JP7206537B2 - Water discharge device - Google Patents

Description

The present invention relates to a water discharging device.

In Patent Document 1, an inference condition regarding the situation of providing a shower is detected, and a shower temperature inference means infers a comfortable shower temperature for the shower user based on the inference condition with a neural network, and based on the inferred shower temperature. A technique is described for providing a shower with

JP-A-7-322967

The technology described in Patent Document 1 targets only the shower temperature for learning and control. For this reason, when the technique described in Patent Document 1 is applied to the water discharge control of the washing stand, for example, water is wasted from the water discharge unit, or the amount of water discharged from the water discharge unit or the water discharge time is insufficient. In addition, there is a high possibility that there will be an excess or deficiency in the water discharge amount or water discharge time. However, regarding the amount of water to be discharged or the duration of water discharge desired by a user who has arrived in front of the water discharge unit, the user may decide whether to use the water by hand or by fetching water with a bucket. Therefore, it is difficult to control the amount of spouted water and the duration of spouted water according to the user's situation.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a water discharger capable of appropriately controlling at least one of water discharge amount and water discharge time according to the situation of a user arriving in front of the water discharger. is.

The water discharging device according to the first aspect of the present invention comprises a photographing unit for photographing a predetermined area in which a user who has arrived in front of the water discharging unit is shown, an image photographed by the photographing unit during past water discharge from the water discharging unit, and Based on a trained model learned from information including water discharge conditions including at least one of a water discharge amount and a water discharge time in past water discharge from the water discharge unit, and an image currently captured by the imaging unit, the water discharge is performed. a water discharge condition acquisition unit that acquires a water discharge condition including at least one of a water discharge amount and a water discharge time in the current water discharge from a unit; and controlling water discharge from the water discharge unit according to the water discharge conditions acquired by the water discharge condition acquisition unit. a water discharge controller for detecting a water discharge condition in the water discharge from the water discharge unit; and a learning unit that updates the learned model based on the water discharge condition detected in the water discharge in response to the instruction and the image captured by the imaging unit in the water discharge in response to the instruction. I'm in.

In the first aspect of the present invention, since the photographing section photographs the predetermined area in which the user arriving in front of the water discharging section is captured, the situation of the user arriving in front of the water discharging section is photographed as an image by the photographing section. Also, the water discharge condition acquisition unit acquires water discharge conditions including at least one of the water discharge amount and the water discharge time of the current water discharge from the water discharge unit based on the learned model and the image captured this time by the image capturing unit. Here, the learning model used by the water discharge condition acquisition unit to acquire the water discharge condition is at least one of an image captured by the imaging unit during past water discharge from the water discharge unit and a water discharge volume and water discharge time during past water discharge from the water discharge unit. It is a model learned from information including discharge conditions including

For this reason, the water discharge condition acquisition unit, based on the situation of the user who arrived in front of the water discharge unit this time, sets the water discharge conditions for the current water discharge to be similar to the water discharge conditions in the past use of the water discharge unit with similar situations. Conditions will be obtained. The water discharge control section controls water discharge from the water discharge section according to the water discharge condition acquired by the water discharge condition acquisition section. Therefore, according to the first aspect of the invention, it is possible to appropriately control at least one of the water spouting amount and the water spouting time according to the situation of the user who arrives in front of the water spouting section.
Further, in the invention according to claim 1, when water discharge from the water discharge unit is instructed via the water discharge instruction unit, the learning is completed based on the water discharge conditions in water discharge according to the instruction and the image photographed by the photographing unit. Update your model. As a result, it is possible to improve the accuracy of controlling the water discharge conditions according to the situation of the user who has arrived in front of the water discharger. Further, when water discharge from the water discharge unit is instructed via the water discharge instruction unit, the learned model is automatically updated. It is possible to save the user's trouble.

The invention according to claim 2 is the invention according to claim 1, wherein the water discharge condition used for learning of the trained model includes at least one of water discharge volume and water discharge time and at least one of water temperature and water pressure. The water discharge condition acquisition unit acquires the water discharge condition including at least one of water temperature and water pressure in addition to at least one of water discharge amount and water discharge time.

In the second aspect of the invention, since the water discharge conditions include at least one of water temperature and water pressure in addition to at least one of water discharge amount and water discharge time, the condition of the user arriving in front of the water discharge unit can be adjusted. At least one of water temperature and water pressure can be appropriately controlled in addition to at least one of water spouting amount and water spouting time.

The invention according to claim 3 is based on the invention according to claim 1 or claim 2, wherein the water discharge condition acquisition unit detects the water discharge condition when a user arriving in front of the water discharge unit is photographed by the photographing unit. Obtaining a condition, the water discharge control unit controls water discharge from the water discharge unit when the water discharge condition obtained by the water discharge condition obtaining unit indicates that water is to be discharged.

In the third aspect of the present invention, the water discharge condition acquired by the water discharge condition acquisition unit is used to determine whether or not water is required to be discharged. , the water discharge from the water discharge unit is controlled. This makes it possible to simplify the configuration of the water discharger as compared with the case where the judgment as to whether or not water discharge is required is performed by analyzing the image captured by the imaging unit.

ADVANTAGE OF THE INVENTION This invention has the effect that at least one of the water spouting amount and water spouting time can be controlled appropriately according to the conditions of the user who arrived in front of the water spouting part.

It is a schematic block diagram of a water discharger. Fig. 2 is a perspective view showing the arrangement of a water discharger, a water discharge instruction lever, and a camera; FIG. 4 is a conceptual diagram showing an example of a trained model; It is a flow chart which shows water discharge processing.

An example of an embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows a water discharging device 10 installed indoors of a house. The water discharge device 10 includes a pipe 12 having one end connected to a water supply source 60, the other end of the pipe 12 being connected to the other end of a pipe 14 having one end connected to a hot water supply system 62 and the other end connected to a water discharge section 28. and the other end of the pipe 16, respectively.

A flow rate control valve 18 is provided in the middle of the pipe 12 , and the flow rate of water supplied from the water supply source 60 and flowing through the pipe 12 is adjusted by the flow rate control valve 18 . A flow control valve 20 is provided in the middle of the pipe 14 , and the flow control valve 20 regulates the flow rate of water (hot water) supplied from the hot water supply equipment 62 and flowing through the pipe 14 . Therefore, the temperature of the water supplied from the pipes 12 and 14 and flowing through the pipe 16 is adjusted according to the valve opening ratio of the flow control valves 18 and 20 .

A flow control valve 22 , a water pressure sensor 24 and a water temperature sensor 26 are provided in the middle of the pipe 16 . A flow control valve 22 is used to adjust the flow rate of the water supplied from the pipes 12 and 14 and flow through the pipe 16 and to stop water discharge. A water pressure sensor 24 detects the water pressure of water flowing through the pipe 16 , and a water temperature sensor 26 detects the temperature of water flowing through the pipe 16 . In addition, the water pressure sensor 24 and the water temperature sensor 26 are examples of the detection unit.

As shown in FIG. 2 , the water discharger 28 is provided above a tank 52 provided in a sink 50 so as to discharge water toward the bottom of the tank 52 . An infrared sensor 31 is provided at the tip of the water discharger 28 adjacent to the water discharge port of the water discharger 28 . The infrared sensor 31 detects an object (for example, a user's hand) approaching the tip of the water discharger 28 .

A water discharge instruction lever 30 is provided in the vicinity of the water discharge part 28 . The water discharge instruction lever 30 is turned in the direction of the arrow A in FIG. 2 to instruct to stop water discharge from the water discharger 28 and to adjust the amount of water discharged, and is turned in the direction of the arrow B in FIG. Then, adjustment of the temperature of the water discharged from the water discharger 28 is instructed. The rotational position of the water discharge instruction lever 30 is detected by a sensor incorporated in the water discharge instruction lever 30 . Note that the infrared sensor 31 and the water discharge instruction lever 30 are examples of a water discharge instruction unit.

A camera 32 is attached to the sink 50, and a room temperature sensor 34 (see FIG. 1) is provided. The camera 32 shoots a predetermined area set so that an object (for example, a user such as a resident of a house in which the water discharger 10 is installed) is captured in front of the water discharger 28 (in front of the sink 50). . Note that the camera 32 is an example of an imaging unit.

As shown in FIG. 1, the water discharger 10 includes a computer 36, flow control valves 18, 20, 22, a water pressure sensor 24, a water temperature sensor 26, a water discharge instruction lever 30, an infrared sensor 31, a camera 32, and a room temperature sensor 34. is connected to computer 36 .

The computer 36 includes a CPU, a memory, a non-volatile storage section storing programs for executing various processes, a network interface, and the like. The computer 36 functionally includes an information acquisition section 38 , a learning data storage section 40 , a learning section 42 , a learned model storage section 44 , a water discharge condition acquisition section 46 and a water discharge control section 48 .

The information acquisition unit 38 obtains the water pressure detected by the water pressure sensor 24, the water temperature detected by the water temperature sensor 26, the rotation position of the water discharge instruction lever 30 detected by the built-in sensor of the water discharge instruction lever 30, and the object detected by the infrared sensor 31. As a result of the detection, the image captured by the camera 32 and the room temperature detected by the room temperature sensor 34 are acquired.

The learning data storage unit 40 stores an image taken by the camera 32 when water is spouted from the water spouting unit 28 based on the object detection by the infrared sensor 31 or an instruction via the water spouting instruction lever 30, the room temperature, and the date and time. , water discharge conditions (an example of a teacher signal) such as water discharge amount, water discharge time, water temperature, and water pressure when water is discharged from the water discharge unit 28 based on object detection by the infrared sensor 31 or instructions via the water discharge instruction lever 30; A plurality of learning data associated with are stored.

Based on a plurality of pieces of learning data stored in the learning data storage unit 40, the learning unit 42 uses the images, the room temperature, and the date and time captured by the camera 32 to determine the water discharge unit in the situation represented by these images, the room temperature, and the date and time. A learned model is generated by learning a model for outputting an appropriate water discharge condition from 28, or an already generated learned model is updated. In this embodiment, a neural network as shown in FIG. 3 can be used as an example of a model, and deep learning can be used as an example of a learning algorithm. The learned model storage unit 44 stores the learned model generated by the learning unit 42 . The learning unit 42 is an example of a learning unit.

Note that the learned model storage unit 44 may store in advance a learned model generated from learning data obtained by one or more other water discharging devices 10 . In this case, the learning unit 42 updates the learned model stored in the learned model storage unit 44 . When compared with the water discharger 10, in other water dischargers 10, the user who arrives in front of the water discharger 28 is different, and there is a possibility that the user who has arrived in front of the water discharger 28 may also give different water discharge instructions. Therefore, a learned model generated from learning data obtained from one or more other water discharging devices 10 is not optimized according to the house in which the water discharging device 10 is installed. However, the time required for optimizing the learned model can be shortened compared to the case where the learned model is generated/updated using only the learning data obtained by the water discharging device 10 . In this embodiment, a mode in which a learned model is pre-stored in the learned model storage unit 44 will be described, but the present invention is not limited to this.

The water discharge condition acquisition unit 46 acquires the information acquired by the information acquisition unit 38 when the user arrives in front of the water discharge unit 28, the infrared sensor 31 does not detect an object, and the user does not operate the water discharge instruction lever 30. Water discharge conditions from the water discharge unit 28 are acquired by performing signal processing of the multi-layer neural network based on the image, room temperature, date and time, and the learned model stored in the learned model storage unit 44 . In addition, when an object is detected by the infrared sensor 31 or when the water discharge instruction lever 30 is operated by the user who has arrived in front of the water discharger 28, the water discharge condition acquisition unit 46 detects the object by the infrared sensor 31, A water discharge condition is acquired according to the rotation position of the water discharge instruction lever 30 . Note that the water discharge condition acquisition unit 46 is an example of a water discharge condition acquisition unit.

The water discharge control unit 48 controls the valve opening degrees of the flow control valves 18, 20, and 22 according to the water discharge conditions acquired by the water discharge condition acquisition unit 46, thereby discharging water from the water discharge unit 28 according to the water discharge conditions. I do. Note that the water discharge control unit 48 is an example of a water discharge control unit.

Next, as an operation of the embodiment, the water discharge process executed by the computer 36 will be described with reference to FIG. In step 100 of the water discharge process, the information acquisition section 38 acquires from the camera 32 an image of a predetermined area in front of the water discharge section 28 photographed by the camera 32 .

In step 102, the information acquisition section 38 determines whether or not an object such as a user has arrived in front of the water discharge section 28 (sink 50) based on the image acquired in step 100. FIG. To determine whether an object has arrived, for example, the image acquired in step 100 is compared with a pre-stored reference image (an image in which no object has arrived), and the difference (object area) is It can be realized by determining whether or not there is. Objects other than the user that come in front of the water discharger 28 include, for example, pets raised by the user.

If the object has not arrived in front of the water discharger 28, the determination in step 102 is negative and the process returns to step 100. Accordingly, when an object has not arrived in front of the water discharger 28, image acquisition from the camera 32 is repeated, and whether or not an object has arrived in front of the water discharger 28 is monitored.

Also, when the object arrives in front of the water discharger 28, the determination in step 102 is affirmative, and the process proceeds to step 104. FIG. At step 104 , the information acquiring section 38 acquires from the camera 32 an image of a predetermined area in front of the water discharging section 28 photographed by the camera 32 . In the next step 106 , the information acquiring section 38 determines whether or not an object such as a user has left the front of the water discharging section 28 based on the image acquired in step 104 .

If the determination in step 106 is affirmative, the process returns to step 100, but if the determination in step 106 is negative, the process proceeds to step 108. In step 108 , the water discharge condition acquisition unit 46 determines whether or not the infrared sensor 31 has detected an object approaching the water discharge port of the water discharge unit 28 . If the determination in step 108 is negative, the process proceeds to step 110. In step 110, the water discharge condition acquisition unit 46 determines that the water discharge instruction lever 30 is operated by an object (a user in this case) that has arrived in front of the water discharge unit 28. determine whether or not

The determination in step 108 is made, for example, after the external shape of the object area in the image captured by the camera 32 has changed corresponding to the action of reaching toward the water discharge instruction lever 30, and then the water discharge instruction lever 30 can be performed based on whether or not a change in the rotation position of the water discharge instruction lever 30 is detected by the built-in sensor.

If the determination in step 110 is negative, the process proceeds to step 112. In step 112, the water discharge condition acquisition unit 46 stores the image, the room temperature, and the date and time acquired by the information acquisition unit 38 in the learned model storage unit 44. The condition for water discharge from the water discharge unit 28 is acquired by performing signal processing of the multi-layer neural network based on the learned model.

In the next step 114, the water discharge control unit 48 determines whether or not the water discharge amount, the water discharge time, and the water pressure included in the water discharge conditions acquired in step 112 are values greater than zero. It is determined whether water discharge is necessary. If the situation of the object arriving in front of the water discharger 28 photographed by the camera 32 is not similar to when water was discharged from the water discharger 28 in the past, the water discharge conditions acquired in step 112 are satisfied. The included water discharge amount, water discharge time, and water pressure become 0. In this case, the determination at step 114 is negative and the process returns to step 104 . In this way, when an object arrives in front of the water discharger 28, until the arriving object leaves from the front of the water discharger 28, until the determination in any of steps 108, 110, 114 is affirmative. , steps 104 to 114 are repeated, and the water discharge unit 28 does not discharge water.

For example, when a user who has arrived in front of the water discharger 28 makes an action of holding out his/her hand to the water discharge port of the water discharger 28, an object (for example, the user's hand) approaching the water discharge port of the water discharger 28 ) is detected by the infrared sensor 31 and the determination in step 108 is affirmative, so that the process proceeds to step 116 . Further, for example, when the water discharge instruction lever 30 is operated by a user who has arrived in front of the water discharger 28, the determination in step 110 is affirmative, and the process proceeds to step 116. FIG. At step 116 , the information acquiring section 38 acquires from the camera 32 an image of a predetermined area in front of the water discharging section 28 photographed by the camera 32 .

In step 118, when an object is detected by the infrared sensor 31, the water discharge condition acquisition unit 46 moves the water discharge instruction lever 30 to the rotating position in the direction of arrow B in FIG. A water discharge condition for discharging water of the corresponding temperature is acquired. Further, the water discharge condition acquisition unit 46 acquires the water discharge condition according to the rotation position of the water discharge instruction lever 30 when the water discharge instruction lever 30 is operated. The water discharge control unit 48 controls the valve opening degrees of the flow control valves 18 , 20 , 22 according to the water discharge conditions acquired by the water discharge condition acquisition unit 46 . As a result, water is discharged from the water discharger 28 under water discharge conditions corresponding to the user's action of extending his/her hand to the water discharge port of the water discharger 28 or operation of the water discharge instruction lever 30 .

The user pulls his or her hand out of the spout of the water spouting portion 28, or the user operates the water spouting instruction lever 30 to stop the water. Then, in step 120, the information acquisition unit 38 acquires the water discharge conditions for the current water discharge. That is, the information acquisition unit 38 acquires the valve opening degree of the flow control valve 22 as the water discharge amount from the water discharge condition acquisition unit 46, and acquires the valve opening time of the flow control valve 22 as the water discharge time from the water discharge condition acquisition unit 46. , the water temperature is obtained from the water temperature sensor 26 and the water pressure is obtained from the water pressure sensor 24 .

In step 122 , the information acquisition section 38 acquires the room temperature from the room temperature sensor 34 and acquires the date and time from the built-in timer of the computer 36 . Then, the information acquisition unit 38 associates the image acquired in step 116, the room temperature and the date and time acquired, and the water discharge conditions acquired in step 120, and stores them in the learning data storage unit 40 as new learning data. Let

At step 124, the learning unit 42 determines whether or not the timing for updating the learned model stored in the learned model storage unit 44 has arrived. The learned model may be updated each time new learning data is stored in the learning data storage unit 40 (in this case, the determination in step 124 is affirmative each time), or the learning data storage unit It may be performed each time new learning data is stored in the memory 40 to a predetermined amount or more. Also, the learned model may be updated at regular intervals. If the determination in step 124 is negative, the process returns to step 104 .

If the determination in step 124 is affirmative, the process proceeds to step 126. In step 126, the learning unit 42 performs learning based on the new (unlearned) learning data stored in the learning data storage unit 40. , updates the learned model stored in the learned model storage unit 44 . Note that an algorithm such as error backpropagation can be applied to update the learned model.

Further, if the situation of the object arriving in front of the water discharger 28 photographed by the camera 32 is similar to when water was discharged from the water discharger 28 in the past, the water discharge obtained in step 112 is The spouting amount, spouting time, and water pressure included in the conditions become values greater than zero. In this case, the determination in step 114 is affirmative, and the process proceeds to step 128. In step 128, the water discharge control unit 48 adjusts the opening degrees of the flow control valves 18, 20, and 22 according to the water discharge conditions acquired in step 112. is controlled, water is discharged from the water discharger 28 in accordance with the water discharge conditions.

As a result, when the situation of the object arriving in front of the water discharger 28 photographed by the camera 32 is similar to when water was discharged from the water discharger 28 in the past, the user can move the water discharger 28 forward. Water is spouted from the water spouting portion 28 without performing an action of holding out a hand or the like to the water spout or operating the water spouting instruction lever 30. - 特許庁again,

In the next step 130 , the water discharge control unit 48 determines whether or not the amount of hot water supplied from the hot water supply equipment 62 is equal to or greater than a predetermined value under the water discharge conditions acquired in step 116 . If the determination in step 130 is negative, the process returns to step 100 without performing any processing. Further, when the determination in step 130 is affirmative, the process proceeds to step 132, and in step 132, the water discharge control unit 48 instructs the hot water supply equipment 62 to store hot water. In this case, the hot water supply equipment 62 stores a predetermined amount of hot water in the hot water tank.

An example of the effect by performing the water discharge process mentioned above is enumerated below.

The image captured by the camera 32 shows the user and other objects that have arrived in front of the water discharge unit 28, and the user who has arrived may perform an action of holding out his/her hand or the like to the water discharge port of the water discharge unit 28. , when the water discharge instruction lever 30 is operated, an object held out in front of the water discharge unit 28 by the user (for example, a hand, a cup, a toothbrush, a dust cloth, etc.), the state of the object (shape, dirt, etc.), The user's actions (for example, actions of washing hands, washing face, drinking water, washing vegetables, polishing rice, etc.) are recorded. An image showing such a thing is stored as learning data together with the room temperature, the date and time, and the water discharge condition, and is used to generate (update) the learned model.

As a result, for example, when the first user arrives in front of the water discharger 28 and the first user puts out his hand in front of the water discharger 28, the first user, for example, Water discharge conditions when washing hands in the past, water discharge amount, water discharge time, water pressure, water temperature (for example, depending on room temperature at that time) according to the dirtiness of the hands of the first user, presence or absence of foam Water is spouted from the water spouting portion 28 at the same water temperature.

Further, for example, when a second user arrives in front of the water discharger 28 and the second user presents a toothbrush in front of the water discharger 28, the second user is at the sink 50, for example. Water is spouted from the water spouting unit 28 at a water spout amount, water spouting time, water pressure, and water temperature (for example, normal temperature) according to the water spouting conditions when the toothbrush was washed in the past.

Also, for example, when a third user arrives in front of the water discharger 28 and this third user presents a cup in front of the water discharger 28, the third user is at the sink 50, for example. Water is spouted from the water spouting part 28 with the water spouting amount, water spouting time, water pressure, and water temperature according to the water spouting conditions, etc. when drinking water using a cup in the past.

Furthermore, for example, when a pet arrives in front of the water discharger 28, the object was not detected by the infrared sensor 31 when the pet arrived in front of the water discharger 28 in the past, and the water discharge instruction lever 30 was also operated. Therefore, the judgment in step 114 is negative, and water is not discharged from the water discharger 28 this time as well as in the past.

Also, for example, if any user arrives in front of the water discharger 28 and this user presents a rag in front of the water discharger 28, for example, this user has washed the rag in the sink 50 in the past. Water is spouted from the water spouting portion 28 with the water spout amount, water spouting time, water pressure, and room temperature according to the water spouting conditions, etc.

Also, for example, if any user arrives in front of the water discharger 28 and this user holds out a bucket in front of the water discharger 28, for example, this user has used the bucket in the sink 50 in the past. Water is spouted from the water spouting portion 28 with a spouting amount, water spouting time, water pressure, and water temperature according to the water spouting conditions and the like when the water is scooped up.

In this way, the learned model stored in the learned model storage unit 44 is updated by the user in the house where the water discharging device 10 is installed, as the learned model is updated by the learning unit 42. Optimized according to water discharge accompanying instructions. Then, by acquiring the water discharge conditions using the optimized learned model and controlling the water discharge from the water discharge unit 28, the optimum water discharge conditions according to the situation of the user who arrives in front of the water discharge unit 28, etc. Water can be discharged from the water discharger 28 at .

Further, for example, when a dust cloth is washed in the sink 50 or a bucket is used to draw water, the amount of water used under the water discharge conditions acquired by the water discharge condition acquisition unit 46 becomes relatively large (at least one of the water discharge amount and the water discharge time). becomes larger). For this reason, for example, in general cleaning in which washing of dust cloths and drawing water with a bucket are repeated, the amount of hot water supplied from the hot water supply equipment 62 becomes equal to or greater than a predetermined value, and the determination in step 130 is affirmative. Hot water storage is instructed to the facility 62, and the hot water supply facility 62 stores hot water in the hot water storage tank. As a result, even in a situation where the amount of hot water supplied from the hot water supply equipment 62 is large, such as during general cleaning, it is possible to prevent the hot water supply from the hot water supply equipment 62 from failing in time due to the hot water storage tank becoming empty.

In addition, although the aspect in which the water discharging apparatus 10 is installed indoors of the house was demonstrated above, it is not limited to this and the water discharging apparatus may be installed outdoors.

Also, in the above description, a mode of learning a neural network model as an example of a model by deep learning has been described, but the present invention is not limited to this. For example, a model different from the neural network model may be learned by a learning method different from deep learning.

10 Water discharge device 18 Flow rate adjustment valve 20 Flow rate adjustment valve 22 Flow rate adjustment valve 24 Water pressure sensor 26 Water temperature sensor 28 Water discharge unit 30 Water discharge instruction lever 31 Infrared sensor 32 Camera 34 Room temperature sensor 36 Computer 42 Learning unit 46 Water discharge condition acquisition unit 48 Water discharge control unit 50 sink 62 hot water supply equipment

Claims (3)

a photographing unit for photographing a predetermined area in which a user arriving in front of the water discharge unit is captured;
a trained model learned from information including water discharge conditions including at least one of water discharge amount and water discharge time during past water discharge from the water discharge unit and images captured by the imaging unit during past water discharge from the water discharge unit; a water discharge condition acquisition unit that acquires a water discharge condition including at least one of a water discharge amount and a water discharge time in the current water discharge from the water discharge unit, based on the image captured this time by the imaging unit;
a water discharge control unit that controls water discharge from the water discharge unit according to the water discharge condition acquired by the water discharge condition acquisition unit;
a detection unit that detects water discharge conditions in the water discharge from the water discharge unit;
When water discharge from the water discharge unit is instructed via a water discharge instruction unit for instructing water discharge, the water discharge condition for water discharge according to the instruction detected by the detection unit, and water discharge according to the instruction. a learning unit that updates the learned model based on the image captured by the imaging unit in
Water discharge device including. The water discharge condition used for learning the trained model includes at least one of water discharge volume and water discharge time and at least one of water temperature and water pressure,
The water discharger according to claim 1, wherein the water discharge condition acquisition unit acquires the water discharge condition including at least one of water temperature and water pressure in addition to at least one of water discharge amount and water discharge time. The water discharge condition acquisition unit acquires the water discharge condition when a user arriving in front of the water discharge unit is photographed by the imaging unit,
The water discharger according to claim 1 or 2, wherein the water discharge control unit controls water discharge from the water discharge unit when the water discharge condition acquired by the water discharge condition acquisition unit is water discharge.

Images