Control method for a smart home system and smart home system

Abstract

The present invention relates to the field of smart home technology, and specifically provides a control method and a smart home system for a smart home system, and aims to solve the problem that the conventional smart home system cannot reasonably coordinate clothes and recommend a washing program. Therefore, the control method of the present invention includes the steps of: a household appliance obtains an image of current clothes; a household appliance transmits the image of the current clothes to a server; a server obtains an ID account of the household appliance that transmits the image of the current clothes; a server determines the identity of the household appliance that transmits the image of the current clothes based on the ID account; a server determines a target demand based on the identity; and a server selectively recommends a washing program or a clothing coordination plan based on the image of the current clothes based on the target demand, so as to not only recommend a reasonable and accurate washing program, but also recommend a reasonable clothing coordination plan to the user, and achieve an optimal or relatively good coordination effect, so as to better meet the different needs of users and improve the user experience.

Description

[Technical field]

[0001] The present invention relates to the field of smart home technology, and in particular provides a control method for a smart home system and a smart home system. [Background technology]

[0002] A smart home uses the home as a platform to connect various household devices using Internet of Things technology to create a smart living environment, build an efficient management system for housing facilities and household schedule affairs, and improve the safety, convenience, and comfort of home life.

[0003] In home life, with the improvement of living standards, people pay more and more attention to managing their personal image, including the demand for high-quality dressing, for example, different clothing coordination for different occasions, etc. Therefore, the types and designs of clothes owned by users are becoming richer and richer, and various coordination is possible, but in many cases, users do not know well how to coordinate clothes to achieve an optimal or relatively good coordination effect, and there are too many types and designs of clothes, so how to wash different clothes to avoid damage to clothes has become a problem that troubles users.

[0004] Therefore, the field needs a new control method for a smart home system and a smart home system to solve the above problems. Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention aims to solve the above technical problem, that is, to solve the problem that the conventional smart home system cannot rationally coordinate clothes and recommend a washing program. [Means for solving the problem]

[0006] According to a first aspect, the present invention provides a control method for a smart home system, the smart home system including a household device and a server, the household device can communicate with the server, the control method includes steps of the household device obtaining a current clothing image, the household device sending the current clothing image to the server, the server obtaining an ID account of the household device that sent the current clothing image based on the ID account, the server determining an identity of the household device that sent the current clothing image based on the ID account, the server determining a target demand based on the identity, and based on the target demand, the server selectively recommending a washing program or a clothing coordination idea based on the current clothing image.

[0007] In a preferred technical solution of the above control method, the step of "the server determines a target demand based on the identity" specifically includes a step of determining that the target demand is washing clothes if the ID account of the household device is an ID account of a first type device, and the step of "based on the target demand, the server selectively recommends a washing program or a clothing coordination plan based on the current clothing image" specifically includes a step of the server recommending a washing program based on the current clothing image if the target demand is washing clothes.

[0008] In a preferred technical solution of the above control method, the step of "the server determines a target demand based on the identity" further includes a step of determining that the target demand is clothing coordination advice if the ID account of the household device is an ID account of a second type device, and the step of "based on the target demand, the server selectively recommends a washing program or a clothing coordination suggestion based on the current clothing image" further includes a step of the server recommending a clothing coordination suggestion based on the current clothing image if the target demand is clothing coordination advice.

[0009] In a preferred technical solution of the above control method, the step of "the server determines a target demand based on the identity" further includes a step of the household device obtaining a control command if the ID account of the household device is an ID account of a third type device, and a step of the household device sending the control command to the server, and the step of "based on the target demand, the server selectively recommends a washing program or a clothing coordination suggestion based on the current clothing image" further includes a step of the server determining whether the content of the control command is a recommendation of a washing program, and if the content of the control command is a recommendation of a washing program, the server recommending a washing program based on the current clothing image, and / or if the content of the control command is a recommendation of a clothing coordination suggestion, the server recommending a clothing coordination suggestion based on the current clothing image.

[0010] In a preferred technical solution of the above control method, the server has a clothing coordination database and a clothing private database, and the step of "the server recommends clothing coordination suggestions based on the current clothing image" specifically includes the steps of the server pre-processing the current clothing image to obtain a first body image of the current clothing, the server determining whether tops and bottoms exist simultaneously in the first body image, and if tops and bottoms exist simultaneously in the first body image, the server determining whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database, and / or if only tops or bottoms exist in the first body image, the server recommends target coordination suggestions based on the first body image, the clothing coordination database and the clothing private database.

[0011] In a preferred technical proposal of the above control method, the step of "the server determining whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database" specifically includes the steps of the server determining a first compatibility degree of tops and bottoms based on the first body image, the server determining a second compatibility degree of tops and bottoms based on the first body image and the clothing coordination database, the server determining a total compatibility degree based on the first compatibility degree and the second compatibility degree, and the server determining whether the current clothing coordination is reasonable based on the total compatibility degree.

[0012] In a preferred technical proposal of the above control method, the step of "the server determining a first compatibility degree between the top and the bottom based on the first body image" specifically includes the steps of the server determining the color of the top and the color of the bottom based on the first body image, judging whether the color of the top and the color of the bottom conform to a preset rule, and determining the first compatibility degree of the color of the top and the color of the bottom based on the judgment result.

[0013] In a preferred technical proposal of the above control method, the step of "the server determines a second compatibility degree of the top and the bottom based on the first body image and the clothing coordination database" specifically includes the steps of the server determining a current top image and a current bottom image based on the first body image, the server calculating a first similarity degree between the current top image and each standard top image stored in the clothing coordination database, or the current bottom image and each standard bottom image stored in the clothing coordination database, the server determining a target bottom image from the clothing coordination database based on each of the first similarities, or determining a target top image from the clothing coordination database based on each of the first similarities, the server calculating a second similarity degree between each of the target bottom images and the current bottom image, or each of the target top images and the current top image, and the server calculating the second compatibility degree based on each of the second similarities and the corresponding first similarity degree.

[0014] In a preferred technical solution of the above control method, the second compatibility is calculated by the method F2=D1×D2, where F2 is the second compatibility, D1 is the first similarity, and D2 is the second similarity.

[0015] In a preferred technical solution of the above control method, the step of "the server determining a total fitness based on the first fitness and the second fitness" specifically includes a step of the server calculating the total fitness based on the first fitness and each of the second fitnesses respectively.

[0016] In a preferred technical solution of the above control method, the total fitness is calculated by the method F0=F1×k1+F2×k2, where F0 is the total fitness, F1 is the first fitness, F2 is the second fitness, k1 is the first weight value, and k2 is the second weight value.

[0017] In a preferred technical proposal of the above control method, the step of "the server judges whether the current clothing coordination is reasonable based on the total suitability" specifically includes the steps of the server comparing all the total suitability and determining a maximum suitability, judging whether the maximum suitability is greater than a preset suitability, if the maximum suitability is greater than the preset suitability, judging that the current clothing coordination is reasonable, and the server sending the judgment result to the home appliance, thereby displaying "the current clothing coordination is reasonable" on the home appliance, and / or if the maximum suitability is less than or equal to the preset suitability, judging that the current clothing coordination is not reasonable, and the server recommending a target coordination suggestion based on the first body image, the clothing coordination database and the clothing private database.

[0018] In a preferred technical solution of the above control method, the step of "the server recommends a target coordination plan based on the first body image, the clothing coordination database and the clothing private database" specifically includes the steps of: the server matching a plurality of candidate clothing coordination plans from the clothing coordination database according to a preset method based on the color of current clothing; the server calculating a third similarity between a candidate clothing image corresponding to each of the candidate clothing coordination plans and a clothing image owned by each user stored in the clothing private database; determining whether the third similarity is greater than a first preset similarity; if the third similarity is greater than the first preset similarity, the server determining the candidate clothing coordination plan corresponding to the candidate clothing image as the target coordination plan and sending the target coordination plan to the household device to display the target coordination plan on the household device; and / or if the third similarity is equal to or less than the first preset similarity, not determining the candidate clothing coordination plan corresponding to the candidate clothing image as the target coordination plan.

[0019] In a preferred technical proposal of the above control method, the server further has a clothing image database, and the step of "the server recommends a washing program based on the image of the current clothing" specifically includes the steps of the server pre-processing the image of the current clothing to obtain a second body image of the current clothing, the server determining attribute information based on the second body image, the server comparing the second body image with the historical clothing images stored in the clothing image database, and the server judging based on the comparison result whether the current clothing is one of the historical clothing, if the current clothing is one of the historical washing clothing, the server determining a washing program based on the attribute information and the historical washing data of the current clothing and sending the washing program to the household appliance, and / or if the current clothing is not one of the historical washing clothing, the server determining a washing program based on the attribute information and sending the washing program to the household appliance.

[0020] In a preferred technical proposal of the above control method, the step of "the server compares the second body image with the historical clothing images stored in the clothing image database" specifically includes a step of the server calculating a fourth similarity between the second body image and each of the historical clothing images, and the step of "the server determining whether the current clothing is one of the historical clothing based on the comparison result" specifically includes a step of determining whether the fourth similarity is greater than a second preset similarity, a step of determining that the current clothing is one of the historical clothing if the fourth similarity is greater than the second preset similarity, and / or a step of determining that the current clothing is not one of the historical clothing if the fourth similarity is less than or equal to the second preset similarity.

[0021] According to a second aspect, the present invention provides a control method for a smart home system, the smart home system including a household device and a server, the household device can communicate with the server, and the control method includes a step of the server receiving the current clothing image from the household device after the household device obtains the current clothing image, a step of the server obtaining an ID account of the household device that sent the current clothing image based on the ID account, a step of the server determining an identity of the household device that sent the current clothing image based on the ID account, a step of the server determining a target demand based on the identity, and a step of the server selectively recommending a washing program or a clothing coordination idea based on the current clothing image based on the target demand.

[0022] According to a third aspect, the present invention provides a control method for a smart home system, the smart home system including a household device and a server, the household device can communicate with the server, the control method includes the steps of: the household device obtaining an image of a current garment; and the household device sending the image of the current garment to the server, after the server receives the image of the current garment, obtaining an ID account of the household device that sent the image of the current garment based on the ID account, and having the server determine an identity of the household device that sent the image of the current garment based on the ID account, and having the server determine a target demand based on the identity, and based on the target demand, having the server selectively recommend a washing program or a clothing coordination idea based on the image of the current garment.

[0023] According to a fourth aspect, the present invention provides a smart home system, the smart home system including a household device and a server, the household device can communicate with the server, the household device is used to obtain a current clothing image, the household device is used to send the current clothing image to the server, the server is used to obtain an ID account of the household device that sent the current clothing image based on the ID account, the server is used to determine an identity of the household device that sent the current clothing image based on the ID account, the server is used to determine a target demand based on the identity, and based on the target demand, the server can selectively recommend a washing program or a clothing coordination idea based on the current clothing image. Effect of the Invention

[0024] In a preferred technical proposal of the control method of the present invention, the household appliance obtains an image of current clothes, and the household appliance sends the image of the current clothes to the server, the server obtains an ID account of the household appliance that sent the image of the current clothes, and the server determines the identity of the household appliance that sent the image of the current clothes based on the ID account, and the server determines target demand based on the identity, and based on the target demand, the server selectively recommends a washing program or a clothing coordination suggestion based on the current clothes image.

[0025] The present invention can accurately determine the target needs of a user based on the identity of the household appliance that sends the current clothing image to the server, and based on the target needs of the user, the server can selectively recommend a washing program or clothing coordination plan based on the current clothing image, so as to recommend a reasonable and accurate washing program, thereby cleaning the clothes thoroughly, avoiding damage to the clothes, improving the clothing washing effect, and recommending a reasonable clothing coordination plan to the user, thereby achieving an optimal or relatively good coordination effect, better meeting the different needs of users and further improving the user's usage experience.

[0026] Furthermore, if it is determined that the user's target demand is clothing coordination advice, the server pre-processes the image of the current clothing to obtain a first body image of the current clothing, and removes the influence of factors such as background environment, and the server judges whether a top and a bottom exist simultaneously in the first body image, and if a top and a bottom exist simultaneously in the first body image, it indicates that the user has coordinated a piece of clothing, but has not yet determined whether the current coordination is reasonable, so that the server can know the result of the current coordination by judging whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database, and further improves the user's usage experience; and if only a top or a bottom exists in the first body image, it indicates that the user wants to wear this top or bottom, but has not yet determined how to coordinate it to achieve an optimal or relatively good coordination effect, so that the server recommends a target coordination suggestion based on the first body image, the clothing coordination database and the clothing private database, to provide the user with a relatively good clothing coordination suggestion, and further meets the user's usage demand.

[0027] Furthermore, when it is determined that the user's target demand is washing clothes, the server pre-processes the image of the current clothes to obtain a second body image of the current clothes, and removes the influence of factors such as background environment, and the server can accurately determine the attribute information of the current clothes based on the second body image, and the server compares the second body image with the historical clothes images stored in the clothes image database, and based on the comparison result, the server judges whether the current clothes are one of the historical clothes, and can accurately determine whether the current clothes have been washed, and if the current clothes are one of the historical washed clothes, it indicates that the clothes have been washed, and in order to accurately determine the washing program, the server determines the washing program based on the attribute information and the historical washing data of the current clothes, and performs the washing process. By considering not only the historical washing status of the type but also the attributes of the clothes, the recommended washing program is more reasonable and accurate. If the current clothes is not one of the historical washed clothes, it will indicate that the clothes have not been washed yet. In order to accurately determine the washing program, the server determines the washing program based on the attribute information, so that the recommended washing program is more reasonable and accurate. After determining the washing program, the washing program is sent to the washing machine, so that the washing machine can wash the clothes more cleanly and thoroughly based on the recommended washing program, and can more effectively avoid damage to the clothes, improve the washing effect of the clothes, better meet the usage needs of the user, and further improve the usage experience of the user. [Brief description of the drawings]

[0028] The control method of the present invention will be described below in conjunction with a washing machine with reference to the drawings. [Figure 1] FIG. 2 is a structural diagram of the smart home system of the present invention; [Diagram 2] 2 is a main flowchart of a control method of the present invention. [Diagram 3] 2 is a flow chart of a control method for determining and selectively recommending a washing program or clothing coordination scheme based on target demand according to the present invention. [Figure 4]4 is a flowchart of a control method for recommending clothing coordination suggestions according to the present invention. [Diagram 5] 1 is a flow chart 1 of a control method for determining whether a current clothing coordination is reasonable according to the present invention; [Figure 6] 2 is a flow chart 2 of the control method of determining whether the current clothing coordination is reasonable in the present invention. [Figure 7] 4 is a flowchart of a control method for recommending target coordination proposals according to the present invention. [Figure 8] 4 is a flow chart of a control method for recommending a wash program according to the present invention. [Figure 9] 4 is a flow chart of a control method for determining whether a current garment is one of the history garments of the present invention. [Figure 10] 1 is a logic diagram of the control method of the present invention. [Figure 11] 2 is a logic diagram 2 of the control method of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. Those skilled in the art should understand that these embodiments are merely for interpreting the technical principles of the present invention, and are not intended to limit the scope of protection of the present invention. For example, the present application is described in conjunction with a washing machine, a care machine, and a dressing mirror, but the technical solution of the present invention is not limited thereto, and the smart home system and control method can obviously be used in other household appliances such as a washer / dryer, a dryer, and a smart closet, and such modifications do not deviate from the principle and scope of the present invention.

[0030] It is to be noted that in describing the present invention, the terms "first," "second," "third," "fourth," "fifth," "sixth," "seventh," and "eighth" are used for descriptive purposes only and should not be understood as indicating or implying relative importance.

[0031] It should be further explained that in the description of the present invention, unless otherwise expressly specified and limited, the terms "connection" and "installation" should be interpreted in a broad sense, for example, may be a fixed connection, a removable connection, an integral connection, a mechanical connection, an electrical connection, a direct connection, an indirect connection via an intermediate medium, or an internal communication between two elements. Those skilled in the art may understand the specific meaning of the above terms in the present invention based on the specific situation.

[0032] Based on the technical problems proposed in the background art, the present invention provides a control method for a smart home system and a smart home system, which can accurately determine the target needs of a user based on the identity of a household device that sends a current clothing image to a server, and based on the target needs of the user, the server can selectively recommend a washing program or clothing coordination plan based on the current clothing image, so as to recommend a reasonable and accurate washing program, thereby cleaning the clothes thoroughly, avoiding damage to the clothes, improving the clothing washing effect, and recommending a reasonable clothing coordination plan to the user, thereby achieving an optimal or relatively good coordination effect, so as to better meet different needs of users and further improve the user's usage experience.

[0033] First, the smart home system of the present invention will be described with reference to Fig. 1. Here, Fig. 1 is a structural diagram of the smart home system of the present invention.

[0034] As shown in FIG. 1, the smart home system of the present invention includes a household device 1 and a server 2, the household device 1 can communicate with the server 2, the household device 1 is used to obtain an image of current clothing, the household device 1 is used to send the image of the current clothing to the server 2, the server 2 is used to obtain an ID account of the household device 1 that sent the image of the current clothing based on the ID account, the server 2 is used to determine the identity of the household device 1 that sent the image of the current clothing based on the ID account, the server 2 is used to determine target demand based on the identity, and based on the target demand, the server 2 can selectively recommend a washing program or clothing coordination plan based on the image of the current clothing.

[0035] Preferably, the household appliance 1 is provided with an image acquisition module, which captures the current clothing image, where the image acquisition module can be, but is not limited to, a webcam, a camera, etc.

[0036] Furthermore, the household appliances 1 may be divided into three types, the first type 11 including washing machines, dryers, washer-dryers, shoewashers, etc., which are appliances that users can use to wash clothes, the second type 12 including dressing mirrors, smart closets, etc., which are appliances that users can use to put on clothes, and the third type 13 including clothing care machines, etc., which are equipped with dressing mirrors, which users can use to wash clothes and put on clothes. Of course, the classification method of the household appliances 1 is not limited to the above-listed methods, and the household appliances 1 may be divided into four types, and the fourth type includes ironing machines, etc., which can only iron and care for clothes, and those skilled in the art can flexibly adjust and set the classification method of the household appliances 1 according to actual usage demands, etc.

[0037] Preferably, the server 2 is set with a clothing coordination database, a clothing private database and a clothing image database, in which the clothing coordination database stores a wide variety of clothing images of different styles and coordination suggestions of different styles, the clothing private database stores clothing images owned by the user, the clothing image database stores historical clothing images, historical clothing attributes and historical washing data, the historical clothing attributes include clothing color, material and type, the historical washing data includes historical washing times and historical color depth, and the database can further store current clothing images, current clothing color, material and type.

[0038] Here, the server 2 may be a cloud server 2 and a background server 2, but is not limited to them.

[0039] Preferably, the home device 1 can communicate with the server 2 via communication methods such as wide area network, local area network, WIFI, WIFI+Access Internet Router, Bluetooth, ZIGBEE, NFC, 3G / 4G / 5G, etc.

[0040] Hereinafter, the control method for a smart home system of the present invention will be described with reference to Fig. 2. Here, Fig. 2 is a flow chart of the control method of the present invention.

[0041] As shown in FIG. 2, the control method of the present invention includes the following steps.

[0042] S100, the home appliance, obtains the image of the current clothing, S200, the home appliance sends the image of the current clothing to the server; S300, the server obtains the ID account of the home appliance that sent the current clothing image; S400, the server determines the identity of the home appliance that sent the current clothing image based on the ID account; S500, the server determines the target demand based on the identity; S600, based on the target demand, the server selectively recommends a washing program or a clothing coordination scheme based on the current clothing image.

[0043] In step S100, the home appliance takes an image of the current garment with an image acquisition module, such as a webcam, a camera, etc.

[0044] In addition, in the above process, the order and manner of execution of steps S200 and S300 are not limited to the order and manner listed above. Step S300 may be executed first and then step S200, or steps S200 and S300 may be executed simultaneously. Those skilled in the art can flexibly adjust and set the order of execution of steps S200 and S300, and the present invention does not impose any restrictions in this regard.

[0045] Hereinafter, the control method of the present invention for determining and selectively recommending a washing program or a clothing coordination plan based on a target demand will be described with reference to Figures 3 to 8. Here, Figure 3 is a flowchart of the control method of the present invention for determining and selectively recommending a washing program or a clothing coordination plan based on a target demand, Figure 4 is a flowchart of the control method of the present invention for recommending a clothing coordination plan, Figure 5 is a flowchart of the control method of the present invention for determining whether a current clothing coordination is reasonable, Figure 6 is a flowchart of the control method of the present invention for recommending a target coordination plan, Figure 7 is a flowchart of the control method of the present invention for recommending a washing program, and Figure 8 is a flowchart of the control method of the present invention for determining whether a current garment is one of the history garments.

[0046] As shown in FIG. 3, in step S500, the step of “the server determines the target demand based on the identity” specifically includes the following steps:

[0047] S511, if the ID account of the household appliance is the ID account of the first type of appliance, determining that the target demand is laundry washing; S512, if the ID account of the household device is the ID account of the second type device, determining that the target demand is clothing coordination advice; S513, if the ID account of the home device is the ID account of a third-type device, the home device obtains a control command; S514, the home appliance transmits a control command to the server.

[0048] In step S511, if the ID account of the household device is the ID account of a first type of device, it indicates that the device is an device that can wash clothes, such as a washing machine, and it is considered that the user wants to wash clothes, so it can be determined that the target demand is washing clothes.

[0049] In step S512, if the ID account of the household device is an ID account of a second type of device, it indicates that the device is a device that a user uses to put on clothes, such as a dressing mirror, and it is considered that the user is trying to put on clothes, so it can be determined that the target demand is clothing coordination advice.

[0050] In steps S513 and S514, if the ID account of the household device is an ID account of a third type device, it indicates that the device is a device that a user can use to wash clothes and put on clothes, such as a care machine, and obtains a control command from the user to determine the user's demand, and sends the control command to the server, so that the server determines the target demand according to the content of the control command.

[0051] In addition, it can obtain the user's voice control command, or obtain the control command input by the user through the control end of the care machine.

[0052] Continuing to refer to FIG. 3, in step S600, the step of "based on target demand, the server selectively recommends a washing program or a clothing coordination plan based on the current clothing image" specifically includes the following steps:

[0053] If the target demand is clothing coordination advice, execute step S612; If the target demand is washing clothes, execute step S613; If the server receives the control command, execute step S611; S611, the server judges whether the content of the control command is a recommendation of a washing program, if not, executes step S612, if yes, executes step S613; S612, the server recommends clothing coordination suggestions based on the current clothing image; S613, the server recommends a washing program based on the current clothes image.

[0054] In step S612, if the target demand is clothing coordination advice or the content of the control command is a clothing coordination suggestion recommendation, it indicates that the user is going to put on the clothing, and the server recommends a clothing coordination suggestion based on the current clothing image to meet the user's usage demand.

[0055] In step S613, if the target demand is washing clothes or the content of the control command is recommending a washing program, indicating that the user is going to wash clothes, the server recommends clothing coordination suggestions based on the current clothing image to meet the user's usage demand.

[0056] In addition, in the above process, step S612 and step S613 have no priority and are parallel, and are related only to the contents of the target demand or control command, and corresponding steps may be executed based on the contents of different target demands or control commands.

[0057] Next, referring to FIG. 4, the server has a clothing coordination database and a clothing private database, and a target detection model is pre-stored in the server. In step S612, the step of "the server recommends clothing coordination suggestions based on the current clothing image" specifically includes the following steps:

[0058] S621, the server pre-processes the image of the current garment to obtain a first body image of the current garment; S622, the server judges whether the first main image has both tops and bottoms, if so, executes step S623, if not, executes step S624; S623, the server determines whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database; S624, the server recommends a target coordination suggestion according to the first body image, the clothing coordination database and the clothing private database.

[0059] In step S621, the server calls the target detection model to pre-process the current clothing image to obtain the current clothing body image. Taking the FCOS algorithm as an example, the target detection model adopts the FCOS algorithm to identify the current clothing image, identifies the clothing in the current clothing image and the position in the image where the clothing is located, and removes the background of the current clothing image according to the position to obtain a first clothing body image, and stores the first clothing body image in the clothing private database as the user's existing clothing image, and can be directly retrieved when coordinating the clothing again.

[0060] Of course, the target detection model may adopt the Faster R-CNN algorithm, the SPPNet algorithm, the SSD algorithm, the YOLO algorithm, etc. to preprocess the current clothing image to obtain the first body image, and regardless of which algorithm the target detection model adopts, the specific method of preprocessing the current clothing image corresponding to any algorithm should not constitute any limitation on the present invention.

[0061] In step S623, if the first body image contains both tops and bottoms, it indicates that the user has coordinated a piece of clothing, but has not determined whether the current coordination is reasonable, so that the server determines whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database, allowing the user to know the result of the current coordination, and further improving the user's usage experience.

[0062] In step S624, if only a top or bottom exists in the first body image, it indicates that the user wants to wear this top or bottom, but is not sure how to coordinate it to achieve the optimal or relatively good coordination effect, so the server recommends target coordination suggestions based on the first body image, the clothing coordination database and the private clothing database, provides the user with relatively good clothing coordination suggestions, and further meets the user's usage needs.

[0063] In addition, in the above process, steps S623 and S624 have no priority and are parallel, and are related only to the judgment result of whether a top and a bottom are simultaneously present in the first main image, and corresponding steps may be executed based on different judgment results.

[0064] Next, referring to FIG. 5, the server has a deep learning model pre-stored in it, and in step S623, the step of "the server determines whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database" specifically includes the following steps:

[0065] S631, the server determines a first matching degree between the top and the bottom based on the first body image; S632, the server determines a second matching degree of the top and the bottom based on the first body image and the clothing coordination database; S633, the server determines a total fitness based on the first fitness and the second fitness; S634, the server determines whether the current clothing coordination is reasonable based on the total suitability.

[0066] As shown in FIG. 6, in step S631, the step of "the server determines a first compatibility degree between the top and the bottom based on the first body image" specifically includes the following steps:

[0067] S641, the server determines a top color and a bottom color based on the first body image; S642, determining whether the color of the top and the color of the bottom conform to a preset rule; S643, determining the first compatibility between the color of the top and the color of the bottom based on the judgment result; In step S641, the server calls the target detection model to identify the first body image and identify the top color and bottom color of the current clothing. Here, the target detection model may adopt the FCOS algorithm, the Faster R-CNN algorithm, the SPPNet algorithm, the SSD algorithm, the YOLO algorithm, etc. to identify the first body image and identify the top color and bottom color of the current clothing. Regardless of which algorithm the target detection model adopts, the specific method of identifying the first body image corresponding to any algorithm should not constitute any limitation on the present invention.

[0068] In step S642, the colors of the clothes are divided into cool colors, warm colors, and neutral colors, where cool colors include cool colors such as cyan, blue, purple, green, and gray, warm colors include warm colors such as red, orange, and yellow, and neutral colors include black and white. Accordingly, the preset rules may coordinate cool colors with cool colors, warm colors with warm colors, and neutral colors with cool colors, warm colors, and neutral colors. Furthermore, when coordinated according to the preset rules, the color compatibility is 1, and when coordinated not according to the preset rules, the color compatibility is 0.5. Of course, the color compatibility listed above is only illustrative and not limiting, and those skilled in the art may flexibly adjust and set the color compatibility according to actual coordination needs.

[0069] In step S643, if the coordination is done according to the above-mentioned preset rules, for example, if the color of the top is white and the color of the bottom is blue, which matches the coordination rule of coordinating a neutral color with a cool color, the first compatibility degree between the color of the top and the color of the bottom is determined to be 1. If the coordination is not done according to the above-mentioned preset rules, for example, if the color of the top is cyan and the color of the bottom is yellow, which does not match any of the coordination rules listed above, the first compatibility degree between the color of the top and the color of the bottom is determined to be 0.5.

[0070] Continuing to refer to FIG. 6, the matching based on tops will be further described as an example. In step S632, the step of "the server determines a second matching degree between the tops and bottoms based on the first body image and the clothing coordination database" specifically includes the following steps:

[0071] S644, the server determines a current top image and a current bottom image based on the first body image; S645, the server calculates a first similarity between the current top image and each standard top image stored in the clothing coordinate database, where the first similarity may be expressed using a first cosine distance or a first Euler distance; S646, the server determines a target bottom wear image from the clothing coordination database based on each first similarity degree; S647, the server calculates a second similarity between each target bottom image and the current bottom image, where the second similarity may be expressed by adopting a second cosine distance or a second Euler distance; S648, the server calculates a second compatibility measure based on each second similarity measure and its corresponding first similarity measure.

[0072] In step S644, the server calls the target detection model to identify the first body image and identify the current top image and the current bottom image of the current clothing. Here, the target detection model may adopt the FCOS algorithm, the Faster R-CNN algorithm, the SPPNet algorithm, the SSD algorithm, the YOLO algorithm, etc. to identify the first body image and identify the current top image and the current bottom image, and regardless of which algorithm the target detection model adopts, the specific method of identifying the first body image corresponding to any algorithm should not constitute any limitation to the present invention.

[0073] In step S645, assuming that three standard top images are stored in the clothing coordination database, the server calls the deep learning model to extract a first vector matrix of the current top image and a second vector matrix of the three standard top images, calculates a first similarity between each of the extracted first vector matrices and the three second vector matrices, and adopts a first cosine distance to represent the first similarity, for example, the first cosine distance between the first vector matrix and the three second vector matrices is 0.9, 0.8, 0.3, respectively. Here, the deep learning model may adopt ResNet101 algorithm, CNN algorithm, HRNet algorithm, DeeplabV3+ algorithm and ResNeXt algorithm, etc. to calculate the first similarity between the current top image and the standard top image, and regardless of which algorithm the deep learning model adopts, the specific method of calculating the first similarity between the current top image and the standard top image corresponding to any algorithm should not constitute any limitation on the present invention.

[0074] In step S646, the server determines whether the first similarity is greater than a set similarity, where the set similarity can be expressed by adopting a set cosine distance or by adopting a set Euler distance, and if the first similarity is greater than the set similarity, the server determines the target bottom image based on the standard top image corresponding to the first cosine distance. If the first similarity is equal to or less than the set similarity, the server does not determine the target bottom image based on the first cosine distance.

[0075] For example, the set cosine distance is 0.7, and the first cosine distances calculated in step S645 are 0.9, 0.8, and 0.3, respectively, and 0.9, 0.8, and 0.3 are compared with 0.7, respectively, where 0.9 is greater than 0.7, and 0.8 is greater than 0.7 and greater than the set similarity, indicating that the standard top image with the first cosine distances of 0.9 and 0.8 respectively has a very high similarity with the current top image, and therefore the target bottom image can be determined based on the standard top image with the first cosine distances of 0.9 and 0.8 respectively. Because the clothing coordination database stores different styles of coordination plans, the standard top image can determine the target bottom image to be coordinated with it according to the coordination plans stored in advance, that is, two target bottom images are determined.

[0076] Furthermore, 0.3 is smaller than 0.7 and smaller than the set similarity, indicating that the standard top image corresponding to the first cosine distance of 0.3 has a very low similarity to the current top image, and therefore the target bottom image is not determined based on the first cosine distance of 0.3.

[0077] In step S647, in step S646, two target bottom images are determined, and the server calls the deep learning model to extract the third vector matrix of the current bottom image and the fourth vector matrices of the two target bottom images, calculates the second similarity between each extracted third vector matrix and the two fourth vector matrices, and adopts the second cosine distance to represent the second similarity, for example, the second cosine distance between the third vector matrix and the two fourth vector matrices is 0.8 and 0.6, respectively.

[0078] In step S648, the server may calculate the product of each second similarity and the corresponding first similarity, and use this product as the second compatibility, specifically, by referring to formula (1).

[0079] F2 = D1 × D2 (1) In the above formula (1), F2 is the second fitness, D1 is the first similarity, and D2 is the second similarity.

[0080] For example, if the second cosine distances between the third vector matrix and the two fourth vector matrices calculated in step S647 are 0.8 and 0.6, respectively, and the corresponding first cosine distances are 0.9 and 0.8, respectively, the second fitnesses calculated according to formula (1) are 0.72 and 0.48, respectively, that is, F 21 = 0.8 × 0.9 = 0.72, and F 22 =0.6×0.8=0.48.

[0081] Alternatively, in an alternative embodiment, matching may be performed based on the bottoms, and the execution method of steps S645 to S648 may be adjusted accordingly. The method for the part that needs to be adjusted is specifically as follows:

[0082] The server calculates a first similarity between the current bottom image and each standard bottom image stored in the clothing coordinate database, where the first similarity may be expressed using a first cosine distance or a first Euler distance; The server determines a target top image based on each first similarity; The server calculates a second similarity between each target top image and the current top image, where the second similarity may be expressed by adopting a second cosine distance or by adopting a second Euler distance; For the specific execution methods of the above steps, refer to the execution methods of steps S645 to S648, and no further description will be given here.

[0083] Continuing to refer to FIG. 6, in step S633, "the server determines a total fitness degree based on the first fitness degree and the second fitness degree" specifically includes the following steps:

[0084] S649, the server calculates a total fitness score based on the first fitness score and each of the second fitness scores; In step S649, the server calculates the total fitness score in the following manner: F0 = F1 × k1 + F2 × k2 (2) In the above formula (2), F0 is the total fitness, F1 is the first fitness, F2 is the second fitness, k1 is the first weight value, and k2 is the second weight value.

[0085] For example, k1 is 0.5, k2 is 0.5, and in step S643, the first compatibility score of the top color and the bottom color is determined to be 1. The second compatibility scores calculated in step S648 are 0.72 and 0.48, respectively. In this case, the total compatibility scores calculated according to formula (2) are 0.86 and 0.74, respectively, that is, F 01 = 1 × 0.5 + 0.72 × 0.5 = 0.86, so F 02 =1×0.5+0.48×0.5=0.74.

[0086] For example, if k1 is 0.5, k2 is 0.5, and in step S643, the first compatibility between the top color and the bottom color is determined to be 0.5, and the second compatibility calculated in step S648 is 0.72 and 0.48, respectively, then the total compatibility calculated according to formula (2) is 0.61 and 0.49, respectively, i.e., F 03 = 0.5 × 0.5 + 0.72 × 0.5 = 0.61, so F 04 =0.5×0.5+0.48×0.5=0.49.

[0087] Continuing to refer to FIG. 6, in step S634, "the server determines a total relevance based on the first relevance and the second relevance" specifically includes the following steps:

[0088] S650, the server compares all the total matching scores and determines the maximum matching score; S651, judge whether the maximum matching degree is greater than a preset matching degree, if so, execute step S652, if not, execute step S653; S652, determining that the current clothing coordination is reasonable, and the server transmits the determination result to the home appliance, thereby displaying "the current clothing coordination is reasonable" on the home appliance; S653, determining that the current clothing coordination is not reasonable, the server recommends a target coordination suggestion according to the first body image, the clothing coordination database and the clothing private database.

[0089] In steps S650 to S653, for example, the preset compatibility is 0.8, and the total compatibility calculated in step S649 is 0.86 and 0.74, respectively. 0.86 is compared with 0.74, and if 0.86 is greater than 0.74, 0.86 is determined as the maximum compatibility. Then, 0.86 is compared with 0.8, and 0.8 is greater than 0.8, i.e., greater than the preset compatibility, and it is described that the piece of clothing coordinated by the user is very reasonable. The server sends the judgment result to the home device, and displays "the current clothing coordination is reasonable" on the home device, so that the user can know the result of the current coordination, and further improves the user's usage experience.

[0090] For example, the preset compatibility is 0.8, and the total compatibility calculated in step S649 is 0.61 and 0.49 respectively. By comparing 0.61 and 0.49, if 0.61 is greater than 0.49, 0.61 is determined as the maximum compatibility. Then, by comparing 0.61 and 0.8, 0.61 is smaller than 0.8, i.e., smaller than the preset compatibility, indicating that the outfit coordinated by the user is not reasonable, the server recommends a target outfit suggestion based on the first body image, the clothing coordination database and the clothing private database, so as to provide the user with a relatively good clothing outfit suggestion and further meet the user's usage needs.

[0091] It should be noted that the above-listed first cosine distance, second cosine distance, first compatibility, second compatibility, total compatibility, maximum compatibility, and preset compatibility are merely illustrative and not limiting. Those skilled in the art can calculate the cosine distance and compatibility based on actual clothing images in actual applications, and flexibly adjust the preset compatibility according to actual accuracy requirements.

[0092] In addition, in the above process, the order and manner of execution of steps S631 and S632 are not limited to the order and manner listed above, and step S632 may be executed first and then step S631, or steps S631 and S632 may be executed simultaneously, and the order and manner of execution of steps S641 and S644 are not limited to the order and manner listed above, and step S644 may be executed first and then step S641, or steps S641 and S644 may be executed simultaneously, and the present invention does not make any restrictions thereon. Steps S652 and S653 are parallel and have no priority, and are only related to the judgment result of whether the maximum compatibility is greater than the preset compatibility, and corresponding steps may be executed according to different judgment results.

[0093] Next, referring to FIG. 7, in steps S624 and S653, the steps of "the server recommends a target coordination suggestion based on the first body image, the clothing coordination database and the clothing private database" specifically include the following steps:

[0094] S661, the server matches a plurality of candidate clothing coordination plans from a clothing coordination database according to a preset method based on the color of the current clothing; S662, the server calculates a third similarity between the candidate clothing images corresponding to each candidate clothing coordination plan and the clothing images owned by each user stored in the clothing private database, where the third similarity may be expressed by adopting a third cosine distance or a third Euler distance; S663, judge whether the third similarity is greater than the first preset similarity, and if so, execute step S664; if not, execute step S665, where the first preset similarity may be expressed by adopting a first preset cosine distance or a first preset Euler distance; S664, the server determines the candidate clothing coordination plan corresponding to the candidate clothing image as a target coordination plan, and sends the target coordination plan to the home appliance to display the target coordination plan on the home appliance; S665, the candidate clothing coordination plan corresponding to the candidate clothing image is not determined as the target coordination plan.

[0095] In step S661, in the case of step S653, since the current clothing color, top color and bottom color have already been obtained in step S641, they can be directly called and do not need to be obtained again; in the case of step S624, the server calls the target detection model to identify the first body image and identify the current clothing color, for example, the top color or bottom color. For the specific identification method, refer to the description in step S641 and no further description will be given here. Hereinafter, the current clothing needs to be coordinated with tops and bottoms as an example for further description.

[0096] Furthermore, the preset method corresponds to the above preset rule, and the preset method may be to coordinate a cool color with a cool color, a warm color with a warm color, a neutral color with a cool color, and a warm color with a neutral color. For example, the color of the top is cyan, and the colors of the bottoms that can be coordinated are blue, purple, green, and gray, and there are a total of two bottoms that match this color in the clothing coordination database.

[0097] In step S662, the server calls the deep learning model to extract the fifth vector matrices of two pieces of bottoms that match this color coordination in the clothing coordination database and the sixth vector matrices of all (e.g., two) pieces of bottoms stored in the clothing private database, calculates a third similarity between each of the two extracted fifth vector matrices and the two sixth vector matrices, and adopts a third cosine distance to represent the third similarity, for example, the third cosine distance between each of the two fifth vector matrices and the two sixth vector matrices is 0.9, 0.8, 0.7, and 0.3, respectively.

[0098] In step S664 and step S665, for example, the first preset cosine distance is 0.75, and the third cosine distance calculated in step S653 is 0.9, 0.8, 0.7, 0.3, respectively, and 0.9, 0.8, 0.7, 0.3 are compared with 0.75, respectively, where 0.9 is greater than 0.75, and 0.8 is greater than 0.75 and greater than the first preset cosine distance, and the third cosine distance corresponds to 0.9 and 0.8, respectively, in the bottoms stored in the clothing private database. If the image indicates that the bottoms match well with the current clothing (i.e., the top to be coordinated), then the bottoms stored in the clothing private database with the third cosine distances of 0.9 and 0.8 respectively are determined as the target coordination suggestion, that is, the candidate clothing coordination suggestion corresponding to the candidate clothing image is determined as the target coordination suggestion, and the server sends the target coordination suggestion to a home device, such as a dressing mirror or a care machine, thereby displaying the target coordination suggestion on the dressing mirror or the care machine.

[0099] Furthermore, 0.7 is smaller than 0.75, and 0.3 is smaller than 0.75 and smaller than the first preset cosine distance, indicating that the bottoms stored in the clothing private database corresponding to the third cosine distance of 0.7 and 0.3 respectively do not match the current clothing (i.e., the top to be coordinated) at all, and therefore the bottoms stored in the clothing private database corresponding to the third cosine distance of 0.3 are not determined as the target coordination plan, i.e., the candidate clothing coordination plan corresponding to the candidate clothing image is not determined as the target coordination plan.

[0100] It should be noted that the third cosine distance and the first preset cosine distance listed above are merely illustrative and not limiting. Those skilled in the art can calculate the third cosine distance based on actual clothing images in actual applications, and flexibly adjust the first preset cosine distance according to actual accuracy requirements.

[0101] In the above process, steps S664 and S665 have no priority and are parallel, and are related only to the judgment result of whether the third similarity is greater than the first preset similarity, and corresponding steps may be executed based on different judgment results.

[0102] Now referring to FIG. 8, the server further has a clothes image database, and in step S612, the step of "Server recommends a washing program based on the current clothes image" further includes the following steps:

[0103] S671, the server pre-processes the image of the current garment to obtain a second body image of the current garment; S672, the server determines attribute information of the current clothing based on the second body image; S673, the server compares the second body image with a historical clothing image stored in the clothing image database; S674, based on the comparison result, the server judges whether the current garment is one of the history garments, if so, execute step S675, if not, execute step S676; S675, if the current clothes is one of the historical washing clothes, the server determines a washing program according to the attribute information and the historical washing data of the current clothes, and sends the washing program to the household appliance; S676, if the current laundry is not one of the history laundry, the server determines a washing program based on the attribute information, and sends the washing program to the household appliance.

[0104] In step S671, the server calls the target detection model to pre-process the image of the current clothing, and obtains the second body image of the current clothing. For the specific pre-processing method, refer to the description in step S621, and will not be further described here.

[0105] In step S672, the server calls the target detection model to identify the second body image and identify the attribute information of the current clothing, such as color, type, material, etc. Here, the target detection model may adopt the FCOS algorithm, the Faster R-CNN algorithm, the SPPNet algorithm, the SSD algorithm, the YOLO algorithm, etc. to identify the second body image and identify the attribute information of the current clothing, such as color, type, material, etc. Regardless of which algorithm the target detection model adopts, the specific method of identifying the second body image corresponding to any algorithm should not constitute any limitation on the present invention.

[0106] In step S673, the server calls the deep learning model to compare the second body image with the historical clothing image, and calculates a fourth similarity between the second body image and each historical clothing image, where the fourth similarity can be expressed by adopting a fourth cosine distance or by adopting a fourth Euler distance. If there are three historical clothing images in total, the deep learning model extracts the seventh vector matrix of the second body image and the eighth vector matrix of the three historical clothing images, calculates the fourth similarity between each extracted seventh vector matrix and the three eighth vector matrices, and adopts a fourth cosine distance to represent the fourth similarity, for example, the fourth cosine distance between the seventh vector matrix and the three eighth vector matrices is 0.9, 0.6, 0.3, respectively, or the fourth cosine distance between the seventh vector matrix and the three eighth vector matrices is 0.7, 0.4, 0.1, respectively.

[0107] Next, referring to FIG. 9, in step S674, the step of "based on the comparison result, the server determines whether the current garment is one of the history garments" specifically includes the following steps:

[0108] S681, judge whether the fourth similarity is greater than the second preset similarity, if so, execute step S682, if not, execute step S683; S682, determining that the current garment is one of the historical garments; S683, determining that the current garment is not one of the historical garments.

[0109] Here, the second preset similarity may be expressed by adopting a second preset cosine distance, or may be expressed by adopting a second preset Euler distance.

[0110] In step S682, if the fourth similarity is greater than the second preset similarity, for example, the second preset cosine distance is 0.8, and the fourth cosine distance calculated in step S673 is 0.9, 0.6, and 0.3 respectively, 0.9, 0.6, and 0.3 are compared with 0.8 respectively, where 0.9 is greater than 0.8 and greater than the preset cosine distance, indicating that the similarity between this garment and one of the three historical garments is extremely high, and therefore it can be determined that the current garment is one of the historical garments, that is, the current garment is a garment that has already been washed, and the historical washing data of this garment is stored in the database.

[0111] Furthermore, in step S675, since the current clothes is one of the historically washed clothes, it indicates that the clothes have already been washed, and the historical washing data of the clothes is stored in the database. In order to accurately determine the washing program, the server determines the washing program based on the attribute information and the historical washing data of the current clothes, not only taking into account the historical washing status of the clothes, but also taking into account the attributes of the clothes, so that the recommended washing program is more reasonable and accurate, which can not only wash the clothes cleanly, but also avoid damage to the clothes, improve the washing effect of the clothes, better meet the user's usage needs, and further improve the user's usage experience.

[0112] In step S683, if the fourth similarity is lower than the second preset similarity, for example, the second preset cosine distance is 0.8, and the fourth cosine distance calculated in step S673 is 0.7, 0.4, and 0.1, respectively, by comparing 0.7, 0.4, and 0.1 with 0.8, they are all smaller than 0.8 and smaller than the preset cosine distance, indicating that the similarity between this clothing and the three historical clothing pieces is relatively low, and since the current clothing and the three historical clothing pieces compared to it are not the same clothing, it is determined that the current clothing is not one of the historical clothing pieces, and the historical washing data of this clothing piece is not stored in the database.

[0113] Furthermore, in step S676, if the current clothes is not one of the historical washed clothes, it indicates that the clothes have not been washed yet, and the attribute information of the clothes such as color, material and type is not stored in the clothes image database, the server determines a washing program based on the attribute information such as color, material and type, and the recommended washing program takes into account the attributes of the clothes, i.e., color, material and type, so that the recommended washing program is more reasonable and accurate, and can wash the clothes more cleanly and thoroughly. By sending the washing program to the washing machine, the washing machine can more effectively avoid damage to the clothes, improve the washing effect of the clothes, better meet the user's usage needs, and further improve the user's usage experience.

[0114] Furthermore, the server stores attribute information such as the identified color, material, and type in a clothing image database, and when the clothing is washed again, the attribute information such as the color, material, and type of the clothing can be directly obtained, without the need to identify the clothing again.

[0115] In addition, after the washing program is determined, the washing program is sent to a household appliance, such as a washing machine, so that the washing machine can wash the clothes more cleanly and thoroughly based on the recommended washing program, more effectively avoid damage to the clothes, improve the washing effect of the clothes, better meet the usage needs of the user, and further improve the usage experience of the user.

[0116] It should be noted that the fourth cosine distance and the second preset cosine distance listed above are merely illustrative and not limiting. Those skilled in the art can calculate the fourth cosine distance based on actual clothing images in actual applications, and flexibly adjust the second preset cosine distance according to actual accuracy requirements.

[0117] In the above process, steps S675 and S676 are parallel and have no priority, and only relate to the judgment result of whether the current garment is one of the history garments, and execute corresponding steps according to different judgment results. Steps S682 and S683 are parallel and have no priority, and only relate to the judgment result of whether the fourth similarity is greater than the second preset similarity, and execute corresponding steps according to different judgment results.

[0118] Hereinafter, a possible control flow of the present invention will be introduced with reference to Figs. 10 and 11. Here, Fig. 10 is a logic diagram 1 of the control method of the present invention, and Fig. 11 is a logic diagram 2 of the control method of the present invention.

[0119] As shown in FIG. 10 and FIG. 11, a possible complete flow of the control method of the present invention is as follows:

[0120] S701, the home appliance obtains an image of the current clothing; S702, the home appliance sends a current clothing image to the server; S703, the server obtains the ID account of the home appliance that sent the current clothing image; S704, the server determines the identity of the home appliance that sent the current clothing image according to the ID account; If the ID account of the home device is the ID account of the first type device, execute step S705; If the ID account of the home device is the ID account of the second type device, execute step S706; If the ID account of the home device is a third-type device ID account, execute step S707; S705, determine that the target demand is laundry washing; after step S705, execute step S731; S706, determine that the target demand is clothing coordination advice; after step S706, execute step S710; S707, household appliances obtain control instructions; S708, the home appliance sends a control command to the server; S709, the server judges whether the content of the control command is a recommendation of a washing program, if not, executes step S710, if yes, executes step S731; S710, the server pre-processes the image of the current garment to obtain a first body image of the current garment; S711, the server judges whether the first main image has both tops and bottoms; if so, execute step S712; if not, execute step S726; S712, the server determines the color of the top and the color of the bottom according to the first body image, that is, the color of the current clothing; S713, determine whether the color of the top and the color of the bottom meet the preset rule, if so, execute step S714, if not, execute step S715; S714, the first conformance is 1; S715, the first conformance is 0.5; After steps S714 and S715, step S716 is executed; S716, the server determines a current top image and a current bottom image based on the first body image; S717, the server calculates a first similarity between the current top image and each standard top image stored in the clothing coordination database; S718, the server determines a target bottom wear image from the clothing coordination database according to each first similarity; S719, the server calculates a second similarity between each target bottom image and the current bottom image; S720, the server calculates a second compatibility score based on each second similarity score and the corresponding first similarity score; S721, the server calculates a total fitness score based on the first fitness score and each of the second fitness scores; S722, the server compares all the total matching scores and determines the maximum matching score; S723, judge whether the maximum matching degree is greater than the preset matching degree, if so, execute step S724, if not, execute step S725; S724, determining that the current clothing coordination is reasonable, the server transmits the determination result to the home appliance, and displays "the current clothing coordination is reasonable" on the home appliance; S725, determining that the current clothing coordination is not reasonable, and executing step S726 after step S725; S726, the server matches a plurality of candidate clothing coordination ideas from the clothing coordination database according to a preset method based on the color of the current clothing; S727, the server calculates a third similarity between the candidate clothing image corresponding to each candidate clothing coordination plan and the clothing image owned by each user stored in the clothing private database; S728, judge whether the third similarity is greater than the first preset similarity, if so, execute step S729, if not, execute step S730; S729, the server determines the candidate clothing coordination plan corresponding to the candidate clothing image as a target coordination plan, and sends the target coordination plan to the home device to display the target coordination plan on the home device; S730, not determining the candidate clothing coordination plan corresponding to the candidate clothing image as the target coordination plan; S731, the server pre-processes the image of the current garment to obtain a second body image of the current garment; S732, the server determines attribute information of the current clothing according to the second body image, including material, type and color; S733, the server calculates a fourth similarity between the second body image and each historical clothing image stored in the clothing image database; S734, judge whether the fourth similarity is greater than the second preset similarity, if so, execute step S735, if not, execute step S737; S735, determining that the current garment is one of the historical garments; S736, the server determines a washing program according to the attribute information and the current laundry history data, and sends the washing program to the home appliance; S737, determining that the current garment is not one of the historical garments; S738, the server determines a washing program according to the attribute information, and sends the washing program to the household appliance.

[0121] It should be noted that the above examples are only used to explain the principles of the method of the present invention in the preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. In practical application, those skilled in the art may allocate the above functions and complete them through different steps as necessary, that is, re-decompose or combine the steps in the embodiments of the present invention. For example, the steps in the above examples may be integrated into one step, or further decomposed into multiple sub-steps to complete all or part of the functions described above. The names of the steps in the embodiments of the present invention are only for distinguishing each step, and are not considered as limitations on the present invention.

[0122] The above describes the technical solution of the present invention in conjunction with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art may make equivalent modifications or replacements to the relevant technical features, and the technical solutions after such modifications or replacements all fall within the protection scope of the present invention. [Explanation of symbols]

[0123] 1, household equipment 11, Type 1 equipment 12, Type 2 equipment 13, Type 3 equipment 2. Server

Claims

1. A control method for a smart home system, the smart home system including a home appliance and a server, the home appliance being capable of communicating with the server, The control method includes: said household appliance acquiring an image of a current garment; said home appliance sending an image of said current garment to said server; The server obtains an ID account of the home appliance that sent the current clothing image; The server determines the identity of the home appliance that sent the current clothing image according to the ID account; The server determines target demand based on the identity; and based on the target demand, the server selectively recommends a washing program or a clothing coordination plan based on the current clothing image.

2. The step of "the server determines the target demand based on the identity" specifically includes: If the ID account of the household appliance is a first type of appliance ID account, determining that the target demand is laundry washing; The step of "selectively recommending a washing program or a clothing coordination plan based on the image of the current clothing by the server based on the target demand" specifically includes:

2. The method of claim 1, further comprising the step of: if the target demand is washing clothes, the server recommending a washing program based on the current clothes image.

3. The step of "the server determines target demand based on the identity" includes: If the ID account of the household device is a second type device ID account, determining that the target demand is clothing coordination advice; The step of "selectively recommending a washing program or a clothing coordination plan based on the image of the current clothing by the server based on the target demand" includes: The control method according to claim 1 , further comprising: when the target demand is clothing coordination advice, the server recommending a clothing coordination suggestion based on the image of the current clothing.

4. The step of "the server determines target demand based on the identity" includes: When the ID account of the household device is a third-type device ID account, the household device obtains a control command; and the home appliance transmitting the control command to the server; The step of "selectively recommending a washing program or a clothing coordination plan based on the image of the current clothing by the server based on the target demand" includes: a step of the server determining whether the content of the control command is a recommendation of a washing program; If the content of the control command is a recommendation of a washing program, the server recommends a washing program based on the image of the current clothes; and / or The control method according to claim 1 , further comprising the step of: when the content of the control command is a clothing coordination suggestion recommendation, the server recommends the clothing coordination suggestion based on the image of the current clothing.

5. The server has a clothing coordinate database and a clothing private database; Specifically, the step of "the server recommends a clothing coordination plan based on the image of the current clothing" includes: The server pre-processes the image of the current garment to obtain a first body image of the current garment; the server determining whether a top and a bottom are simultaneously present in the first body image; When the first body image includes a top and a bottom, the server determines whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database; and / or The control method according to claim 3 or 4, further comprising a step of: when only tops or bottoms are present in the first body image, the server recommends a target coordination suggestion based on the first body image, the clothing coordination database and the clothing private database.

6. Specifically, the step of "the server determines whether the current clothing coordination is reasonable based on the first body image and the clothing coordination database" is as follows: The server determines a first compatibility between the top and the bottom based on the first body image; The server determines a second matching degree between the top and the bottom based on the first body image and the clothing coordination database; the server determining a total fitness based on the first fitness and the second fitness; The control method according to claim 5, further comprising a step of: the server determining whether the current clothing coordination is reasonable based on the total fitness score.

7. Specifically, the step of "the server determines a first compatibility between the top and the bottom based on the first body image" includes: The server determines a color of a top and a color of a bottom based on the first body image; determining whether the color of the top and the color of the bottom match a preset rule; The control method according to claim 6, further comprising a step of determining the first compatibility between the color of the top and the color of the bottom based on a result of the judgment.

8. Specifically, the step of "the server determines a second matching degree between the top and the bottom based on the first body image and the clothing coordination database" includes: The server determines a current top image and a current bottom image based on the first body image; The server calculates a first similarity between the current top image and each standard top image stored in the clothing coordinate database, or between the current bottom image and each standard bottom image stored in the clothing coordinate database; The server determines a target bottom wear image from the clothing coordinate database based on each of the first similarities, or determines a target top wear image from the clothing coordinate database based on each of the first similarities; The server calculates a second similarity between each of the target bottom images and the current bottom image, or between each of the target top images and the current top image; and a step of the server calculating the second degree of compatibility based on each of the second degrees of similarity and the corresponding first degree of similarity.

9. F ２ =D １ ×D ２ The second fitness is calculated by the method Here, F ２ is the second fitness, and D １ is the first similarity, and D ２ 9. The control method according to claim 8, wherein: is the second similarity.

10. Specifically, the step of "the server determines a total fitness based on the first fitness and the second fitness" includes:

9. The method of claim 8, further comprising the step of: the server calculating the total fitness based on the first fitness and each of the second fitnesses, respectively.

11. F ０ =F １ ×k １ +F ２ ×k ２ The total fitness is calculated by the method Here, F ０ is the total fitness, F １ is the first fitness, F ２ is the second fitness, and k １ is the first weight value, k ２ 11. The method of claim 10, wherein: is a second weight value.

12. Specifically, the step of "the server determines whether the current clothing coordination is reasonable based on the total suitability score" is as follows: the server comparing all the total fitness scores to determine the maximum fitness score; determining whether the maximum fitness is greater than a preset fitness; If the maximum compatibility score is greater than the preset compatibility score, determining that the current clothing coordination is reasonable, and the server transmits a determination result to the home appliance, thereby displaying "the current clothing coordination is reasonable" on the home appliance; and / or The control method of claim 10, further comprising: if the maximum matching degree is less than or equal to the preset matching degree, determining that the current clothing coordination is not reasonable, and the server recommends a target coordination suggestion based on the first body image, the clothing coordination database and the clothing private database.

13. The step of "the server recommends a target coordination plan based on the first body image, the clothing coordination database, and the clothing private database" specifically includes: The server matches a plurality of candidate clothing coordination ideas from the clothing coordination database according to a preset method based on the color of the current clothing; The server calculates a third similarity between a candidate clothing image corresponding to each of the candidate clothing coordination ideas and a clothing image owned by each user stored in the private clothing database; determining whether the third similarity is greater than a first preset similarity; if the third similarity is greater than the first preset similarity, the server determines the candidate clothing outfit suggestion corresponding to the candidate clothing image as the target outfit suggestion, and transmits the target outfit suggestion to the home device, thereby displaying the target outfit suggestion on the home device; and / or 13. The control method according to claim 12, further comprising a step of not determining the candidate clothing coordination idea corresponding to the candidate clothing image as the target coordination idea if the third similarity is equal to or lower than the first preset similarity.

14. The server further includes a clothing image database; The step of "the server recommends a washing program based on the image of the current clothes" specifically includes: The server pre-processes the image of the current garment to obtain a second body image of the current garment; The server determines attribute information of the current clothing based on the second body image; the server comparing the second body image with historical clothing images stored in the clothing image database; Based on the comparison result, the server determines whether the current garment is one of the historical garments; If the current laundry is one of the historical laundry, the server determines a washing program based on the attribute information and the historical laundry data of the current laundry, and sends the washing program to the household appliance; and / or The control method according to claim 2 or 4, further comprising a step of: if the current laundry is not one of the previously washed laundry, the server determines a washing program based on the attribute information and transmits the washing program to the household appliance.

15. Specifically, the step of "the server compares the second body image with a historical clothing image stored in the clothing image database" is as follows: The server calculates a fourth similarity between the second body image and each of the historical clothing images; Specifically, the step of "based on the comparison result, the server determines whether the current garment is one of the history garments" includes: determining whether the fourth similarity is greater than a second preset similarity; determining that the current garment is one of the history garments if the fourth similarity is greater than the second preset similarity; and / or The control method of claim 14, further comprising: determining that the current garment is not one of the history garments if the fourth similarity is less than or equal to the second preset similarity.

16. A control method for a smart home system, the smart home system including a home appliance and a server, the home appliance being capable of communicating with the server, The control method includes: receiving the current clothing image from the home appliance by the server after the home appliance has acquired the current clothing image; The server obtains an ID account of the home appliance that sent the current clothing image; The server determines the identity of the home appliance that sent the current clothing image according to the ID account; The server determines target demand based on the identity; and based on the target demand, the server selectively recommends a washing program or a clothing coordination plan based on the current clothing image.

17. A control method for a smart home system, the smart home system including a home appliance and a server, the home appliance being capable of communicating with the server, The control method includes: said household appliance acquiring an image of a current garment; The control method includes the steps of: having the household appliance send an image of the current clothes to the server; after the server receives the image of the current clothes, obtaining an ID account of the household appliance that sent the image of the current clothes, having the server determine the identity of the household appliance that sent the image of the current clothes based on the ID account, and having the server determine target demands based on the identity, and based on the target demands, having the server selectively recommend a washing program or a clothing coordination plan based on the image of the current clothes.

18. A smart home system, the smart home system including a home appliance and a server, the home appliance being capable of communicating with the server; The home appliance is used to capture images of the current garment; The home appliance is adapted to transmit an image of the current garment to the server; The server is used to obtain an ID account of the home appliance that transmitted the current clothing image; The server is used to determine the identity of the home appliance that sent the current clothing image based on the ID account; The server is used to determine target demand based on the identity; A smart home system, characterized in that, based on the target demand, the server can selectively recommend a washing program or a clothing coordination plan based on the current clothing image.

Images