagnosis information, the date and time information and degree of occurrence of dirt, deterioration, damage, misalignment, etc. of the camera 10 may be stored in a predetermined memory. You may make it implement as an initial check.

Further, according to the camera state detection system 100 of the present embodiment, the integral value ∫ _p | SH | of the high frequency component | SH | of the image P obtained by photographing by the camera 10 is set as a reference value set in advance. The presence or absence of a failure (dirt or misalignment) of the camera 10 can be easily detected by a simple process of simply comparing the size with Ki.

  In the present embodiment, the occupant selects and sets a part of the region range p, but may be set in advance as a specific range in the image P.

  Further, the set area range p is not limited to a part of the entire area of the image P, and may be the entire area. That is, the integral value ∫ | SH | of the absolute value | SH | of the high-frequency component SH for the entire region of the image P may be compared with the corresponding reference value K.

Further, since the integration circuit 40 calculates the integration value ∫ _p | SH | only for a part of the region range p specified by the area selection circuit 62, this calculation process can be speeded up.

  The BPF 20 may also extract the high-frequency component SH only for a partial region range p specified by the area selection circuit 62. With such a configuration, the extraction process is speeded up. be able to.

  Of course, the absolute value circuit 30 may be subjected to the absolute value processing only for a part of the region range p specified by the area selection circuit 62. Processing can be speeded up.

Further, by designating, by the area selection circuit 62, a partial region range p having a relatively large contrast or tone difference (luminance difference or density difference) in the entire region of the image P, integration for the entire region is performed. Even if an appropriate determination result cannot be obtained by comparison with the value ∫ | SH |, the integrated value ∫ _p | SH | for a partial region range p having a relatively large contrast or tone difference. An appropriate determination result can be obtained depending on the comparison.

  The camera state detection system 100 according to the present embodiment determines whether or not there is a failure based on the image P taken in a state where the vehicle 200 is stopped. The form is not limited.

That is, when the vehicle 200 on which the camera state detection system 100 is mounted is parked with the camera 10 facing a wall or a wall having a substantially uniform luminance distribution, or when the vehicle 200 is parked on a road without a road surface sign. Since the high-frequency component SH of the image P is small, even if the camera 10 is not obstructed, the integral value ∫ _p | SH | (the same applies to the integral value ∫ | SH | for the entire region) is extremely small. There is a risk of erroneous determination of the presence or absence of a failure in the camera 10.

  Therefore, as shown in FIG. 4, the camera state detection system 100 further includes a vehicle speed sensor 80 (travel detection means) that detects the travel of the vehicle 200, and the dirt determination circuit 50 uses the vehicle speed sensor 80 to drive the vehicle 200. In the detected state, the presence / absence of the failure of the camera 10 is repeatedly detected, and in the state where the traveling of the vehicle 200 is not detected, the presence / absence of the failure of the camera 10 may not be detected.

  According to the camera state detection system 100 of the embodiment configured as described above, in a state where the vehicle 200 is traveling, the dirt determination circuit 50 is based on the vehicle speed signal input from the vehicle speed sensor 80. Can be detected.

  When the vehicle 200 is traveling, the image P taken by the camera 10 changes from moment to moment. Therefore, there cannot be a situation in which the obtained image P always contains only a small amount of the high-frequency component SH.

  For example, when the object to be photographed by the camera 10 includes a road surface that is running, there is always an opportunity to photograph a sign drawn on the road surface while traveling, and the image P including the road sign is displayed in the outline portion of the sign. A high frequency component SH having a high spatial frequency is included.

  Accordingly, when the vehicle 200 is traveling, that is, when traveling of the vehicle 200 is detected by the dirt determination circuit 50 by the vehicle speed sensor 80, the dirt determination circuit 50 repeatedly detects the failure state of the camera 10, thereby stopping the vehicle 200. It is possible to detect the presence or absence of a failure of the camera 10 with higher accuracy and accuracy than inside.

  On the other hand, when the vehicle 200 is stopped, that is, when the traveling speed of the vehicle 200 is not detected by the dirt determination circuit 50 by the vehicle speed sensor 80, the dirt determination circuit 50 does not detect the failure state of the camera 10. Can be prevented.

It is a block diagram which shows an example of the camera state detection system which concerns on embodiment of this invention. It is a figure which shows an example of the image (left-right reversal) image | photographed by the camera in the camera state detection system shown in FIG. It is a figure explaining the effect | action of the camera state detection system shown in FIG. 1, (a) is image information, (b) is luminance distribution, (c) is a high frequency component, (d) is the absolute value of a high frequency component. Each is shown. It is a block diagram which shows the modification of the camera state detection system shown in FIG.

Explanation of symbols

10 Camera 20 Band pass filter (High-frequency component extraction means)
30 Absolute value circuit (integral value calculation means)
40 Integration circuit (integral value calculation means)
50 Dirt determination circuit (camera state detection means)
61 Area selection part (area selection means)
62 Area selection circuit (area selection means)
71 Motif type selection unit 72 Memory 100 Camera state detection system 200 Vehicle

Claims (7)

A camera that is mounted on a vehicle and photographs the periphery of the vehicle;
High-frequency component extraction means for extracting high-frequency components from image information obtained by photographing with the camera;
An integral value calculating means for calculating an integral value of a part or all of the image information of the high frequency component extracted by the high frequency component extracting means;
A camera state detection system comprising camera state detection means for detecting a failure of the camera based on the integral value calculated by the integral value calculation means.   The camera state detection system according to claim 1, wherein the camera state detection unit determines whether or not there is a failure by comparing the integral value with a preset reference value.   The camera state detection system according to claim 1, wherein the camera state detection unit also serves as at least one of the high-frequency component extraction unit and the integral value calculation unit.   The camera state detection system according to claim 1, wherein the camera includes an optical system having a fixed focal length. It further comprises travel detection means for detecting the travel of the vehicle,
5. The camera state detection unit according to claim 1, wherein the camera state detection unit repeatedly detects a failure of the camera in a state in which the traveling detection unit detects the traveling of the vehicle. The camera state detection system according to item. An area selection means for selecting a partial area of the entire area of the image information obtained by photographing with the camera;
The high-frequency component extraction means extracts the high-frequency component only for the partial region selected by the region selection means,
Alternatively, the integral value calculating unit calculates the integral value only for the partial region selected by the region selecting unit. The camera state detection system described.   The said camera is a camera used for the vehicle-mounted camera system provided with at least the vehicle-mounted monitor which displays the image information obtained by image | photographing as a visible image, Any one of Claim 1 to 6 characterized by the above-mentioned. The camera state detection system according to item.

Images