Washing machine and detergent placement method

Abstract

The present application discloses a washing machine and a method for dispensing detergent. The washing machine provided by the present application can determine at least one target recognition model from a plurality of clothing recognition models based on a captured image of the clothes to be washed, and use the at least one target recognition model to recognize the captured image, thereby determining the characteristic information of the clothes to be washed. Since there are sample images whose similarity with the captured image is greater than a similarity threshold in the sample image set corresponding to each target recognition model, and the target recognition model is obtained by training based on a corresponding sample image set, the accuracy of the characteristic information of the clothes to be washed determined by the at least one target recognition model can be ensured. Furthermore, since the washing machine can determine the detergent dispensing information based on the determined characteristic information, and control the detergent storage box to automatically dispense detergent into the washing drum according to the dispensing information, the efficiency and accuracy of detergent dispensing are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of electronic technology, and in particular to a washing machine and a method for dispensing detergent. Background Art

[0002] To ensure a good washing effect of the clothes, the user can add an appropriate amount of detergent to the washing drum in the washing machine according to the material and quantity of the clothes to be washed. During the washing process, the washing machine can put the detergent in the detergent storage box into the washing drum.

[0003] However, the above-mentioned method of adding detergent is relatively inefficient. Summary of the invention

[0004] The present application provides a washing machine and a method for adding detergent, which can solve the problem of low efficiency of the method for adding detergent in the related art. The technical solution is as follows:

[0005] In one aspect, a washing machine is provided, comprising: a camera, a washing drum, a controller and a detergent storage box:

[0006] The camera is used to photograph the laundry to be washed to obtain a photographed image, wherein the laundry to be washed is the laundry put into the washing drum;

[0007] The controller is used to:

[0008] Based on the captured image, at least one target recognition model is determined from a plurality of clothing recognition models, wherein the plurality of clothing recognition models correspond one-to-one to a plurality of sample image sets, each clothing recognition model is trained based on a corresponding sample image set, and the sample image set corresponding to each target recognition model contains a sample image whose similarity with the captured image is greater than a similarity threshold;

[0009] Inputting the captured image into the at least one target recognition model to obtain feature information of the laundry to be washed, wherein the feature information includes at least one of material, color and dirt information;

[0010] Determining detergent delivery information based on the characteristic information of the laundry to be washed, wherein the delivery information includes the type and delivery amount of the detergent;

[0011] The detergent storage box is controlled to dispense detergent into the washing tub according to the dispensing information.

[0012] Optionally, the controller is used to:

[0013] Inputting the captured image into the at least one target recognition model to obtain a recognition result output by each target recognition model, wherein the recognition result includes a probability that the laundry to be washed meets each of a plurality of candidate feature information;

[0014] The candidate feature information with the highest probability in the recognition result output by the at least one target recognition model is determined as the feature information of the clothes to be washed.

[0015] Optionally, the controller is used to:

[0016] Calculating the similarity between the captured image and sample images in the plurality of sample image sets;

[0017] Based on the similarity, at least one object recognition model is determined from the plurality of clothing recognition models.

[0018] Optionally, the controller is further configured to:

[0019] receiving the plurality of sample image sets uploaded by the terminal, and the recognition result of each sample image in each of the sample image sets;

[0020] Each of the sample image sets and the recognition results is used to train the initial model to obtain a clothing recognition model corresponding to each of the sample image sets.

[0021] Optionally, the controller is further configured to:

[0022] If there is no sample image in the plurality of sample image sets whose similarity with the captured image is greater than the similarity threshold, inputting the captured image into a reference recognition model to obtain feature information of the laundry to be washed;

[0023] The reference recognition model is a model trained after the multiple clothing recognition models.

[0024] Optionally, the controller is used to:

[0025] If, based on the characteristic information of the laundry to be washed, it is detected that the laundry to be washed includes a first type of laundry of a target material, determining that the type of detergent includes a laundry softener;

[0026] If it is detected that the laundry to be washed includes a second type of laundry of a target color, based on the characteristic information of the laundry to be washed, determining that the type of detergent includes a laundry bleach;

[0027] If it is detected based on the characteristic information of the laundry to be washed that the plurality of laundry to be washed include a third type of laundry with dirt, then it is determined that the type of the detergent includes a laundry detergent containing enzymes;

[0028] If it is detected based on the characteristic information of the laundry to be washed that the plurality of laundry to be washed do not include the third type of laundry with dirt, it is determined that the type of the detergent includes laundry liquid without enzyme.

[0029] Optionally, the controller is further configured to:

[0030] If the type of detergent includes fabric softener, determining a first dosage of the fabric softener based on the total volume of the first type of clothing, the first dosage being positively correlated with the total volume of the first type of clothing;

[0031] If the type of the detergent includes a clothing bleach, then based on the total volume of the second type of clothing, determining a second dosage of the clothing bleach, the second dosage being positively correlated with the total volume of the second type of clothing;

[0032] If the type of detergent includes laundry liquid, a third dosage of the laundry liquid is determined based on the total weight of the laundry to be washed, and the third dosage is positively correlated with the total weight of the laundry to be washed.

[0033] Optionally, the first dosage V of the fabric softener is 1 Satisfaction: V 1 =V 01 *ρ 1 *K 1 ;

[0034] Among them, V 01 is the total volume of the first type of clothing, ρ 1 is the density of the first type of clothing, K 1 is the preset first delivery coefficient;

[0035] The second dosage V of the clothing bleaching agent 2 Satisfaction: V 2 =V 02 *ρ 2 *K 2 ;

[0036] Among them, V 02 is the total volume of the second type of clothing, ρ 2 is the density of the second type of clothing, K 2 It is the preset second delivery coefficient.

[0037] Optionally, the controller is further configured to:

[0038] Determining water quality in an area where the washing machine is located based on the location information of the washing machine;

[0039] If it is determined that the water quality in the area where the washing machine is located is hard water, a prompt message is issued, and the prompt message is used to instruct the addition of a water softener.

[0040] On the other hand, a method for dispensing detergent is provided, the method comprising:

[0041] Acquire a photographed image obtained by photographing clothes to be washed, wherein the clothes to be washed are clothes put into a washing drum of a washing machine;

[0042] Based on the captured image, at least one target recognition model is determined from a plurality of clothing recognition models, wherein the plurality of clothing recognition models correspond one-to-one to a plurality of sample image sets, each clothing recognition model is trained based on a corresponding sample image set, and the sample image set corresponding to each target recognition model contains a sample image whose similarity with the captured image is greater than a similarity threshold;

[0043] Inputting the captured image into the at least one target recognition model to obtain feature information of the laundry to be washed, wherein the feature information includes at least one of material, color and dirt information;

[0044] Determining detergent delivery information based on the characteristic information of the laundry to be washed, wherein the delivery information includes the type and delivery amount of the detergent;

[0045] The detergent storage box of the washing machine is controlled to dispense detergent into the washing drum according to the dispensing information.

[0046] On the other hand, a washing machine is provided, comprising a processor and a memory, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement the detergent dispensing method described in the above aspect.

[0047] On the other hand, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the detergent dispensing method described in the above aspects.

[0048] On the other hand, a computer program product comprising instructions is provided, and when the computer program product is run on a computer, the computer is enabled to execute the detergent dosing method described in the above aspects.

[0049] The beneficial effects of the technical solution provided by this application include at least:

[0050] The present application provides a washing machine and a method for dispensing detergent. The washing machine can determine at least one target recognition model from a plurality of clothing recognition models based on a captured image of the clothes to be washed, and use the at least one target recognition model to recognize the captured image, thereby determining the characteristic information of the clothes to be washed. Since the sample image set corresponding to each target recognition model contains sample images whose similarity with the captured image is greater than a similarity threshold, and the target recognition model is trained based on a corresponding sample image set, the accuracy of the characteristic information of the clothes to be washed determined by the at least one target recognition model can be ensured. Furthermore, since the washing machine can determine the detergent dispensing information based on the determined characteristic information, and control the detergent storage box to automatically dispense detergent into the washing drum according to the dispensing information, the efficiency and accuracy of detergent dispensing are effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 It is a structural schematic diagram of a detergent delivery system provided in an embodiment of the present application;

[0053] Figure 2 is a structural schematic diagram of another detergent delivery system provided in an embodiment of the present application;

[0054] Figure 3 It is a schematic diagram of a flow chart of a method for dispensing detergent provided in an embodiment of the present application;

[0055] Figure 4 is a schematic flow chart of another method for dispensing detergent provided in an embodiment of the present application;

[0056] Figure 5 It is a schematic diagram of prompt information displayed on a display screen of a washing machine provided in an embodiment of the present application;

[0057] Figure 6 It is a schematic diagram of a display screen of a washing machine provided in an embodiment of the present application displaying operation information;

[0058] Figure 7 It is a schematic diagram of a display screen of a terminal provided in an embodiment of the present application displaying prompt information and operation information;

[0059] Figure 8 It is a schematic diagram of a display screen of a washing machine provided in an embodiment of the present application showing dispensing information;

[0060] Fig. 9 It is a structural schematic diagram of a washing machine provided in an embodiment of the present application;

[0061] Fig.10 It is a structural schematic diagram of another washing machine provided in an embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0063] Figure 1 Schematic diagram of a detergent delivery system provided in an embodiment of the present application. Figure 1 The system may include a server 110 and a washing machine 120. A wired or wireless communication connection is established between the server 110 and the washing machine 120.

[0064] The server 110 may be a single server, or a server cluster consisting of several servers, or a cloud computing service center. The washing machine 120 may be a pulsator washing machine, a drum washing machine, an agitator washing machine, an ultrasonic washing machine, an electromagnetic decontamination washing machine, or an active oxygen washing machine.

[0065] refer to Figure 1 The washing machine 120 may include a camera 121, which can capture images of the laundry to be washed when the user puts the laundry into the washing drum. After that, the washing machine 120 can send the captured images to the server 110. The server 110 can process the captured images of the laundry to be washed to determine the placement information of the detergent required in the washing process of the laundry. The placement information may include the type of detergent and the amount of each detergent. After that, the server 110 can send the placement information to the washing machine 120, so that the washing machine 120 controls the detergent storage box to place detergent into the washing drum according to the placement information.

[0066] It is understandable that the washing machine 120 may also be capable of processing the photographed image of the laundry to be washed and determining the detergent delivery information. Accordingly, the detergent delivery system may also not include the server 110.

[0067] Optionally, refer to Figure 2The system may further include at least one terminal 130. The at least one terminal 130 establishes a wired or wireless communication connection with the server 110 and the washing machine 120. The terminal 130 may be a mobile phone, a tablet computer, a laptop computer, a desktop computer, a vehicle-mounted terminal, a smart home device or a wearable device. For example, see Figure 2 , the terminal 130 can be a mobile phone.

[0068] In the embodiment of the present application, a washing machine management application (APP) may be installed in the terminal 130, and the server 110 may be a background server of the washing machine management APP. The terminal 130 may control the working process of the washing machine 120 through the washing machine management APP. In addition, the terminal 130 may also be used to send sample images to the server 110 so that the server 110 can train a clothing recognition model.

[0069] Figure 3 is a schematic diagram of a method for dispensing detergent provided in an embodiment of the present application, and the method can be applied to a detergent dispensing device. The detergent dispensing device can be Figure 1 The server 110 or washing machine 120 in the scene shown. The following description takes the detergent delivery device as the washing machine 120 as an example, and reference Figure 3 , the method comprising:

[0070] Step 101: Acquire an image of the laundry to be washed.

[0071] In an embodiment of the present application, the washing machine includes a camera. When a user puts the laundry to be washed into the washing drum, the camera can take a picture of the laundry to be washed to obtain a photographed image.

[0072] Step 102: Determine at least one target recognition model from a plurality of clothing recognition models based on the captured image.

[0073] In an embodiment of the present application, a plurality of clothing recognition models are pre-stored in the washing machine, and the plurality of clothing recognition models correspond one-to-one to a plurality of sample image sets, and each clothing recognition model is trained based on a corresponding sample image set. Each sample image set may include a plurality of sample images, and the plurality of sample images may be images obtained by photographing a plurality of clothing of different colors, different materials, with dirt and / or without dirt.

[0074] After acquiring the captured image of the clothes to be washed, the washing machine can determine the similarity between the captured image and the sample images in the multiple sample sets. Then, based on the calculated similarity, at least one target recognition model can be determined from the multiple clothes recognition models. Among them, in the sample image set corresponding to each target recognition model, there is a sample image whose similarity with the captured image is greater than the similarity threshold.

[0075] Optionally, the multiple clothing recognition models and the multiple sample image sets may be pre-stored in the washing machine. Alternatively, the multiple sample image sets may be sent to the washing machine by a terminal (such as a mobile phone) in different time periods, and the multiple clothing recognition models are obtained by training the washing machine based on the multiple sample image sets. Alternatively, the multiple clothing recognition models and the multiple sample image sets may be sent to the washing machine by a server.

[0076] Step 103: input the captured image into at least one target recognition model to obtain feature information of the clothes to be washed.

[0077] In an embodiment of the present application, after determining at least one target recognition model, the washing machine can input a captured image of the laundry to be washed into the at least one target recognition model. The at least one target recognition model can analyze and process the captured image to obtain feature information of the laundry to be washed. The feature information includes at least one of material, color, and dirt information. The dirt information includes one of dirt and no dirt.

[0078] It can be understood that, since the sample image set corresponding to each target recognition model contains sample images whose similarity with the captured image is greater than the similarity threshold, the target recognition model can accurately recognize the feature information of the clothes to be washed in the captured image.

[0079] Step 104: Determine the detergent dosage information based on the characteristic information of the laundry to be washed.

[0080] In the embodiment of the present application, the washing machine pre-stores the correspondence between the characteristic information of the clothes and the information of the detergent. After the washing machine determines the characteristic information of the clothes to be washed, it can determine the information of the detergent required for the clothes to be washed during the washing process based on the correspondence. The information includes the type and amount of detergent.

[0081] Optionally, the types of detergents may include bleach, water softener, conditioner, enzyme-containing laundry detergent and enzyme-free laundry detergent, etc.

[0082] Step 105: Control the detergent storage box to put detergent into the washing drum according to the putting information.

[0083] After determining the dispensing information of the detergent required in the washing process of the clothes to be washed and starting the washing program, the washing machine can control its detergent storage box to dispense the detergent into the washing drum according to the dispensing information.

[0084] It is understandable that the washing machine may include a plurality of detergent storage boxes, and the types of detergents contained in the plurality of detergent storage boxes are different from each other. In addition, a flow sensor may be installed in each detergent storage box, and the flow sensor may detect the amount of detergent put into the detergent storage box. After the washing machine starts the washing program, the washing machine may control at least one target storage box among the plurality of detergent storage boxes to put detergent according to the amount in the putting information according to the putting information. The target storage box is used to hold the detergent indicated by the putting information.

[0085] It can also be understood that if the detergent dispensing device is a server, then in the above step 101, the server can obtain the captured image sent by the washing machine, and can determine the detergent dispensing information through the methods shown in the above steps 102 to 104. In the above step 105, the server can send the detergent dispensing information to the washing machine, so that the washing machine controls its detergent storage box to dispense detergent into the washing drum according to the dispensing information.

[0086] In summary, an embodiment of the present application provides a method for dispensing detergent. The method can determine at least one target recognition model from a plurality of clothing recognition models based on a captured image of the clothes to be washed, and use the at least one target recognition model to recognize the captured image, thereby determining the characteristic information of the clothes to be washed. Since there are sample images whose similarity with the captured image is greater than a similarity threshold in the sample image set corresponding to each target recognition model, and the target recognition model is trained based on a corresponding sample image set, the accuracy of the characteristic information of the clothes to be washed determined by the at least one target recognition model can be ensured. Furthermore, since the method can determine the detergent dispensing information based on the determined characteristic information, and control the detergent storage box to automatically dispense detergent into the washing drum according to the dispensing information, the efficiency and accuracy of detergent dispensing are effectively improved.

[0087] Figure 4 is a flow chart of another method for dispensing detergent provided in an embodiment of the present application, which method can be applied to a detergent dispensing device. The detergent dispensing device can be Figure 1 The server 110 or the washing machine 120 in the illustrated scenario. The following description will be made by taking the washing machine 120 as an example in which the detergent delivery device is used. Figure 4 , the method comprising:

[0088] Step 201: receiving multiple sample image sets uploaded by a terminal, and a recognition result of each sample image in each sample image set.

[0089] Among them, each sample image set may include multiple sample images, and the multiple sample images may be images obtained by photographing clothes of multiple different colors, different materials, with dirt and / or without dirt. The recognition result of each sample image may include at least one of the color, material and dirt information of the clothes in the photographed image. The dirt information may include one of dirt and no dirt. In an embodiment of the present application, the terminal can send the sample image and its recognition result to the washing machine through the washing machine management APP installed thereon. The washing machine can then receive the sample images and recognition results uploaded by multiple terminals.

[0090] It is understandable that each sample image set may include sample images uploaded by multiple terminals and received by the washing machine within a time period, and the multiple sample image sets may be received by the washing machine within different time periods.

[0091] Optionally, the number of sample images included in each sample image set may be greater than or equal to a quantity threshold, which may be 5000. That is, when the number of sample images received by the washing machine within a certain time period reaches the quantity threshold, the plurality of sample images received within the time period may be regarded as a sample image set.

[0092] Step 202: Use each sample image set and the recognition result to train the initial model, and obtain a clothing recognition model corresponding to each sample image set.

[0093] In an embodiment of the present application, an initial model is pre-stored in the washing machine. The initial model may be an image recognition model. For each acquired sample image set, the washing machine may train the initial model using the sample image set and the recognition result to obtain a clothing recognition model corresponding to the sample image set. Since the washing machine can acquire multiple sample image sets, multiple clothing recognition models may be trained. Furthermore, different clothing recognition models may learn the features of sample images in different sample image sets.

[0094] It is understandable that the total number of sample images included in the multiple sample image sets is relatively large. Therefore, if the multiple sample image sets are directly used to train the initial model, it will take a long training time to obtain the clothing recognition model. In the embodiment of the present application, since the number of sample images included in each sample image set is much smaller than the total number of sample images included in the multiple sample image sets, each image sample set is used to train the initial model separately, so that a corresponding clothing recognition model can be obtained in a shorter training time. As a result, the training efficiency of the clothing recognition model is effectively improved.

[0095] It can also be understood that after the washing machine obtains multiple clothing recognition models through model training, each clothing recognition model and the sample image set corresponding to the clothing recognition model can be stored accordingly.

[0096] Step 203: Determine the water quality of the area where the washing machine is located based on the location information of the washing machine.

[0097] In an embodiment of the present application, the washing machine has a positioning function and can determine its positioning information based on the positioning function. The washing machine may pre-store a correspondence between an area and water quality. The washing machine may determine the water quality of the area where it is located based on the correspondence. The water quality may be hard water or soft water.

[0098] Optionally, the washing machine may periodically obtain the location information of the washing machine. For example, the washing machine may obtain its own location information every three months.

[0099] Step 204: If it is determined that the water quality in the area where the washing machine is located is hard water, a prompt message is issued.

[0100] In the embodiment of the present application, if the washing machine determines that the water quality in the area where the washing machine is located is hard water based on the corresponding relationship between the area and the water quality, a prompt message may be issued. In addition, the washing machine may also send a prompt message to the terminal so that the terminal displays the prompt message. The prompt message is used to indicate the addition of a water softener.

[0101] Optionally, when the washing machine determines that the water quality in the area where it is located is hard water, it can first detect the detergent storage box in its detergent storage box for storing softener to determine whether the detergent storage box still stores softener. If the washing machine detects that there is no softener stored in the detergent storage box, or the storage amount of softener is less than the storage threshold, the prompt information can be displayed on its display screen. For example, refer to Figure 5 The prompt information displayed on the display screen of the washing machine may include: This area belongs to a hard water area, and it is recommended to add a water softener.

[0102] If the washing machine detects that softener is stored in the detergent storage box and the storage amount of softener is greater than the storage threshold, the operation information can be displayed on the display screen of the washing machine. Figure 6 The operation information displayed on the display screen of the washing machine may include: this area belongs to a hard water area, and water softener is automatically added during the washing process.

[0103] Furthermore, if the washing machine sends a prompt message and an operation message to the terminal, the schematic diagram of displaying the prompt message on the display screen of the terminal can refer to Figure 7 (a) in the figure, the schematic diagram showing the operation information can be referred to Figure 7 (b) in .

[0104] Step 205: Acquire an image of the laundry to be washed.

[0105] In an embodiment of the present application, the washing machine includes a camera, which can be located on a control panel of the washing machine. When a user puts the laundry to be washed into the washing drum, the camera can take a picture of the laundry to be washed to obtain a photographed image.

[0106] Optionally, the camera can take photos or videos of the laundry to be washed, thereby obtaining a photographed image of the laundry to be washed. Since the laundry to be washed is not in a static state during the process of being put into the washing machine, the washing machine can obtain multiple candidate images of each piece of laundry to be washed. Afterwards, the washing machine can select the image with the best image quality from the multiple candidate images as the photographed image of the laundry to be washed.

[0107] Step 206: Calculate the similarity between the captured image and the sample images in the plurality of sample image sets.

[0108] In the embodiment of the present application, the washing machine may pre-store an image similarity algorithm. When the washing machine obtains a captured image of the laundry to be washed, the image similarity algorithm may be used to calculate the similarity between the captured image and sample images in a plurality of sample image sets.

[0109] It is understandable that if the similarity between the captured image and the sample image is high, it can be determined that the captured object (ie, clothing) in the captured image is likely to be the same type of clothing as the captured image in the sample image.

[0110] Step 207: Determine at least one target recognition model from the multiple clothing recognition models based on the similarity.

[0111] In an embodiment of the present application, after calculating the similarity between the captured image and the sample images in the plurality of sample image sets, the washing machine can determine at least one target recognition model from the plurality of clothing recognition models based on the calculated similarity. Among them, in the sample image set corresponding to each target recognition model, there is a sample image whose similarity with the captured image is greater than a similarity threshold. For example, if the image similarity ranges from 0 to 1, the similarity threshold can be 0.5.

[0112] Step 208: input the captured image into at least one target recognition model to obtain a recognition result output by each target recognition model.

[0113] In an embodiment of the present application, after determining at least one target recognition model, the washing machine can input a captured image of the laundry to be washed into the at least one target recognition model. The at least one target recognition model can analyze and process the captured image and output a recognition result of the laundry to be washed.

[0114] It can be understood that, since the sample image set corresponding to each target recognition model contains sample images whose similarity with the captured image is greater than the similarity threshold, the target recognition model can accurately recognize the captured image and output a recognition result.

[0115] The recognition result may include the probability that the laundry to be washed meets each of the multiple candidate feature information. The multiple candidate feature information may include material, color and dirt information. Each candidate feature information may include multiple feature values. The probability of each candidate feature information may include the probability of multiple feature values ​​in the candidate feature information.

[0116] For example, the multiple feature values ​​included in the material in the candidate feature information can be: chemical fiber, pure cotton and polyester cotton. The multiple feature values ​​included in the color in the candidate feature information can be: white, light color and dark color. The feature values ​​included in the dirt information in the candidate feature information can be: with dirt and without dirt.

[0117] For a photographed image T of a piece of clothing to be washed that is made of pure cotton, is white in color, and has dirt, if the photographed image T of the piece of clothing to be washed is input into the target recognition model M1, the recognition results output by the target recognition model M1 may include: the probability that the material of the clothing to be washed is chemical fiber is 10%, the probability that the material is pure cotton is 80%, and the probability that the material is polyester-cotton is 10%; the probability that the color of the clothing to be washed is white is 75%, the probability that the color is light-colored is 20%, and the probability that the color is dark-colored is 5%; the probability that the clothing to be washed has dirt is 85%, and the probability that it has no dirt is 15%.

[0118] Step 209: Determine the candidate feature information with the highest probability among the recognition results output by at least one target recognition model as the feature information of the clothes to be washed.

[0119] In the embodiment of the present application, after the washing machine obtains the recognition result output by at least one target recognition model, it can determine the candidate feature information with the highest probability in the result output by the at least one target recognition model as the feature information of the clothes to be washed. That is, for the same candidate feature information in the recognition result output by the at least one target model, the washing machine can determine the feature value with the highest probability in the candidate feature information as the feature information of the clothes to be washed.

[0120] For example, for the photographed image T of the laundry to be washed mentioned above, if the photographed image T of the laundry to be washed is input into the target recognition models M1 and M2, the recognition results output by the target recognition models M1 and M2 may be as shown in Table 1.

[0121] Table 1

[0122]

[0123] Based on the above Table 1, it can be seen that for the material in the candidate feature information, the probability that the material output by the target recognition model M2 is pure cotton is the highest. For the color in the candidate feature information, the probability that the color output by the target recognition model M1 is white is the highest. For the dirt information in the candidate feature information, the probability that the dirt information output by the target recognition model M2 is dirt is the highest. Therefore, the washing machine can identify the material in the feature information of the piece of clothing T to be washed as pure cotton, the color as white, and the dirt information as dirt.

[0124] Step 210: If there is no sample image in the plurality of sample image sets whose similarity with the captured image is greater than a similarity threshold, the captured image is input into a reference recognition model to obtain feature information of the clothes to be washed.

[0125] The reference recognition model may be obtained by training the initial model using a newly acquired sample image set. That is, the reference recognition model may be a model trained after multiple clothing recognition models and stored in the washing machine.

[0126] In the above step 207, if the washing machine detects that there is no sample image with a similarity greater than a similarity threshold with the captured image in the multiple sample image sets corresponding to the multiple clothing recognition models, the captured image can be input into the reference recognition model. The reference recognition model can analyze and process the captured image and obtain feature information of the clothing to be washed.

[0127] It is understandable that in the above step 207, the washing machine first calculates the similarity between the captured image and the sample images in the plurality of sample image sets, but does not calculate the similarity between the captured image and the sample images in the sample image set corresponding to the reference recognition model, thereby improving the calculation efficiency of the washing machine. When the washing machine determines that there is no sample image in the plurality of sample image sets whose similarity with the captured image is greater than the similarity threshold, the captured image is then input into the reference recognition model for processing, thereby ensuring that the washing machine can determine the characteristic information of the clothes to be washed.

[0128] It is also understandable that in the above step 205, the washing machine can obtain one or more images of the laundry to be washed. For each image of the laundry to be washed, the washing machine can determine the characteristic information of the laundry to be washed by the method shown in the above steps 206 to 210.

[0129] Step 211: If, based on the characteristic information of the laundry to be washed, it is detected that the laundry to be washed includes a first type of laundry of a target material, it is determined that the type of detergent includes a laundry softener.

[0130] In an embodiment of the present application, the washing machine pre-stores a correspondence between characteristic information of the clothes and information on the placement of detergents. After the washing machine determines the characteristic information of the clothes to be washed in the above step 209 or step 210, it can determine the placement information of the detergents required for the clothes to be washed during the washing process based on the correspondence. The placement information may include the type and amount of detergent. Optionally, the types of detergents may include bleach, water softener, softener, enzyme-containing laundry detergent, enzyme-free laundry detergent, etc.

[0131] If the washing machine detects that the first type of clothes including the target material are included in the clothes to be washed based on the characteristic information of the clothes to be washed, the detergent dispensing information corresponding to the target material can be determined based on the correspondence between the characteristic information and the detergent dispensing information. Optionally, the target material can include at least one of chemical fiber, cotton, polyester-cotton, wool, acrylic, polyester, silk, linen, denim, polyester, rayon and viscose. Correspondingly, the type of detergent corresponding to the target material can be a fabric softener.

[0132] It is understandable that by adding fabric softener during the washing process of the first type of clothes made of the target material, the static electricity of the first type of clothes can be effectively reduced and pilling of the first type of clothes can be avoided.

[0133] Step 212: Determine a first dosage of fabric softener based on the total volume of the first category of clothing.

[0134] In an embodiment of the present application, the washing machine can also determine the volume of each piece of clothing to be washed based on the captured image of the clothing to be washed, and then determine the total volume of the first category of clothing. Afterwards, the washing machine can determine the first amount of fabric softener to be dispensed based on the total volume of the first category of clothing. The first amount of fabric softener to be dispensed can be positively correlated with the total volume of the first category of clothing. The volume of each piece of clothing to be washed can be determined based on the area and thickness of the clothing to be washed.

[0135] Optionally, after the washing machine obtains the captured image, it can determine the area of ​​the laundry to be washed based on the captured image of the laundry to be washed. Wherein, when the shooting distance of the camera is fixed, the washing machine can determine the actual area of ​​the laundry to be washed based on the imaging area of ​​the laundry to be washed in the captured image and the shooting distance.

[0136] As an optional implementation, the camera may be a monocular camera, and for an image of the laundry captured by the monocular camera, the washing machine may identify the front piece and the back piece of the laundry from the captured image. Afterwards, the washing machine may determine a first distance between the front piece of the laundry and the monocular camera, and a second distance between the back piece of the laundry and the monocular camera. The absolute value of the difference between the first distance and the second distance is the thickness of the laundry.

[0137] As another optional implementation, the washing machine may directly determine the pre-stored thickness value as the thickness of the laundry to be washed. For example, the thickness threshold may be 2 mm.

[0138] After the washing machine determines the area and thickness of each piece of clothing to be washed, it can calculate the volume of the clothing to be washed. The volume V' of the clothing to be washed can satisfy:

[0139] V`=S*TH;

[0140] Wherein, S is the area of ​​the laundry to be washed, and TH is the thickness of the laundry to be washed.

[0141] Optionally, the first dosage V of the fabric softener is 1 Can satisfy:

[0142] V 1 =V 01 *ρ 1 *K 1 ;

[0143] Among them, V 01 is the total volume of the first type of clothing, ρ 1 K is the density of the first type of clothing. 1is the preset first dosage coefficient. The first dosage coefficient may be the dosage of fabric softener required for the first type of clothing per unit weight. If the unit of weight is 1 kilogram (kg) and the unit of dosage is milliliter (ml), K 1 The value of V is 20, then it can be determined that the amount of fabric softener required to wash 1 kg of the first type of clothing is 20 ml. It can also be understood that the product of the volume of a substance and its density is the weight of the object. Therefore, V 01 *ρ 1 The calculation result is the total weight of the first category of clothing.

[0144] It can be understood that if the laundry to be washed includes a plurality of first-category laundry, the density ρ 1 The density of each first-category clothing may refer to the weight of the first-category clothing per unit volume, and the density is related to the material of the clothing.

[0145] It can also be understood that in order to ensure that the washing machine accurately obtains the characteristic information and volume of the clothes to be washed, the user can unfold each piece of clothes to be washed at a certain distance from the camera during the process of putting the clothes to be washed for the camera to take pictures.

[0146] Step 213: If, based on the characteristic information of the laundry to be washed, it is detected that the laundry to be washed includes the second type of laundry of the target color, it is determined that the type of detergent includes laundry bleach.

[0147] In the process of determining the information of dispensing detergent based on the characteristic information of the laundry to be washed, if the laundry to be washed includes the second type of laundry of the target color, it can be determined that the type of detergent includes laundry bleach. The target color can include at least white, or can also include gray, silvery white and other colors that are close to white.

[0148] It is understandable that clothing bleach has strong sterilization, deodorization and bleaching functions. Therefore, when the plurality of clothes to be washed include the second type of clothes of the target color, clothing bleach can be added during the washing process of the plurality of clothes to be washed to ensure a better washing effect for the second type of clothes in the plurality of clothes to be washed.

[0149] Step 214: Determine a second dosage of the clothing bleach based on the total volume of the second type of clothing.

[0150] In an embodiment of the present application, the washing machine can determine the total volume of the second category of clothes after determining the volume of each piece of clothing in the laundry to be washed. Then, the washing machine can determine the second dosage of clothing bleach based on the total volume of the second category of clothes. The second dosage can be positively correlated with the total volume of the second category of clothes.

[0151] Optionally, the second dosage V of the clothing bleaching agent is 2 Can meet: V 2 =V 02 *ρ 2 *K 2 ;

[0152] Among them, V 02 is the total volume of the second type of clothing, ρ 2 is the density of the second type of clothing, K 2 is the preset second dosage coefficient. The second dosage coefficient may be the dosage of the clothing bleach required for the second type of clothing per unit weight. For example, if the unit weight is 1 kg and the dosage unit is ml, K 2 The value of V is 10, then it can be determined that the amount of clothing bleach required to wash 1 kg of the second type of clothing is 10 ml. 02 *ρ 2 The result of the calculation is the total weight of the second category of clothing.

[0153] It can be understood that if the laundry to be washed includes a plurality of second-category laundry, the density ρ 2 It may be the average density of the plurality of second-category clothes. After determining the plurality of second-category clothes in the clothes to be washed, the washing machine may determine the average density of the materials corresponding to the plurality of second-category clothes as the density of the second-category clothes.

[0154] Step 215: If it is detected that the plurality of laundry items to be washed include laundry items of the third type with dirt based on the characteristic information of the laundry items to be washed, it is determined that the type of detergent includes laundry detergent containing enzymes.

[0155] In the process of determining the detergent delivery information based on the characteristic information of the laundry to be washed, if the washing machine detects that the laundry to be washed includes the third type of laundry with dirt, it can be determined that the type of detergent in the delivery information can include enzyme-containing laundry liquid. The dirt can include at least one of oil stains, blood stains, sweat stains, chocolate stains and sweat stains.

[0156] It is understandable that the components of dirt are generally proteins. The enzymes in the enzyme-containing laundry detergent have the function of decomposing proteins. Therefore, the use of enzyme-containing laundry detergent to wash clothes can remove the dirt in the third category of clothes.

[0157] Step 216: If it is detected based on the characteristic information of the laundry to be washed that the third type of laundry with dirt is not included in the plurality of laundry to be washed, it is determined that the type of detergent includes enzyme-free laundry liquid.

[0158] When the washing machine determines the detergent dispensing information based on the characteristic information of the laundry to be washed, if it is detected that the third type of laundry with dirt is not included in the multiple laundry to be washed, it can be determined that the type of detergent in the dispensing information may include enzyme-free laundry detergent.

[0159] Step 217: Determine a third dosage of laundry detergent based on the total weight of the laundry to be washed.

[0160] In an embodiment of the present application, the washing machine can detect the total weight of the clothes to be washed in the washing drum after the user puts the clothes to be washed into the washing drum. Optionally, the washing machine may also include a weight sensor, by which the washing machine can detect the total weight of the clothes to be washed in the washing drum. Afterwards, the washing machine can send the total weight of the clothes to be washed to the washing machine together with the captured image of the clothes to be washed. Among them, if the washing machine determines that the type of detergent includes enzyme-containing laundry detergent, it can determine the third amount of enzyme-containing laundry detergent based on the total weight of the clothes to be washed. If the washing machine determines that the type of detergent includes enzyme-free laundry detergent, it can determine the third amount of enzyme-free laundry detergent based on the total weight of the clothes to be washed. The third amount of enzyme-containing or enzyme-free laundry detergent may be positively correlated with the total weight of the clothes to be washed.

[0161] Step 218: Control the detergent storage box to put detergent into the washing drum according to the putting information.

[0162] After determining the information of the detergent required for the washing process of the laundry, the washing machine can send the information to the washing machine. The washing machine can respond to the information and control its detergent storage box to put detergent into the washing drum according to the information after the washing machine starts the washing program.

[0163] It is understandable that the washing machine may include a plurality of detergent storage boxes, and the types of detergents contained in the plurality of detergent storage boxes are different from each other. In addition, a flow sensor may be installed in each detergent storage box, and the flow sensor may detect the amount of detergent put into the detergent storage box. After the washing machine starts the washing program, the washing machine may control at least one target storage box among the plurality of detergent storage boxes to put detergent according to the amount in the putting information according to the putting information. The target storage box is a storage box for the detergent indicated by the putting information.

[0164] Optionally, the washing machine can also display the dispensing information on its display screen, so that the user can view the type and amount of detergent dispensed by the washing machine during the washing process. Figure 8 The content of the dispensing information displayed on the display screen 121 of the washing machine may include: fabric softener: 30 ml, fabric bleach: 10 ml, and enzyme-containing laundry detergent 60 ml.

[0165] It is understandable that the sequence of the steps of the method for dispensing detergent provided in the embodiment of the present application can be appropriately adjusted, and the steps can also be increased or decreased accordingly according to the situation. For example, step 203 and step 204 can be deleted according to the situation. That is, the washing machine does not need to determine the water quality in its area. Alternatively, step 210 can be performed before step 208. Alternatively, step 211, step 213 and step 215 can be performed synchronously. Alternatively, step 211, step 213 and step 216 can be performed synchronously. Any technician familiar with the technical field can easily think of the method of change within the technical scope disclosed in this application, which should be covered within the scope of protection of this application, so it will not be repeated.

[0166] It can also be understood that some or all of the steps in the above-mentioned detergent delivery method can also be executed by the server. For example, the server can directly execute the method shown in steps 201 to 218 above. Alternatively, the server can only execute steps 201 and 202 above, and can send the trained multiple clothing recognition models and the corresponding multiple sample image sets to the washing machine, and the washing machine can then execute the method shown in steps 205 to 218 above. Alternatively, the washing machine can send the acquired photographed images of the shifted clothes to be washed to the server, and the server can execute the method shown in steps 206 to 210 above. After the server determines the characteristic information of the clothes to be washed, it can send the characteristic information to the washing machine, and the washing machine can then execute the method shown in steps 211 to 218 above.

[0167] In summary, an embodiment of the present application provides a method for dispensing detergent. The method can determine at least one target recognition model from a plurality of clothing recognition models based on a captured image of the clothes to be washed, and use the at least one target recognition model to recognize the captured image, thereby determining the characteristic information of the clothes to be washed. Since there are sample images whose similarity with the captured image is greater than a similarity threshold in the sample image set corresponding to each target recognition model, and the target recognition model is trained based on a corresponding sample image set, it is possible to ensure the accuracy of the characteristic information of the clothes to be washed determined by the at least one target recognition model. Afterwards, the method can determine the detergent dispensing information based on the characteristic information. As a result, the washing machine can automatically dispense detergent into the washing drum based on the dispensing information, effectively improving the efficiency of dispensing detergent and ensuring the accuracy of dispensing detergent.

[0168] Fig. 9 A structural diagram of a washing machine provided in an embodiment of the present application can execute the detergent delivery method provided in the above method embodiment. Figure 1 The washing machine in the scene. Fig. 9 The washing machine 120 includes: a camera 121, a washing drum 122, a controller 123 and a detergent storage box 124.

[0169] The camera 121 is used to obtain an image of the laundry to be washed, where the laundry to be washed is the laundry put into the washing drum 122 .

[0170] The controller 123 is used to:

[0171] Based on the captured image, at least one target recognition model is determined from multiple clothing recognition models, wherein the multiple clothing recognition models correspond one-to-one to multiple sample image sets, each clothing recognition model is trained based on a corresponding sample image set, and in the sample image set corresponding to each target recognition model, there exists a sample image whose similarity with the captured image is greater than a similarity threshold.

[0172] The captured image is input into at least one target recognition model to obtain feature information of the clothes to be washed, wherein the feature information includes at least one of material, color and dirt information.

[0173] Based on the characteristic information of the laundry to be washed, the detergent placement information is determined, and the placement information includes the type and amount of detergent.

[0174] The detergent storage box 124 is controlled to put detergent into the washing tub 122 according to the putting information.

[0175] Optionally, the controller 123 is used to:

[0176] The captured image is input into at least one target recognition model to obtain a recognition result output by each target recognition model, wherein the recognition result includes a probability that the laundry to be washed conforms to each candidate feature information among a plurality of candidate feature information.

[0177] The candidate feature information with the highest probability among the recognition results output by at least one target recognition model is determined as the feature information of the clothes to be washed.

[0178] Optionally, the controller 123 is used to:

[0179] The similarity between the captured image and the sample images in the plurality of sample image sets is calculated.

[0180] Based on the similarity, at least one target recognition model is determined from the plurality of clothing recognition models.

[0181] Optionally, the controller 133 is further configured to:

[0182] The receiving terminal uploads multiple sample image sets and the recognition result of each sample image in each sample image set.

[0183] Each sample image set and the recognition result are used to train the initial model respectively, and a clothing recognition model corresponding to each sample image set is obtained.

[0184] Optionally, the controller 123 is further configured to input the captured image into a reference recognition model to obtain feature information of the laundry to be washed if there is no sample image in the plurality of sample image sets whose similarity to the captured image is greater than a similarity threshold. The reference recognition model is a model trained after the plurality of laundry recognition models.

[0185] Optionally, the controller 123 is used to:

[0186] If, based on the characteristic information of the laundry to be washed, it is detected that the laundry to be washed includes a first type of laundry of the target material, it is determined that the type of detergent includes a laundry softener.

[0187] If the second type of clothes including the target color in the clothes to be washed is detected based on the characteristic information of the clothes to be washed, it is determined that the type of detergent includes clothes bleach.

[0188] If it is detected that the plurality of laundry items to be washed include laundry items of a third type with dirt based on the characteristic information of the laundry items to be washed, it is determined that the type of detergent includes laundry detergent containing enzymes.

[0189] If it is detected based on the characteristic information of the laundry to be washed that the third type of laundry with dirt is not included in the plurality of laundry to be washed, it is determined that the type of detergent includes laundry liquid without enzyme.

[0190] Optionally, the controller 123 is further configured to:

[0191] If the types of detergents include fabric softener, a first dosage of the fabric softener is determined based on the total volume of the first category of clothes, and the first dosage is positively correlated with the total volume of the first category of clothes.

[0192] If the types of detergents include clothing bleach, a second dosage of the clothing bleach is determined based on the total volume of the second category of clothing, and the second dosage is positively correlated with the total volume of the second category of clothing.

[0193] If the types of detergents include laundry liquid, a third dosage of the laundry liquid is determined based on the total weight of the laundry to be washed, and the third dosage is positively correlated with the total weight of the laundry to be washed.

[0194] Optionally, the first dosage V of the fabric softener is 1 Satisfaction: V 1 =V 01 *ρ 1 *K 1 ;

[0195] Among them, V 01 is the total volume of the first type of clothing, ρ 1 is the density of the first type of clothing, K 1 It is the preset first delivery coefficient.

[0196] The second dosage V of the clothing bleaching agent 2 Satisfaction: V 2 =V 02 *ρ 2 *K 2 ;

[0197] Among them, V 02 is the total volume of the second type of clothing, ρ 2 is the density of the second type of clothing, K 2 It is the preset second delivery coefficient.

[0198] Optionally, the controller 123 is further configured to:

[0199] Based on the location information of the washing machine, the water quality of the area where the washing machine is located is determined.

[0200] If it is determined that the water quality in the area where the washing machine is located is hard water, a prompt message is issued, and the prompt message is used to instruct the addition of a water softener.

[0201] In summary, an embodiment of the present application provides a washing machine. The washing machine can determine at least one target recognition model from a plurality of clothing recognition models based on a captured image of the clothes to be washed, and use the at least one target recognition model to recognize the captured image, thereby determining the characteristic information of the clothes to be washed. Since there are sample images whose similarity with the captured image is greater than a similarity threshold in the sample image set corresponding to each target recognition model, and the target recognition model is trained based on a corresponding sample image set, the accuracy of the characteristic information of the clothes to be washed determined by the at least one target recognition model can be ensured. Furthermore, since the washing machine can determine the detergent delivery information based on the determined characteristic information, and control the detergent storage box to automatically deliver detergent to the washing drum according to the delivery information, the efficiency and accuracy of detergent delivery are effectively improved.

[0202] It can be understood that the washing machine provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the detection device can be divided into different functional modules to complete all or part of the functions described above.

[0203] In addition, the washing machine and detergent dispensing method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0204] The present application also provides a washing machine, such as Fig.10 As shown, the washing machine may include a processor 310 and a memory 320, wherein the memory 320 stores instructions, which are loaded and executed by the processor 310 to implement the detergent delivery method provided by the above method embodiment, for example Figure 3 or Figure 4 The method shown.

[0205] The present application embodiment provides a computer-readable storage medium, in which a computer program is stored. The computer program is loaded by a processor and executes the detergent delivery method provided in the above embodiment, for example Figure 3 or Figure 4 The method shown.

[0206] The present application also provides a computer program product including instructions. When the computer program product is run on a computer, the computer executes the detergent delivery method provided by the above method embodiment, for example: Figure 3 or Figure 4 The method shown.

[0207] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0208] It is to be understood that the term "at least one" in the present application means one or more, and the term "plurality" means two or more.

[0209] The "and / or" mentioned in this article indicates that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship.

[0210] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with basically the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on quantity and execution order.

[0211] The above description is only an exemplary embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A washing machine, characterized in that: The washing machine includes: a camera, a washing drum, a controller and a detergent storage box: The camera is used to photograph the laundry to be washed to obtain a photographed image, wherein the laundry to be washed is the laundry put into the washing drum; The controller is used to: Calculating the similarity between the captured image and sample images in a plurality of sample image sets, and determining at least one target recognition model from a plurality of clothing recognition models based on the similarity, wherein the plurality of clothing recognition models correspond to the plurality of sample image sets one by one, each clothing recognition model is trained based on a corresponding sample image set, and the sample image set corresponding to each target recognition model contains a sample image whose similarity with the captured image is greater than a similarity threshold; Inputting the captured image into the at least one target recognition model to obtain feature information of the laundry to be washed, wherein the feature information includes at least one of material, color and dirt information; Determining detergent delivery information based on the characteristic information of the laundry to be washed, wherein the delivery information includes the type and delivery amount of the detergent; Controlling the detergent storage box to dispense detergent into the washing drum according to the dispensing information; The controller is also used for: The plurality of sample image sets uploaded by the receiving terminal, and the recognition result of each sample image in each of the sample image sets; Each of the sample image sets and the recognition results is used to train the initial model to obtain a clothing recognition model corresponding to each of the sample image sets.

2. The washing machine according to claim 1, characterized in that: The controller is used to: Inputting the captured image into the at least one target recognition model to obtain a recognition result output by each target recognition model, wherein the recognition result includes a probability that the laundry to be washed meets each of a plurality of candidate feature information; The candidate feature information with the highest probability in the recognition result output by the at least one target recognition model is determined as the feature information of the clothes to be washed.

3. The washing machine according to claim 1, characterized in that: The controller is also used for: If there is no sample image in the plurality of sample image sets whose similarity with the photographed image is greater than the similarity threshold, inputting the photographed image into a reference recognition model to obtain feature information of the laundry to be washed; The reference recognition model is a model trained after the multiple clothing recognition models.

4. The washing machine according to any one of claims 1 to 3, characterized in that: The controller is used to: If, based on the characteristic information of the laundry to be washed, it is detected that the laundry to be washed includes a first type of laundry of a target material, determining that the type of detergent includes a laundry softener; If it is detected that the laundry to be washed includes a second type of laundry of a target color, based on the characteristic information of the laundry to be washed, determining that the type of detergent includes a laundry bleach; If it is detected based on the characteristic information of the laundry to be washed that the plurality of laundry to be washed include a third type of laundry with dirt, then it is determined that the type of the detergent includes a laundry detergent containing enzymes; If it is detected based on the characteristic information of the laundry to be washed that the plurality of laundry to be washed do not include the third type of laundry with dirt, it is determined that the type of the detergent includes laundry liquid without enzyme.

5. The washing machine according to claim 4, characterized in that: The controller is also used for: If the type of detergent includes fabric softener, determining a first dosage of the fabric softener based on the total volume of the first type of clothing, the first dosage being positively correlated with the total volume of the first type of clothing; If the type of the detergent includes a clothing bleach, then based on the total volume of the second type of clothing, determining a second dosage of the clothing bleach, the second dosage being positively correlated with the total volume of the second type of clothing; If the type of detergent includes laundry liquid, a third dosage of the laundry liquid is determined based on the total weight of the laundry to be washed, and the third dosage is positively correlated with the total weight of the laundry to be washed.

6. The washing machine according to claim 5, characterized in that: The first dosage of the fabric softener satisfy: ; in, is the total volume of the first type of clothing, is the density of the first type of clothing, is the preset first delivery coefficient; The second dosage of the clothing bleaching agent satisfy: ; in, is the total volume of the second type of clothing, is the density of the second type of clothing, It is the preset second delivery coefficient.

7. The washing machine according to any one of claims 1 to 3, characterized in that: The controller is also used for: Determining water quality in an area where the washing machine is located based on the location information of the washing machine; If it is determined that the water quality in the area where the washing machine is located is hard water, a prompt message is issued, and the prompt message is used to instruct the addition of a water softener.

8. A method for dispensing a detergent, characterized in that: The method comprises: Acquire a photographed image obtained by photographing clothes to be washed, wherein the clothes to be washed are clothes put into a washing drum of a washing machine; Calculating the similarity between the captured image and sample images in a plurality of sample image sets, and determining at least one target recognition model from a plurality of clothing recognition models based on the similarity, wherein the plurality of clothing recognition models correspond to the plurality of sample image sets one by one, each clothing recognition model is trained based on a corresponding sample image set, and the sample image set corresponding to each target recognition model contains a sample image whose similarity with the captured image is greater than a similarity threshold; Inputting the captured image into the at least one target recognition model to obtain feature information of the laundry to be washed, wherein the feature information includes at least one of material, color and dirt information; Determining detergent delivery information based on the characteristic information of the laundry to be washed, wherein the delivery information includes the type and delivery amount of the detergent; Controlling the detergent storage box of the washing machine to dispense detergent into the washing drum according to the dispensing information; The method further comprises: The plurality of sample image sets uploaded by the receiving terminal, and the recognition result of each sample image in each of the sample image sets; Each of the sample image sets and the recognition results is used to train the initial model to obtain a clothing recognition model corresponding to each of the sample image sets.

Images