Contamination determining apparatus, contamination determining method, and cleaning system

Abstract

To enable cleaning suitable for each kind of dirt.SOLUTION: A dirt discrimination device for discriminating a type of dirt present on a predetermined surface includes a light source configured to emit light of a plurality of wavelengths, a camera configured to capture an image of the dirt when the light of the plurality of wavelengths emitted from the light source is irradiated onto the dirt, and a controller connected to each of the light source and the camera and configured to discriminate the type of dirt based on the image captured by the camera and output a signal or data indicating one cleaning pattern among a plurality of cleaning patterns according to the type of dirt.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a dirt detection device, a dirt detection method, and a cleaning system. [Background technology]

[0002] Japanese Patent Application Laid-Open Publication No. 2016-009099 (Patent Document 1) describes an imaging device with a cleaning function that includes a wiper drive unit with a wiper blade that wipes the surface of the lens of the camera unit, a cleaning liquid discharge unit with a cleaning liquid discharge nozzle that discharges cleaning liquid onto the surface of the lens of the camera unit, and an image recognition unit that determines the state of dirt adhering to the surface of the lens of the camera unit, and is configured to be switchable between a cleaning liquid drip mode in which the operation of the cleaning liquid discharge unit is controlled, and a wiping mode in which the wiper drive unit is operated in addition to controlling the operation of the cleaning liquid discharge unit, depending on the state of dirt determined by the image recognition unit.

[0003] JP 2022-069805 A describes a sensor surface cleaning device having a first sensor, a second sensor, a first camera, a second camera, multiple cleaning devices that spray cleaning liquid toward each sensor surface, and a driving control ECU that operates according to either an operating mode of a first mode that allows automatic cleaning operation of each cleaning device or a second mode that does not allow automatic cleaning operation of each cleaning device, wherein the driving control ECU switches the operating mode to the second mode when at least one of a first specific condition indicating that the environment is such that the sensor surface is likely to become dirty and a second specific condition indicating that dirt on the sensor surface cannot be removed by cleaning liquid is met while the operating mode is in the first mode.

[0004] However, the above-mentioned conventional techniques have room for improvement in that it is difficult to perform cleaning according to the type of dirt. That is, when using each conventional technique, depending on the type of dirt, there is a possibility that cleaning liquid may be wasted or the dirt may not be completely removed, and in some cases the condition of the dirt may become worse. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-009099 [Patent Document 2] Japanese Patent Publication No. 2022-069805 Summary of the Invention [Problem to be solved by the invention]

[0006] One of the objectives of a specific embodiment of the present disclosure is to provide a technology that enables cleaning according to the type of dirt. [Means for solving the problem]

[0007] [1] A dirt determination device according to one aspect of the present disclosure includes: 1. A device for determining the type of dirt present on a predetermined surface, comprising: a light source configured to be able to emit light of a plurality of wavelengths; a camera that captures an image of the stain when the light beams of the plurality of wavelengths emitted from the light source are irradiated onto the stain; a controller connected to each of the light source and the camera, for determining the type of dirt based on the image captured by the camera, and for outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; The dirt detection device includes: [2] A method for determining dirt according to one aspect of the present disclosure includes: 1. A method for determining the type of contamination present on a surface, comprising: The controller Obtaining an image of the stain when irradiating the stain with light of a plurality of wavelengths; estimating the type of dirt based on the image, and outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; This is a method for determining dirt by performing the above. [3] A cleaning system according to one embodiment of the present disclosure includes: The dirt detection device according to claim 1; a cleaning device connected to the dirt determination device and operating in response to a signal or data indicating the cleaning pattern output from the dirt determination device; A cleaning system comprising:

[0008] According to the above configuration, a technique is obtained that enables cleaning according to the type of dirt. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing the configuration of a cleaning system according to one embodiment. [Figure 2] FIG. 2 is a schematic diagram for explaining the principle of determining dirt. [Figure 3] FIG. 3 is a schematic cross-sectional view illustrating an example of a configuration for guiding light to a camera lens. [Figure 4] FIG. 4 is a flowchart showing the operation procedure of the contamination determining device. DETAILED DESCRIPTION OF THE INVENTION

[0010] 1 is a block diagram showing the configuration of a cleaning system according to one embodiment. The cleaning system according to this embodiment includes a dirt determination device 1 that determines (identifies) the type of dirt adhering to the surface of an outer lens 3 that is provided to protect a camera 10, which is a sensing camera installed in a vehicle, and instructs a cleaning device 2 to select a cleaning mode according to the type of dirt, and the cleaning device 2 that cleans the dirt adhering to the outer lens 3 by performing an operation according to the cleaning mode instructed by the dirt determination device 1.

[0011] The dirt detection device 1 includes a camera 10, a light source 20, and a controller 30. As illustrated in the figure, the dirt detection device 1 may further include an outside air temperature sensor 40 and a raindrop sensor 41.

[0012] The camera 10 is a sensing camera for capturing images of the conditions around the vehicle. In this embodiment, the camera 10 is used to capture images of dirt adhering to the outer lens 3. The camera 10 includes a camera lens 11 and an image sensor 12. The camera lens 11 collects incident light and directs it to the image sensor 12. The image sensor 12 receives the incident light and converts it into an electrical signal to generate an image.

[0013] The light source 20 is configured to be able to emit light of multiple wavelengths. In this example, the light source 20 includes LEDs 21R, 21G, and 21B, which are semiconductor light-emitting elements. The LED 21R emits light in the red wavelength range. The LED 21G emits light in the green wavelength range. The LED 21B emits light in the blue wavelength range.

[0014] The red wavelength range is, for example, a wavelength range of 600 nm or more and less than 800 nm. The green wavelength range is, for example, a wavelength range of 500 nm or more and less than 600 nm. The blue wavelength range is, for example, a wavelength range of 400 nm or more and less than 500 nm. Note that these wavelength range classifications are merely examples and are not limiting.

[0015] The controller 30 controls the overall operation of the dirt determining device 1. The controller 30 can be realized by using, for example, a computer including a processor and a memory, and causing the processor to execute a predetermined program.

[0016] Specifically, the controller 30 controls the operation of the light source 20 to cause light of multiple wavelengths to be incident on the outer lens 3, and determines the type of dirt adhering to the outer lens 3 based on the image captured by the camera 10 at that time. Then, the controller 30 selects a cleaning mode according to the determined type of dirt and instructs the cleaning device 2 to select that cleaning mode.

[0017] The outside air temperature sensor 40 detects the outside air temperature around the outer lens 3 (that is, the outside air temperature around the vehicle). The raindrop sensor 41 detects the amount of rainfall around the outer lens 3 and the diameter of raindrops.

[0018] The cleaning device 2 operates to remove dirt adhering to the outer lens 3, and is composed of a cleaning nozzle 50, a cleaning liquid supply unit 51, an air supply unit 52, an electromagnetic valve 53, and a cleaning liquid tank 54.

[0019] The cleaning nozzle 50 is used to spray cleaning liquid and / or air onto the surface of the outer lens 3, and is positioned with its outlet facing the outer lens 3. The cleaning liquid may be sprayed alone, or may be sprayed simultaneously with air to form a mist. The cleaning liquid supply unit 51 is, for example, a water pump, which operates when necessary to generate water pressure and supply cleaning liquid from the cleaning liquid tank 54 to the cleaning nozzle 50. The air supply unit 52 is, for example, a compressor, which operates when necessary to supply air (compressed air) to the cleaning nozzle 50. The solenoid valve 53 is connected between the cleaning liquid supply unit 51 and the air supply unit 52 and the cleaning nozzle 50, and is used to switch between supplying cleaning liquid, air, or both to the cleaning nozzle 50. The cleaning liquid tank 54 contains cleaning liquid.

[0020] FIG. 2 is a schematic diagram illustrating the principle of dirt discrimination in this embodiment. In this embodiment, light emitted from the light source 20 is incident on the camera 10 from outside the angle of view (for example, the end of the outer lens 3). At this time, at least a portion of the incident light travels through the outer lens 3 by total internal reflection and is reflected by dirt 80 adhering to the outer lens 3. This reflected light is captured by the camera 10. Since the main components differ depending on the type of dirt 80, the type of dirt can be discriminated by utilizing the wavelength dependency of the reflectance of each main component. By locating the light source 20 outside the angle of view of the camera 10, light emitted from the light source 20 does not enter the camera 10 when dirt 80 is not present. In addition to guiding light to the outer lens 3, light can also be guided directly to the camera lens 11. An example of this is described below.

[0021] 3 is a schematic cross-sectional view illustrating an example of a configuration for guiding light to a camera lens. Camera lens 11 is supported by housing 60. Light emitted from light source 20 is made incident on one end of camera lens 11 (the right end in the illustrated example) via light-guiding member 61. Then, due to the difference in refractive index between the resin (e.g., approximately 1.3 to 1.6), which is an example of the material of camera lens 11, and the refractive index of the air outside camera lens 11 (approximately 1.0), at least a portion of the light that has entered camera lens 11 travels through camera lens 11 while being totally reflected.

[0022] Next, the wavelength dependency of the main component of the dirt 80 and the reflectance will be described in detail. When the main component of the dirt 80 is water, water absorbs a lot of light in the red wavelength range, so the reflectance of the dirt 80 is higher for green light and blue light than for red light. On the other hand, when the main component of the dirt 80 is mud (earth), mud contains iron (Fe) and therefore absorbs a lot of blue light, so the reflectance of the dirt 80 is higher for green light and red light than for blue light. Therefore, the type of dirt 80 (water, mud) can be determined based on the combination of the magnitude of the reflectance for red light, green light, and blue light.

[0023] Furthermore, if there is an air gap between the dirt 80 and the surface of the outer lens 3 (or camera lens 11, the same applies below), the dirt 80 is not tightly attached to the surface of the outer lens 3, and therefore the intensity of the light reflected by the dirt 80 is relatively low. In this case, it can be determined that the dirt 80 is not wet, i.e., is relatively dry soil. On the other hand, if there is no air gap between the dirt 80 and the surface of the outer lens 3, the dirt 80 is tightly attached to the surface of the outer lens 3, and therefore the intensity of the light reflected by the dirt 80 is relatively high. In this case, it can be determined that the dirt 80 is wet, i.e., is mud or water. In other words, the type of dirt 80 can be determined based on the intensity of the reflected light.

[0024] Figure 4 is a flowchart showing the operating procedure of the dirt detection device. Note that the order of the processes shown here can be changed as long as no contradictions or inconsistencies occur in the results of the information processing, and other processes not specifically shown here can also be added. The illustrated process is executed repeatedly, for example, at regular intervals. Here, we will explain how to clean dirt adhering to the outer lens 3, but the same applies to the camera lens 11.

[0025] When the vehicle starts (step S11; YES), the controller 30 determines whether or not dirt is attached to the outer lens 3 based on the image captured by the camera 10 (step S12).

[0026] The start of the vehicle can be identified based on the state of the ignition switch, for example. The presence or absence of dirt can be determined by, for example, preparing a reference image obtained by photographing an unsoiled outer lens 3, and calculating the difference between this reference image and an image of the outer lens 3 photographed after the vehicle is started.

[0027] If the vehicle has not started (step S11; NO) or if there is no dirt (step S12; NO), the process returns to step S11.

[0028] If dirt is present (step S12; YES), the controller 30 instructs the light source 20 to turn on (step S13). Specifically, the controller 30 controls the light source 20 so that the LEDs 21R, 21G, and 21B are turned on at different timings.

[0029] Images obtained by light (light with different wavelengths) emitted from LEDs 21R, 21G, and 21B being reflected by dirt are captured by camera 10, and controller 30 acquires these images from camera 10 (step S14).

[0030] The controller 30 analyzes each image (step S15). More specifically, as described above, the magnitude of the reflectance is calculated based on the image for each color (wavelength range) of light incident on the outer lens 3. Specifically, the controller 30 determines the reflectance by using, for example, several pixels included in the image and pixel values ​​(such as brightness values) for each color of light.

[0031] The controller 30 determines the type of dirt based on the combination of the determined magnitudes of reflectance (step S16). Specifically, as described above, if the reflectance with green light or blue light is higher than the reflectance with red light, the type of dirt is determined to be "water," and if the reflectance with green light or red light is higher than the reflectance with blue light, the type of dirt is determined to be "mud or earth."

[0032] Furthermore, if the reflectances of all of the red light, green light, and blue light are higher than a predetermined standard, the controller 30 determines that the outer lens 3 is wet (step S17; YES), and if the reflectances are equal to or lower than the predetermined standard, the controller 30 determines that the outer lens 3 is not wet (step S17; NO). Note that if the type of dirt is determined to be "water" in step S16, step S17 may be omitted and the process may proceed to step S18 described below.

[0033] If wet, the controller 30 instructs the cleaning device 2 to select either the first cleaning mode or the second cleaning mode depending on the type of dirt determined in step S16 (step S18). Specifically, if wet, the type of dirt is either "water" or "mud," so the controller 30 instructs the cleaning device 2 to select the first cleaning mode if the dirt is "water," and to select the second cleaning mode if the dirt is "mud." A preferred cleaning method for the first cleaning mode is to eject a cleaning liquid onto the outer lens 3, followed by spraying only air. A preferred cleaning method for the second cleaning mode is to simultaneously eject a cleaning liquid and air to generate a high-pressure mist, which is then sprayed to blow away the dirt, followed by spraying only air. The cleaning device 2 operates the cleaning liquid supply unit 51 and the air supply unit 52 depending on the cleaning mode instructed by the controller 30, thereby achieving the first or second cleaning mode.

[0034] On the other hand, if there is no wetness, the controller 30 instructs the cleaning device 2 to use the third cleaning mode (step S19). If there is no wetness, the type of dirt is "soil," so the cleaning method should preferably be to blow off the dirt by blowing air without using a cleaning liquid. This is because dripping, applying, or spraying a cleaning liquid can turn the soil into mud, which could actually worsen the dirt on the outer lens 3. Therefore, when the cleaning device 2 is instructed to use the third cleaning mode, it supplies air from the air supply unit 52 to the cleaning nozzle 50.

[0035] Whether the cleaning device 2 is operating in the first, second, or third cleaning mode, the amount and strength of the cleaning fluid and air may be variably set according to the level of the outside air temperature detected by the outside air temperature sensor 40, or according to the amount of rainfall detected by the raindrop sensor 41. For example, when the outside air temperature is low and there is a possibility that the cleaning fluid will freeze, control can be performed such that more cleaning fluid is used and less mist is used. Also, when rain is detected by the raindrop sensor, it is likely that the target area is already wet, so control can be performed such that less cleaning fluid is used.

[0036] According to the above-described embodiment, a technique is obtained that enables cleaning that is appropriate for each type of dirt.

[0037] It should be noted that the present disclosure is not limited to the contents of the above-described embodiment, and various modifications can be made within the scope of the gist of the present disclosure. For example, the above-described embodiment illustrates a cleaning system that identifies and cleans dirt adhering to a predetermined surface such as the outer lens 3 of the camera 10, but the predetermined surface to be cleaned is not limited to this.

[0038] The present disclosure has the following additional features. (Appendix 1) 1. A device for determining the type of dirt present on a predetermined surface, comprising: a light source configured to be able to emit light of a plurality of wavelengths; a camera that captures an image of the stain when the light beams of the plurality of wavelengths emitted from the light source are irradiated onto the stain; a controller connected to each of the light source and the camera, for determining the type of dirt based on the image captured by the camera, and for outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; A dirt detection device comprising: (Appendix 2) the controller selects different cleaning patterns depending on whether the reflectances of the stains obtained from the images corresponding to the light of the plurality of wavelengths are all higher than a predetermined standard or whether the reflectances of the stains are all equal to or lower than the predetermined standard. 2. The dirt detection device according to claim 1. (Appendix 3) the light source is configured to be able to emit light belonging to a first wavelength range, a second wavelength range having a longer wavelength than the first wavelength range, and a third wavelength range having a longer wavelength than the second wavelength range, each of which is a wavelength range of visible light; the controller selects different cleaning patterns for a case where the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range and the second wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the third wavelength range, and a case where the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range. 2. The dirt detection device according to claim 1. (Appendix 4) the light source is configured to be able to emit light belonging to a first wavelength range, a second wavelength range having a longer wavelength than the first wavelength range, and a third wavelength range having a longer wavelength than the second wavelength range, each of which is a wavelength range of visible light; The controller selecting a first cleaning pattern that is one of the plurality of cleaning patterns when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range and the second wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the third wavelength range, and when the reflectances of the dirt obtained from the images corresponding to the first wavelength range, the second wavelength range, and the third wavelength range are all higher than a predetermined standard; selecting a second cleaning pattern that is another one of the plurality of cleaning patterns when the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range, and when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range, the second wavelength range, and the third wavelength range is higher than a predetermined standard; selecting a third cleaning pattern, which is another one of the plurality of cleaning patterns, when the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range, and when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range, the second wavelength range, and the third wavelength range is equal to or lower than a predetermined standard; 2. The dirt detection device according to claim 1. (Appendix 5) the first cleaning pattern is a cleaning pattern that includes dropping a cleaning liquid onto the predetermined surface and then blowing air onto the predetermined surface, the second cleaning pattern is a cleaning pattern that includes simultaneously ejecting the cleaning liquid and the air to generate mist, spraying the mist onto the predetermined surface, and then spraying the air, the third cleaning pattern is a cleaning pattern that includes blowing the air onto the predetermined surface without supplying the cleaning liquid to the predetermined surface; 5. The dirt detection device according to claim 4. (Appendix 6) the first wavelength range is a wavelength range of 400 nm or more and less than 500 nm, the second wavelength range is a wavelength range of 500 nm or more and less than 600 nm, The third wavelength range is a wavelength range of 600 nm or more and less than 800 nm. 6. The dirt detection device according to claim 4 or 5. (Appendix 7) The predetermined surface is an outer surface of an outer lens for protecting the camera or a camera lens attached to the camera. 7. The dirt detection device according to any one of appendices 1 to 6. (Appendix 8) The light of the plurality of wavelengths emitted from the light source travels through the outer lens or the camera lens, is reflected by the dirt, and enters the camera. 8. The dirt detection device according to claim 7. (Appendix 9) 1. A method for determining the type of contamination present on a surface, comprising: The controller Obtaining an image of the stain when irradiating the stain with light of a plurality of wavelengths; estimating the type of dirt based on the image, and outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; To execute How to detect dirt. (Appendix 10) A dirt determination device according to any one of appendices 1 to 8; a cleaning device connected to the dirt determination device and operating in response to a signal or data indicating the cleaning pattern output from the dirt determination device; , a cleaning system. [Explanation of symbols]

[0039] 1: Dirt detection device, 2: Cleaning device, 3: Outer lens, 10: Camera, 11: Camera lens, 12: Image pickup element, 20: Light source, 21R, 21G, 21B: LED, 30: Controller, 40: Outside air temperature sensor, 41: Raindrop sensor, 50: Cleaning nozzle, 51: Cleaning liquid supply unit, 52: Air supply unit, 53: Solenoid valve, 80: Dirt

Claims

1. 1. A device for determining the type of dirt present on a predetermined surface, comprising: a light source configured to be able to emit light of a plurality of wavelengths; a camera that captures an image of the stain when the light beams of the plurality of wavelengths emitted from the light source are irradiated onto the stain; a controller connected to each of the light source and the camera, for determining the type of dirt based on the image captured by the camera, and for outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; A dirt detection device comprising:

2. the controller selects different cleaning patterns depending on whether the reflectances of the stains obtained from the images corresponding to the light of the plurality of wavelengths are all higher than a predetermined standard or whether the reflectances of the stains are all equal to or lower than the predetermined standard. The contamination detection device according to claim 1 .

3. the light source is configured to be able to emit light belonging to a first wavelength range, a second wavelength range having a longer wavelength than the first wavelength range, and a third wavelength range having a longer wavelength than the second wavelength range, each of which is a wavelength range of visible light; the controller selects different cleaning patterns for a case where the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range and the second wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the third wavelength range, and a case where the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range. The contamination detection device according to claim 1 .

4. the light source is configured to be able to emit light belonging to a first wavelength range, a second wavelength range having a longer wavelength than the first wavelength range, and a third wavelength range having a longer wavelength than the second wavelength range, each of which is a wavelength range of visible light; The controller selecting a first cleaning pattern that is one of the plurality of cleaning patterns when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range and the second wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the third wavelength range, and when the reflectances of the dirt obtained from the images corresponding to the first wavelength range, the second wavelength range, and the third wavelength range are all higher than a predetermined standard; selecting a second cleaning pattern that is another one of the plurality of cleaning patterns when the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range, and when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range, the second wavelength range, and the third wavelength range is higher than a predetermined standard; selecting a third cleaning pattern, which is another one of the plurality of cleaning patterns, when the reflectance of the dirt obtained from the image corresponding to the light in each of the second wavelength range and the third wavelength range is relatively higher than the reflectance of the dirt obtained from the image corresponding to the light in the first wavelength range, and when the reflectance of the dirt obtained from the image corresponding to the light in each of the first wavelength range, the second wavelength range, and the third wavelength range is equal to or lower than a predetermined standard; The contamination detection device according to claim 1 .

5. the first cleaning pattern is a cleaning pattern that includes dropping a cleaning liquid onto the predetermined surface and then blowing air onto the predetermined surface, the second cleaning pattern is a cleaning pattern that includes simultaneously ejecting the cleaning liquid and the air to generate mist, spraying the mist onto the predetermined surface, and then spraying the air, the third cleaning pattern is a cleaning pattern that includes blowing the air onto the predetermined surface without supplying the cleaning liquid to the predetermined surface; The dirt detection device according to claim 4.

6. the first wavelength range is a wavelength range of 400 nm or more and less than 500 nm, the second wavelength range is a wavelength range of 500 nm or more and less than 600 nm, The third wavelength range is a wavelength range of 600 nm or more and less than 800 nm. The dirt detection device according to claim 4.

7. The predetermined surface is an outer surface of an outer lens for protecting the camera or a camera lens attached to the camera. The contamination detection device according to claim 1 .

8. The light of the plurality of wavelengths emitted from the light source travels through the outer lens or the camera lens, is reflected by the dirt, and is incident on the camera. The dirt detection device according to claim 7.

9. 1. A method for determining the type of contamination present on a surface, comprising: The controller Obtaining an image of the stain when irradiating the stain with light of a plurality of wavelengths; estimating the type of dirt based on the image, and outputting a signal or data indicating one of a plurality of cleaning patterns according to the type of dirt; To execute How to detect dirt.

10. The dirt detection device according to claim 1 ; a cleaning device connected to the dirt determination device and operating in response to a signal or data indicating the cleaning pattern output from the dirt determination device; , a cleaning system.

Images