JP2020139794A - Dirt detection device and air conditioning system - Google Patents

Abstract

To provide a dirt detection device capable of distinguishing between a plurality of types of dirt.SOLUTION: A dirt detection device for detecting dirt on a target object, a device with a water storage container, is provided, the dirt detection device comprising a control unit configured to process a captured image (40) of the target object. The control unit is configured to perform a process for identifying the dirt on the target object at least either as adhesion dirt or turbidity based on the captured image (40).SELECTED DRAWING: Figure 7

Description

The present disclosure relates to a dirt detector and an air conditioning system for detecting dirt.

As a dirt detection device, the technique described in Patent Document 1 is known. The technique of Patent Document 1 acquires an image of a filter, performs binarization processing on each pixel constituting the image, divides the binarized image into predetermined blocks, and uses data related to the blocks. Find the diagnostic value.

Japanese Unexamined Patent Publication No. 2005-29066

By the way, there are various forms of dirt on the object, not only adhesion. In order to properly manage the object, it is preferable to manage not only adhered stains but also other stains. An object of the present disclosure is to provide a stain detection device capable of discriminating a plurality of types of stains.

The dirt detection device that solves this problem is a dirt detection device that detects dirt on the object by using a device including a container for storing water as the object, and a control unit that processes a photographed image of the object. In addition, the control unit performs a process of discriminating at least adhering stains and turbidity with respect to the stains on the object based on the captured image. According to this configuration, adhered dirt and turbidity can be discriminated.

In the stain detection device, the control unit identifies a discolored region different from the reference color of the object with respect to the plurality of captured images having different times, and among the plurality of captured images, floating objects in the discolored region. The floating object image area corresponding to the movement of the object is specified as turbidity.

As the turbidity flows, suspended matter (eg, suspended matter or granules) moves over time. The region where the suspended matter image is detected in the discolored region can be identified as a turbid region. Since the above configuration identifies turbidity based on such technical matters, turbidity can be detected accurately.

In the stain detection device, the control unit identifies a discolored region different from the reference color of the object in the captured image, specifies a predetermined hue region within the discolored region, and determines the size of the predetermined hue region. When it is larger than the value, it is specified as the adhered stain, and the region other than the region specified as the adhered stain in the discolored region is specified as the turbidity.

Since the adhered stain has a visually recognizable size, the predetermined hue region can be identified as the adhered stain when the size of the predetermined hue region having the predetermined hue is larger than the predetermined value in the captured image. In the discolored area, areas other than the adhered dirt portion can be identified as turbidity. In the above configuration, since the adhered dirt and turbidity are specified based on such technical matters, the adhered dirt and turbidity can be accurately detected.

In the stain detection device, the control unit discriminates between the adhered stain and the turbidity based on the hue of each pixel in the captured image.
The turbidity is caused by suspended matter floating in water and is detected as a reddish color. Adhering stains are caused by mold or the like adhering to the wall, and are detected as a greenish color. Therefore, it is possible to distinguish between adherent stains and turbidity by hue. According to the above configuration, since this fact is used, it is possible to easily distinguish between adhered dirt and turbidity.

In the stain detection device, the control unit holds a first threshold value for determining the progress of the adhered stain and a second threshold value larger than the first threshold value, and the adhered stain is found in the captured image. The area of the specified range is calculated as the progress of the adhered dirt, and when the progress of the adhered dirt exceeds the first threshold value, the external device is notified, and the area of the range determined to be turbid in the captured image is turbid. It is calculated as the degree of progress, and when the degree of progress of the turbidity exceeds the second threshold value, the external device is notified.

Since the turbidity is caused by suspended matter, it has less influence on the trouble of the object than the adhered dirt. In the above configuration, the second threshold value for the progress of turbidity is larger than the first threshold value for the progress of adherent stains. In this way, by making the execution condition of the treatment for the adhered dirt different from the execution condition of the treatment for the turbidity, the external device is notified that the adhered dirt or the turbidity exceeds the threshold value at the timing according to the adhered dirt or the turbidity. it can.

In the stain detection device, the control unit holds a third threshold value for determining the progress of the adhered stains and a fourth threshold value larger than the third threshold value, and the attached stains in the captured image. The area of the determined range is calculated as the progress of the adhered dirt, and when the progress of the adhered dirt exceeds the third threshold value, cleaning is performed, and the area of the range determined to be turbid in the captured image is the progress of turbidity. When the degree of progress of the turbidity exceeds the fourth threshold value, washing is performed.

Since the turbidity is caused by suspended matter, it has less influence on the trouble of the object than the adhered dirt. In the above configuration, the fourth threshold value for the progress of turbidity is larger than the third threshold value for the progress of adherent stains. In this way, by making the execution conditions of the treatment for the adhered stains different from the execution conditions of the treatment for the turbidity, it is possible to wash the object at the timing according to the adhered stains and the turbidity.

In the dirt detection device, the object is a drain pan of an air conditioner.
According to this configuration, it is possible to discriminate adhered dirt and turbidity with respect to the drain pan as an object.

An air conditioning system that solves this problem includes any of the above dirt detectors. According to this configuration, adhered dirt and turbidity can be discriminated.

Front view of the air conditioner. Top view of the internal structure of the indoor unit. Side view of the internal structure of the indoor unit. The figure which shows the range of adhesion dirt and turbidity in a color wheel. The figure which shows the photographed image. The figure which shows the mask. The figure of the photographed image in which the mask is overlapped. The figure which shows the matching of the template photographed image and the photographed image. The figure which shows the relationship between a template mask and a mask. Diagram of captured image converted to emphasized color. A graph showing changes in dirt.

Hereinafter, the dirt detection device 1 according to the present embodiment will be described.
The dirt detection device 1 detects a plurality of dirt on the object 2. In the present embodiment, the dirt detection device 1 detects adhered dirt and turbidity of the object 2. In the following, when the adhered dirt and the turbidity are not distinguished, both are collectively referred to as "dirt". Adhering dirt indicates dirt adhering to an object. The turbidity indicates the turbidity of the water in the container of the object.

A suitable example of the object 2 for dirt detection is a container for storing water. For example, the object 2 for dirt detection is the drain pan 26 of the indoor unit 21 of the air conditioner 20, the drain pan of the outdoor unit, and the dehumidifying / humidifying water tank. In the present embodiment, the dirt detection device 1 detects dirt on the drain pan 26 of the indoor unit 21 of the air conditioner 20. In this embodiment, the object 2 is a drain pan 26.

The dirt detection device 1 includes a control unit 10.
The control unit 10 processes the captured image of the object 2. The control unit 10 determines the adhered dirt and turbidity of the object based on the captured image.

The control unit 10 includes an acquisition unit 11 for acquiring a captured image 40 of the object 2 and a calculation unit 12. Based on the captured image 40, the calculation unit 12 performs a process of discriminating the dirt of the object 2 into at least adhering dirt and turbidity. The acquisition unit 11 is connected to the camera 30. The acquisition unit 11 and the calculation unit 12 may be housed in a case and packaged as one device, and as described below, the acquisition unit 11 and the calculation unit 12 are connected to the network N. , Each installation location may be dispersed. The components of the control unit 10 according to the present embodiment are distributed over the network N.

The acquisition unit 11 is provided on the air conditioner 20 itself or around the air conditioner 20.
The acquisition unit 11 acquires the captured image 40 sent from the camera 30 and stores the captured image 40. The acquisition unit 11 holds the identification information of the object 2 or the device including the object 2. In the present embodiment, the acquisition unit 11 holds the identification information of the air conditioner 20. Preferably, the acquisition unit 11 holds the object 2 or the position information of the device including the object 2. The location information includes the location (latitude and longitude, or address) of the object 2. In the present embodiment, the acquisition unit 11 holds the position information of the air conditioner 20 including the drain pan 26. More preferably, the acquisition unit 11 holds usage information of the device including the object 2. The usage information is information on the usage of the room in which the object 2 is installed, and includes, for example, the type of store. The acquisition unit 11 acquires identification information, position information, and usage information by an input operation. The acquisition unit 11 is connected to the communication unit 13. The communication unit 13 may be a component of the control unit 10.

The communication unit 13 controls communication between the acquisition unit 11 and the calculation unit 12. The communication unit 13 outputs the captured image 40 held in the acquisition unit 11 to the calculation unit 12 based on the internal command and the external command. Preferably, the communication unit 13 outputs at least one of the identification information, the position information, and the usage information held in the acquisition unit 11 to the calculation unit 12 based on the internal command and the external command. The internal command is a command formed at a preset time. For example, the internal command is formed by the internal circuit of the communication unit 13 when a predetermined condition (for example, the reception sensitivity is higher than a predetermined value in radio) is satisfied, or is formed periodically. The external command is a command based on a request from the calculation unit 12 of the cloud server 15. The communication unit 13 and the acquisition unit 11 may be housed in one case.

The installation location of the calculation unit 12 is not limited as long as information can be obtained from the acquisition unit 11. For example, the calculation unit 12 is provided in the cloud server 15 connected to the network N.

As shown in FIG. 1, the dirt detection device 1 may be a component of the air conditioning system S. For example, the air conditioner system S includes a dirt detection device 1 and an air conditioner 20. The air conditioner 20 is connected to the network N via the communication unit 13 of the control unit 10 of the dirt detection device 1.

The air conditioner 20 will be described with reference to FIGS. 2 and 3. FIG. 2 is a plan view of the internal structure of the indoor unit 21 as viewed by removing the upper wall of the indoor unit 21 of the air conditioner 20. FIG. 3 is a side view of the internal structure of the indoor unit 21 as viewed by removing the side wall of the indoor unit 21 of the air conditioner 20.

The model of the air conditioner 20 subject to the dirt inspection is not limited. For example, the ceiling-embedded air conditioner 20 is subject to a dirt inspection. The indoor unit 21 of the ceiling-embedded air conditioner 20 requires time and effort for internal inspection. Therefore, camera monitoring of the indoor unit 21 of the ceiling-embedded air conditioner 20 contributes to improvement of maintenance work efficiency. Therefore, in the present embodiment, the indoor unit 21 of the air conditioner 20 which is a ceiling-embedded type and is connected to the duct behind the ceiling will be described. The indoor unit 21 is connected to the outdoor unit via a refrigerant pipe. The indoor unit 21 is installed behind the ceiling.

As shown in FIGS. 2 and 3, the indoor unit 21 is arranged under the air conditioning control unit 22, the heat exchanger 23, the fan 24, the fan motor 25 for rotating the fan 24, and the heat exchanger 23. The drain pan 26 is provided, a drain pump 27 for discharging the water in the drain pan 26, and a case 28 having a suction port 28a and an outlet 28b. The color of the wall of the drain pan 26 is preferably a color in which stains are conspicuous. The color of the wall of the drain pan 26 is preferably white or a color close to white.

The case 28 is provided with an inspection lid 28c for inspecting the inside of the case 28. The inspection lid 28c is provided near the drain pan 26 and the drain pump 27. The camera 30 is attached to the inside of the inspection lid 28c. The camera 30 is attached to the inspection lid 28c so that at least a part of the bottom of the drain pan 26 is photographed.

The camera 30 includes a photographing unit 31 and a photographing control unit 32 (see FIG. 1). The photographing unit 31 photographs a part of the drain pan 26 at a timing controlled by the photographing control unit 32 to form a captured image 40.

In one example, the photographing unit 31 photographs the drain pan 26 under the condition of a still water surface. The photographing control unit 32 determines whether or not the condition of the still water surface state is satisfied. The still water surface state indicates a state in which the water surface in the drain pan 26 does not move. The imaging control unit 32 determines whether or not the surface is still water, among the changes in the captured image 40 obtained by the operation of the drain pump 27, the rotation of the fan 24, and the comparison of the plurality of continuously captured images 40. , At least one. When the condition of the still water surface state is satisfied, the photographing control unit 32 issues an imaging instruction to the photographing unit 31. For example, the imaging control unit 32 determines the condition of the still water surface state when detecting the adhered dirt.

In another example, the photographing unit 31 photographs the drain pan 26 under the condition of running water. The photographing control unit 32 determines whether or not the condition of the running water state is satisfied. The running water state indicates a state in which the water in the drain pan 26 moves. The photographing control unit 32 determines whether or not the water is flowing based on at least one of the operation of the drain pump 27 and the change in the captured image 40 obtained by comparing the plurality of captured images 40 continuously captured. To do. The shooting control unit 32 issues a shooting instruction to the shooting unit 31 when the condition of running water is satisfied. For example, the photographing control unit 32 determines the condition of the running water state when detecting turbidity as dirt.

The shooting control unit 32 controls the shooting timing as described above. Further, the photographing control unit 32 transmits the photographed image 40 formed by the photographing unit 31 to the acquisition unit 11. The photographing control unit 32 transmits the photographed image 40 to the acquisition unit 11 based on the internal command. The internal command is a preset command.

The captured image 40 is transmitted to the calculation unit 12 as follows. The captured image 40 formed by the photographing unit 31 of the camera 30 is output to the acquisition unit 11 and stored in the storage unit 11a of the acquisition unit 11. The captured image 40 stored in the acquisition unit 11 is transmitted to the calculation unit 12 of the cloud server 15 via the network N by an internal command or an external command of the communication unit 13.

The calculation unit 12 of the control unit 10 will be described with reference to FIG. The calculation unit 12 quantifies the dirt of the object 2 for each type at the command of the user or a predetermined timing. Specifically, the calculation unit 12 discriminates between adhered dirt and turbidity based on either one of the following (A) and (B) in the target range 40a described later.

(A) The calculation unit 12 discriminates between adhering dirt and turbidity based on the change in the captured image. The suspended matter is, for example, dust, dusty matter, suspended matter containing bacteria, etc., and may be translucent or opaque. When there is a stream of water in the drain pan 26, the suspended matter moves with the water. On the other hand, the adhered dirt adhering to the drain pan 26 does not move even if there is a flow of water. In the discolored region, the portion where the image is changed can be identified as turbidity, and the portion where the image is not changed can be identified as adherent stain. In this way, it is possible to discriminate between adhering stains and turbidity based on the image change.

In one example, the calculation unit 12 of the control unit 10 identifies a discolored region different from the reference color of the drain pan 26 in the target range 40a for a plurality of captured images 40 having different times. Specifically, the calculation unit 12 holds the reference color (for example, white) of the drain pan 26. The calculation unit 12 specifies a region other than the reference color as a discoloration region for the target range 40a of the captured image 40. Next, the calculation unit 12 identifies the floating matter image region corresponding to the movement of the floating matter in the discolored region as turbidity among the plurality of captured images 40. For example, the calculation unit 12 detects a change in the hue of the pixels constituting the discolored region, and identifies the region having the change in hue as a floating object image region. In the discolored region, the calculation unit 12 identifies the suspended matter image region as a turbid region, and in the discolored region, identifies the suspended matter image region as an adherent stain region.

(B) The calculation unit 12 discriminates between adhering stains and turbidity based on the hue of the captured image. Here, the dirt on the drain pan 26 will be described. Analyzing the hue of the stain on the drain pan 26, the stain has a reddish yellow-green color (hue 10-30) or green (hue 30-60). The hue of the stain is in the range of 10 or more and 60 or less. The adhered stain is green or a color around green, and the hue of the adhered stain is in the range of 30 or more and 60 or less. The turbidity is yellowish green close to red, and the hue of the turbidity is in the range of 10 or more and less than 30. The hue of the adhered stains and turbidity of the drain pan 26 is different from the hue of the wall of the drain pan 26. Therefore, dirt on the drain pan 26 can be detected based on the hue.

In one example, the calculation unit 12 of the control unit 10 identifies a discolored region different from the reference color of the drain pan 26 in the captured image 40. The identification of the discolored region is as described above. Next, the calculation unit 12 specifies a predetermined hue region within the discoloration region. Specifically, the calculation unit 12 specifies a region having a hue of 30 or more and 60 or less in the discoloration region. The calculation unit 12 determines whether or not the size of the predetermined hue region thus specified is larger than the predetermined value, and when the size of the predetermined hue region is larger than the predetermined value, the predetermined hue region is attached. Identify as a dirty area. Here, the predetermined value indicates the minimum area of adhered dirt. Even if the region has the same hue as the adhered stain, the region smaller than the predetermined value may be suspended. Therefore, in the present embodiment, the region of the predetermined hue smaller than the predetermined value is adhered. Not classified as dirty. The calculation unit 12 identifies a region other than the adhered stain in the discolored region as a turbid region.

The hue of the captured image 40 is obtained as follows. The calculation unit 12 calculates the hue for each pixel constituting the captured image 40 of the drain pan 26. When the captured image 40 is an image formed in the RGB format, the calculation unit 12 converts the captured image 40 based on the conversion formula from the RGB format to the HSV format, and determines the hue (H) value for each pixel. obtain.

In order to accurately detect the dirt on the drain pan 26, it is preferable that the target range 40a for which the dirt is detected is set in the captured image 40. The captured image 40 may include a part of the heat exchanger 23 and a part of the drain pump 27. In this case, the area where the heat exchanger 23 and the drain pump 27 are removed from the captured image 40 is set as the target range 40a for detecting dirt. The target range 40a is preset. The calculation unit 12 detects dirt in the target range 40a.

An example of setting the target range 40a will be described with reference to FIGS. 5 to 9. FIG. 5 is a diagram showing a captured image 40. FIG. 6 is a diagram showing a mask 41. FIG. 7 is a view of the captured image 40 on which the mask 41 is superimposed. In FIG. 7, the dark dot region indicates the adherent dirt region, and the light dot region indicates the turbid region. In the captured image 40 of FIG. 7, it is not possible to clearly distinguish between the area of adhered stains and the area of thin dots visually. FIG. 8 is a diagram showing matching between the template captured image 43 and the captured image 40. FIG. 9 is a diagram showing the relationship between the template mask 42 and the mask 41.

The calculation unit 12 holds a mask 41 for overlaying the captured image 40. The area other than the target range 40a of the photographed image 40 on the mask 41 is black without hue. In the mask 41, the inside of the target range 40a of the captured image 40 is transparent. In the captured image 40 on which the mask 41 is superimposed, the portion other than the target range 40a becomes black. Since the hues of black cannot be associated with each other, the area of the black portion is 0 when the hue area is calculated with respect to the captured image 40. Therefore, when the calculation unit 12 calculates the area of each hue in the entire captured image 40 on which the mask 41 is superimposed, as a result, the area of each hue within the target range 40a of the captured image 40 is calculated. By using the mask 41 in this way, it becomes easy to calculate the area of each hue within the target range 40a of the captured image 40.

The calculation unit 12 holds a template captured image 43 prepared in advance for each model of the air conditioner 20 and a template mask 42 prepared in advance for each model of the air conditioner 20. Even if the model of the air conditioner 20 is the same, the position of the drain pan 26 in the captured image 40 differs depending on the mounting variation of the camera 30. Therefore, it is preferable to use a mask 41 matched to the drain pan 26 of each air conditioner 20 in order to accurately detect the stain on the drain pan 26 that is the target of dirt detection. For example, the calculation unit 12 matches the feature points of the template captured image 43 with the feature points of the captured image 40 to be detected as stains (see FIG. 8), and forms a projective transformation matrix based on the matching result. To do. The calculation unit 12 forms the mask 41 by converting the template mask 42 by the formed projective transformation matrix (see FIG. 9).

The formation of the captured image 40 for display will be described with reference to FIG.
When determining the necessity of maintenance, the captured image 40 of the drain pan 26 may be used. However, the shadow portion of the captured image 40 is blackish and it is difficult to grasp the presence of dirt. For this reason, the captured image 40 is image-processed so that stains can be easily visually recognized. An example of image processing of the captured image 40 is shown below.

As shown in FIG. 10, in the target range 40a of the photographed image 40 among the photographed images 40 converted into the HSV format, the calculation unit 12 sets the region specified as the adhered stain and the region specified as the turbidity. , Converted to an emphasized color that is easy to recognize to form an enhanced image 44. In one example, the region RA identified as adherent stains is blue, the region RB identified as turbidity is light blue, and the region RC other than adherent stains and turbidity is red in the target range 40a. When displaying the captured image 40 of the drain pan 26 on the display device, the calculation unit 12 displays the enhanced image 44 converted to the enhanced color together with the raw captured image 40 or in place of the raw captured image 40. ..

Preferably, the calculation unit 12 calculates the degree of dirt progress as follows. In one example, dirt progression is assessed by enlargement of the dirty area.
(A) In the first example, the calculation unit 12 calculates the degree of dirt progress based on the area of the suspended matter image region detected based on the plurality of captured images 40 at different times. Specifically, the calculation unit 12 forms a mask 41 suitable for the captured image 40 based on the template mask 42 as described above. The calculation unit 12 identifies the discolored area and the suspended matter image area in the target range 40a based on the plurality of captured images 40 on which the masks 41 are overlapped. The identification of the discolored area and the suspended matter image area is as described above. The calculation unit 12 specifies a region obtained by removing the suspended matter image region from the discolored region as adhered stains, and identifies the suspended matter image region as turbidity. When calculating the area of the adhered dirt, the calculation unit 12 obtains the area of the adhered dirt by counting the number of pixels in the region specified as the adhered dirt. When calculating the area of turbidity, the calculation unit 12 obtains the area of turbidity by counting the number of pixels in the floating matter image region specified as turbidity. Further, the calculation unit 12 may calculate the total of the area of adhered dirt and the area of turbidity as the area of dirt. The calculation unit 12 outputs the area of dirt as the progress of dirt. The calculation unit 12 may output the area of the adhered dirt as the "progress of the adhered dirt", or the calculation unit 12 may output the area of the turbidity as the "progress of the turbidity".

(B) In the second example, the calculation unit 12 calculates the progress of the adhered stains based on the area of the region within the predetermined hue range in the target range 40a of the captured image 40, and determines the progress of the adhered stains. The progress of turbidity is calculated based on this. Specifically, the calculation unit 12 forms a mask 41 suitable for the captured image 40 based on the template mask 42 as described above. The calculation unit 12 identifies the discolored region in the target range 40a based on the plurality of captured images 40 on which the masks 41 are superimposed. The identification of the discolored region is as described above. The calculation unit 12 specifies a region having a hue of 30 or more and 60 or less as a region within a predetermined hue range in the discoloration region. Since the regions within the predetermined hue range are often dispersed in the captured image 40, the calculation unit 12 obtains an area for each region specified as a region having a hue of 30 or more and 60 or less. Areas larger than a predetermined value are identified as individual adherent dirt areas. Then, the calculation unit 12 calculates the total area of the areas of the attached dirt as the area of the area of the attached dirt. When calculating the area of each adhered stain, the calculation unit 12 obtains the area of the adhered stain by counting the number of pixels having a hue of 30 or more and 60 or less. The calculation unit 12 identifies a region excluding the region of adhered stains (all regions of individual adhered stains) from the discolored region as a turbid region, and obtains the area of the turbid region. Further, the calculation unit 12 may calculate the total of the area of adhered dirt and the area of turbidity as the area of dirt. The calculation unit 12 outputs the area of dirt as the progress of dirt. The calculation unit 12 may output the area of the adhered dirt as the "progress of the adhered dirt", or the calculation unit 12 may output the area of the turbidity as the "progress of the turbidity".

(C) In the third example, the calculation unit 12 calculates the degree of progress of adherent stains and turbidity based on the area of a region within a predetermined hue range in the target range 40a of the captured image 40. Specifically, the calculation unit 12 forms a mask 41 suitable for the captured image 40 based on the template mask 42 as described above. The calculation unit 12 calculates the area of the region within the predetermined hue range in the target range 40a based on the captured image 40 on which the mask 41 is superimposed. When calculating the area of adhered stains, the calculation unit 12 obtains the area of adhered stains by counting the number of pixels having a hue of 30 or more and 60 or less with respect to the captured image 40 on which the mask 41 is superimposed. .. When calculating the turbidity area, the calculation unit 12 obtains the turbidity area by counting the number of pixels having a hue of 10 or more and less than 30 with respect to the captured image 40 on which the mask 41 is superimposed. Further, the calculation unit 12 may calculate the total of the area of adhered dirt and the area of turbidity as the area of dirt. The calculation unit 12 outputs the area of dirt as the progress of dirt. The calculation unit 12 may output the area of the adhered dirt as the "progress of the adhered dirt", or the calculation unit 12 may output the area of the turbidity as the "progress of the turbidity".

With reference to FIG. 11, changes in the adhered stains on the drain pan 26 will be described.
Adhering stains on the drain pan 26 increase with the passage of time. During the cooling period, the water accumulated in the drain pan 26 is discharged by the drain pump 27. At that time, dirt is also discharged, but some dirt is not discharged and accumulates, and the dirt on the drain pan 26 gradually progresses. During the period when cooling is not performed, the drain pump 27 may not operate due to a small amount of water accumulated in the drain pan 26, so that the drain pan 26 is gradually contaminated. In this way, the dirt on the drain pan 26 increases year by year. Since the turbidity of the drain pan 26 changes depending on the flow of water, there is no periodic change. Since it is not easy to inspect the inside of the drain pan 26, it is preferable to notify or clean the drain pan 26 as follows regarding dirt and turbidity.

The calculation unit 12 holds a first threshold value for determining the progress of adhering dirt and a second threshold value for determining the progress of turbidity. The second threshold is larger than the first threshold. The calculation unit 12 calculates the area of the range specified as the adhered stain in the captured image 40 as the progress of the adhered stain. The calculation unit 12 notifies the external device 16 when the progress of the adhered dirt exceeds the first threshold value. The content of the notification is that the progress of the adhered dirt exceeds the first threshold value, and preferably includes a notification of the necessity of maintenance. The calculation unit 12 calculates the area of the range determined to be turbid in the captured image 40 as the progress of turbidity. The calculation unit 12 notifies the external device 16 when the progress of turbidity exceeds the second threshold value. The content of the notification is that the progress of turbidity exceeds the second threshold value, and preferably includes notification of the need for maintenance.

Further, preferably, the calculation unit 12 holds a third threshold value for determining the progress of adhering dirt and a fourth threshold value for determining the progress of turbidity. The fourth threshold is larger than the third value. The calculation unit 12 calculates the area of the range determined as the adhered stain in the captured image 40 as the progress of the adhered stain. The calculation unit 12 performs cleaning when the progress of adhering dirt exceeds the third threshold value. In one example of cleaning, water is allowed to flow through the drain pan 26 by cooling the heat exchanger 23 so that the heat exchanger 23 condenses. The calculation unit 12 calculates the area of the range determined to be turbid in the captured image 40 as the progress of turbidity. The calculation unit 12 performs cleaning when the progress of turbidity exceeds the fourth threshold value. In one example of cleaning, water is allowed to flow through the drain pan 26 by cooling the heat exchanger 23 so that the heat exchanger 23 condenses.

Preferably, the calculation unit 12 notifies the external device 16 of the information of the object 2 to be inspected for dirt together with the notification items. In the present embodiment, the calculation unit 12 notifies the external device 16 of the content related to dirt and the identification information, position information, and application information of the air conditioner 20. By receiving the identification information, the location information, and the usage information, the user can grasp the model, installation location, and usage of the air conditioner 20, so that the maintenance method can be examined in advance.

The external device 16 is, for example, a user terminal that can be connected to the network N. User terminals include mobile phones, notebook-type personal computers, personal computers, and tablet-type personal computers.

The operation of this embodiment will be described.
Adhering dirt may be peeled off from the wall, which causes clogging of the drain pump 27. Water turbidity may hinder the operation of the drain pump 27. Adhering dirt and turbidity of water are both causes of failure. Since the adhered dirt is a solid substance, the adhered dirt is likely to cause a failure. Since water turbidity is caused by a small substance floating in water, turbidity is unlikely to cause a failure. In this way, adherent dirt and turbidity have different effects on the device. In the present technology, the dirt detection device 1 discriminates between adhered dirt and turbidity. Further, the dirt detection device 1 provides information to the external device 16 used by the user for each of the adhered dirt and the turbidity. As a result, it is possible to alert the user about the necessity of maintenance of the object 2 at an appropriate timing.

The effect of this embodiment will be described.
(1) The control unit 10 of the dirt detection device 1 performs a process of discriminating at least the attached dirt and the turbidity of the dirt of the object 2 based on the captured image 40. According to this configuration, adhered dirt and turbidity can be discriminated.

(2) The control unit 10 identifies a discolored region different from the reference color of the object 2 for a plurality of captured images 40 having different times, and corresponds to the movement of floating objects in the discolored region among the plurality of captured images 40. The floating matter image area is specified as turbidity.

As the turbidity flows, suspended matter (eg, suspended matter or granules) moves over time. The region where the suspended matter image is detected in the discolored region can be identified as a turbid region. Since the above configuration identifies turbidity based on such technical matters, turbidity can be detected accurately.

(3) In one example, the control unit 10 specifies a discolored region different from the reference color of the object 2 in the captured image 40, specifies a predetermined hue region within the discolored region, and the size of the predetermined hue region is greater than a predetermined value. When it is also large, it is specified as the attached stain, and the region other than the region specified as the attached stain in the discolored region is specified as turbidity.

Since the adhered stain has a visually recognizable size, the predetermined hue region can be identified as the adhered stain when the size of the predetermined hue region having the predetermined hue is larger than the predetermined value in the captured image. In the discolored area, areas other than the adhered dirt portion can be identified as turbidity. In the above configuration, since the adhered dirt and turbidity are specified based on such technical matters, the adhered dirt and turbidity can be accurately detected.

(4) In another example, the control unit 10 discriminates between adhering stains and turbidity based on the hue of each pixel in the captured image 40.
The turbidity is caused by suspended matter floating in water and is detected as a reddish color. Adhering stains are caused by mold or the like adhering to the wall, and are detected as a greenish color. Therefore, it is possible to distinguish between adherent stains and turbidity by hue. According to the above configuration, since this fact is used, it is possible to easily distinguish between adhered dirt and turbidity.

(5) Preferably, the control unit 10 holds a first threshold value for determining the progress of adhering dirt and a second threshold value larger than the first threshold value. The control unit 10 calculates the area of the range specified as the adhered dirt in the captured image 40 as the progress of the adhered dirt, and notifies the external device when the progress of the adhered dirt exceeds the first threshold value. Further, the control unit 10 calculates the area of the range determined to be turbid in the captured image 40 as the turbidity progress, and notifies the external device 16 when the turbidity progress exceeds the second threshold value.

Since the turbidity is caused by suspended matter, it has less influence on the trouble of the object than the adhered dirt. In the above configuration, the second threshold value for the progress of turbidity is larger than the first threshold value for the progress of adherent stains. In this way, by making the execution condition of the treatment for the adhered dirt different from the execution condition of the treatment for the turbidity, the external device is notified that the adhered dirt or the turbidity exceeds the threshold value at the timing according to the adhered dirt or the turbidity. it can.

(6) Preferably, the control unit 10 holds a third threshold value for determining the progress of adhering dirt and a fourth threshold value larger than the third threshold value. The control unit 10 calculates the area of the range determined as the adhered dirt in the captured image 40 as the progress of the adhered dirt, and performs cleaning when the progress of the adhered dirt exceeds the third threshold value. Further, the control unit 10 calculates the area of the range determined as turbidity in the captured image 40 as the degree of turbidity progress, and performs cleaning when the degree of turbidity progress exceeds the fourth threshold value.

Since the turbidity is caused by suspended matter, it has less influence on the trouble of the object than the adhered dirt. In the above configuration, the fourth threshold value for the progress of turbidity is larger than the third threshold value for the progress of adherent stains. In this way, by making the execution conditions of the treatment for the adhered stains different from the execution conditions of the treatment for the turbidity, it is possible to wash the object at the timing according to the adhered stains and the turbidity.

(7) In the dirt detection device 1, an example of the object 2 is the drain pan 26 of the air conditioner 20. According to this configuration, it is possible to discriminate adhered dirt and turbidity with respect to the drain pan 26 as the object 2.

(8) The air conditioning system S includes a dirt detection device 1. According to this air conditioning system S, adhered dirt and turbidity can be discriminated. The dirt detection device 1 includes an acquisition unit 11 and a calculation unit 12. The acquisition unit 11 and the calculation unit 12 may be arranged so that the acquisition unit 11 and the calculation unit 12 are dispersed in the network N.

<Modification example>
In addition to the above-described embodiments, the stain detection device 1 of the present disclosure may be a combination of, for example, the following modifications and at least two modifications that do not contradict each other.

The calculation unit 12 may be configured as follows. The calculation unit 12 is configured to be available to a plurality of users. Specifically, the calculation unit 12 is capable of communicating with a plurality of objects 2 and is capable of communicating with a plurality of user terminals. The calculation unit 12 holds an address book for managing information transmission between the plurality of objects 2 and the plurality of user terminals. The address book associates the identification number of the air conditioner 20 with the user terminal of the user of the air conditioner 20 related to this identification number. When the calculation unit 12 acquires the captured image 40 of the object 2, the shooting time, and the identification number from the acquisition unit 11 and calculates the progress of the stain and the notification item, the calculation unit 12 refers to the address book and stains. Outputs the progress and notification items to the user terminal.

The acquisition unit 11 and the control unit 10 of the dirt detection device 1 may be configured as one unit. Such a dirt detection device 1 is arranged near the air conditioner 20. In this case, the dirt detection device 1 may directly communicate with the external device 16 (for example, the user terminal) without going through the network N.

The calculation unit 12 of the dirt detection device 1 may be provided in the external device 16 (for example, a user terminal). The calculation unit 12 may be composed of an application installed in the external device 16 and a calculation circuit in the external device 16.

The acquisition unit 11 and the calculation unit 12 of the dirt detection device 1 may be provided in the external device 16 (for example, the user terminal). For example, a personal computer is provided with an acquisition unit 11 and a calculation unit 12. In this case, the acquisition unit 11 is configured as a recording medium reading device or communication device. The calculation unit 12 may be composed of an application and a calculation circuit. The personal computer acquires the captured image 40 by the input operation of the user.

Although the embodiments of this device have been described above, it will be understood that various modifications of the form and details are possible without departing from the purpose and scope of the device described in the claims. ..

S Air conditioning system A Threshold 1 Stain detection device 2 Object 10 Control unit 16 External device 20 Air conditioner 23 Heat exchanger 26 Drain pan 40 Captured image 40a Target range

Claims (8)

A dirt detection device (1) for detecting dirt on the object (2), with a device including a container for storing water as the object (2).
A control unit (10) for processing a captured image (40) of the object (2) is provided.
The control unit (10) is a stain detection device that performs a process of discriminating at least adhering stains and turbidity from the stains of the object (2) based on the captured image (40). The control unit (10) identifies a discolored region different from the reference color of the object (2) for the plurality of captured images (40) having different times, and among the plurality of captured images (40), the said The dirt detection device according to claim 1, wherein a suspended matter image region corresponding to the movement of the suspended matter is specified as turbidity in the discolored region. The control unit (10) identifies a discolored region different from the reference color of the object (2) in the captured image (40), specifies a predetermined hue region within the discolored region, and determines the size of the predetermined hue region. The stain detection device according to claim 1, wherein when the value is larger than a predetermined value, the stain is specified as the adhered stain, and the region other than the region specified as the adhered stain in the discolored region is specified as the turbidity. The stain detection device according to any one of claims 1 to 3, wherein the control unit (10) discriminates between the adhered stain and the turbidity based on the hue of each pixel in the captured image (40). The control unit (10)
A first threshold value for determining the progress of the adhered stain and a second threshold value larger than the first threshold value are held.
The area of the range specified as the adhered dirt in the photographed image (40) is calculated as the progress of the adhered dirt, and when the progress of the adhered dirt exceeds the first threshold value, the external device (16) is notified.
Claims 1 to 4 calculate the area of the range determined to be turbid in the captured image (40) as the degree of turbidity, and notify the external device (16) when the degree of turbidity exceeds the second threshold value. The dirt detection device according to any one of the above. The control unit (10)
A third threshold value for determining the progress of the adhered stain and a fourth threshold value larger than the third threshold value are held.
The area of the range determined to be the adhered dirt in the photographed image (40) is calculated as the progress of the adhered dirt, and when the progress of the adhered dirt exceeds the third threshold value, cleaning is performed.
According to any one of claims 1 to 5, the area of the range determined to be turbid in the photographed image (40) is calculated as the degree of turbidity, and cleaning is performed when the degree of turbidity exceeds the fourth threshold value. The dirt detector described. The dirt detection device according to any one of claims 1 to 6, wherein the object (2) is a drain pan of an air conditioner. An air conditioning system including the dirt detection device according to any one of claims 1 to 7.