Household appliance cleaning degree identification method and device, electronic device, and storage medium

Abstract

The application discloses a household appliance cleaning degree identification method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence; the method comprises the following steps: acquiring image data of a household appliance to be identified; inputting the image data into a target cleaning degree identification model to obtain a household appliance cleaning degree identification result output by the target cleaning degree identification model; wherein the target cleaning degree identification model is obtained by training an initial cleaning degree identification model based on an image sample pair and label information; the image sample pair comprises a first image sample corresponding to a household appliance before cleaning and a second image sample corresponding to a household appliance after cleaning; and the label information is used for representing the cleaning degree of the second image sample. According to the application, the cleaning degree of the household appliance can be accurately identified by the target cleaning degree identification model, manual observation and experience judgment are avoided, the accuracy of the household appliance cleaning degree evaluation is improved, and the labor cost is reduced.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, electronic device, and storage medium for identifying the cleaning level of household appliances. Background Technology

[0002] With the widespread use of home appliances in daily life, keeping them clean is crucial for their performance, lifespan, and user health. Currently, the assessment of appliance cleanliness relies primarily on manual observation and experience, a method that is highly subjective and inaccurate.

[0003] Therefore, how to accurately identify the cleaning level of home appliances is a problem that urgently needs to be solved. Summary of the Invention

[0004] This application provides a method, apparatus, electronic device, and storage medium for identifying the cleanliness level of household appliances. It can accurately identify the cleanliness level of household appliances, avoid manual observation and experience-based judgment, improve the accuracy of household appliance cleanliness assessment, and reduce labor costs.

[0005] In a first aspect, embodiments of this application provide a method for identifying the cleaning level of household appliances, including:

[0006] Acquire image data of the home appliance to be identified;

[0007] Input the image data into the target cleaning degree recognition model to obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model;

[0008] The target cleaning degree recognition model is obtained by training the initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include the first image sample corresponding to the home appliance before cleaning and the second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

[0009] Secondly, embodiments of this application also provide a device for identifying the cleaning level of household appliances, comprising:

[0010] The acquisition module is used to acquire image data of the home appliance to be identified;

[0011] The recognition module is used to input image data into the target cleaning degree recognition model and obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model;

[0012] The target cleaning degree recognition model is obtained by training the initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include the first image sample corresponding to the home appliance before cleaning and the second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

[0013] Thirdly, embodiments of this application provide an electronic device, including:

[0014] One or more processors;

[0015] Memory, used to store one or more programs.

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the home appliance cleaning degree identification method described in any embodiment of this application.

[0017] Fourthly, embodiments of this application provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the method for identifying the degree of cleaning of household appliances as described in any embodiment of this application.

[0018] This application proposes a method, apparatus, electronic device, and storage medium for identifying the cleanliness level of home appliances. The method includes: acquiring image data of the home appliance to be identified; inputting the image data into a target cleanliness level identification model to obtain the home appliance cleanliness level identification result output by the target cleanliness level identification model; wherein the target cleanliness level identification model is obtained by training an initial cleanliness level identification model based on image sample pairs and label information. The image sample pairs include a first image sample corresponding to the home appliance before cleaning and a second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleanliness level of the second image sample. In other words, in the technical solution of this application, by inputting the image data of the home appliance to be identified into the target cleanliness level identification model and using the target cleanliness level identification model to identify the image data, the cleanliness level of the home appliance can be accurately identified, avoiding manual observation and experience-based judgment, improving the accuracy of home appliance cleanliness level assessment, and reducing labor costs. Simultaneously, it provides users and maintenance personnel with scientific and reasonable cleaning suggestions and decision-making basis, thereby improving the user experience. Attached Figure Description

[0019] Figure 1 A flowchart illustrating a method for identifying the cleanliness level of household appliances according to an embodiment of this application;

[0020] Figure 2 This is a schematic diagram of the training process of a target cleanliness recognition model provided in an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of the structure of a household appliance cleaning degree recognition device provided in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] In this embodiment, image recognition technology is applied to home appliance cleaning services, and an algorithm for judging the degree of home appliance cleaning based on image recognition is proposed.

[0026] Figure 1 This is a flowchart illustrating a method for identifying the cleanliness level of home appliances according to an embodiment of this application. The method can be executed by a home appliance cleanliness level identification device or an electronic device, which can be implemented by software and / or hardware, and can be integrated into any smart device with network communication capabilities.

[0027] like Figure 1 As shown, the method for identifying the cleanliness level of home appliances may include the following steps:

[0028] S110. Acquire image data of the home appliance to be identified.

[0029] In this embodiment of the application, the household appliance to be identified is, for example, a cleaned refrigerator, washing machine, air conditioner, or other household appliance.

[0030] In practical applications, to acquire image data of home appliances, it is necessary to use a high-resolution camera or webcam to capture images of the appliances. When capturing images, ensure sufficient lighting and a comprehensive angle, covering key parts of the appliance, such as internal heating elements, filters, and air ducts, as well as external surface stains and scratches.

[0031] S120. Input the image data into the target cleaning degree recognition model to obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model; wherein, the target cleaning degree recognition model is obtained by training the initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include the first image sample corresponding to the home appliance before cleaning and the second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

[0032] In this embodiment, the target cleaning degree recognition model is a convolutional neural network (CNN).

[0033] A CNN network consists of convolutional layers, pooling layers, fully connected layers, and an output layer. In this case, preprocessed images of household appliances are input as image data into the target cleanliness recognition model. The image data can be color or grayscale, depending on the model design and the characteristics of the training data.

[0034] After inputting the image data into the target cleaning degree recognition model, convolutional layers are used to perform convolution operations on the image data.

[0035] Specifically, feature extraction is first performed on the image data through multiple convolutional layers. Each convolutional layer contains multiple convolutional kernels, which slide across the image and perform convolution operations with local regions of the image data to extract different features.

[0036] For example, smaller convolutional kernels may be used to extract detailed features from image data, such as tiny stains or textures; larger convolutional kernels may be used to capture more macroscopic features, such as large areas of dirt distribution.

[0037] After feature extraction using convolutional layers, pooling layers are used to perform pooling operations on the image features.

[0038] Specifically, convolutional layers are connected to pooling layers. Pooling layers are used to reduce the resolution of image features, reducing computation while preserving key features. In practical applications, pooling operations include max pooling and average pooling. Max pooling selects the maximum value within a local region as the output, while average pooling calculates the average value within the local region as the output.

[0039] After pooling the image features using pooling layers, the extracted features are flattened into one-dimensional vectors and input into fully connected layers. The fully connected layers then integrate and non-linearly transform these features to learn a high-level representation of the image.

[0040] Finally, the output layer outputs the results of the cleaning level recognition of the home appliances.

[0041] In practical applications, the cleaning level identification result of home appliances can be a classification problem. For example, the cleaning level identification result can be divided into three categories: mild, moderate and severe. The output layer can use the Softmax function to calculate the probability and output the probability of each category. The category with the highest probability is the predicted cleaning level identification result.

[0042] For example, the result of the cleaning degree recognition of home appliances can be a regression problem. For instance, the recognition result is a specific cleaning degree value, which is between 0 and 1, where 0 represents no cleaning at all and 1 represents complete cleaning. The output layer can directly output this value.

[0043] In the technical solution of this application, the image data of the home appliance to be identified is input into the target cleaning degree recognition model. The model then identifies the image data, accurately determining the cleaning degree of the appliance. This avoids manual observation and experience-based judgment, improving the accuracy of appliance cleaning degree assessment and reducing labor costs. Simultaneously, it provides users and repair personnel with scientific and reasonable cleaning suggestions and decision-making basis, thereby enhancing the user experience.

[0044] Optionally, the target cleaning level recognition model determines the cleaning level recognition result of the home appliances in the following ways:

[0045] Step 1) Extract features from the image data to obtain the image features corresponding to the image data; the image features include at least one of the following: color features, texture features, and shape features.

[0046] In this embodiment of the application, image features are obtained by extracting features from image data through the convolutional layer in the target cleaning degree recognition model.

[0047] Color features are used to analyze the distribution and variation of colors in image data; for example, stains typically exhibit a specific range of colors.

[0048] Texture features are used to extract texture information such as roughness and regularity from the surface or internal components of home appliances to determine the degree of dirt accumulation.

[0049] Shape features are used to identify shape changes in appliance components, such as shape distortion caused by filter blockage.

[0050] Step 2) Determine the cleaning level of the home appliance based on at least one of the color features, texture features, and shape features.

[0051] In this embodiment of the application, the target cleaning degree recognition model can analyze whether the household appliance to be identified has dirt remaining after cleaning based on at least one of the above-mentioned color features, texture features and shape features, and thus determine the cleaning degree of the household appliance to be identified.

[0052] In some embodiments, the steps for determining the cleaning level recognition result of home appliances based on at least one of color features, texture features, and shape features can be found in [reference needed]. Figure 1 Step S120 will not be repeated here to avoid repetition.

[0053] Optionally, the target cleaning degree recognition model is trained in the following way:

[0054] Step 1) Input the image sample pairs and label information into the initial cleaning degree recognition model;

[0055] Step 2) Iteratively train the initial cleaning degree recognition model using the target loss function until the training stopping condition is met, and obtain the target cleaning degree recognition model;

[0056] The training stopping conditions include: the target loss function reaching a preset threshold, or the number of training iterations reaching a preset value; the target loss function includes the cross-entropy loss function and the L2 regularization loss function.

[0057] In this embodiment, the model is trained using a large number of labeled image samples of appliances before and after cleaning. The label information includes the level of cleaning severity (e.g., light, moderate, heavy, etc.).

[0058] It should be noted that after training the initial cleanliness level recognition model, it is necessary to test it on actual home appliance images and compare the results with human evaluations to calculate metrics such as accuracy and recall, thus verifying the effectiveness of the algorithm. Once the trained initial cleanliness level recognition model passes the test, the target cleanliness level recognition model is obtained.

[0059] Specifically, during the initial training of the cleaning degree recognition model, a target loss function is calculated based on the model's output and the actual label information (i.e., the manually labeled cleaning degree). The target loss function can be cross-entropy loss (used for classification problems), mean squared error loss (used for regression problems), etc.

[0060] Then, optimization algorithms (such as stochastic gradient descent, Adagrad, Adadelta, etc.) are used to adjust the model parameters, such as the weights of convolutional kernels and fully connected layers, based on the value of the loss function, in order to minimize the loss function and make the model's prediction results gradually approach the true cleanliness.

[0061] In the testing phase of the trained initial cleaning level recognition model, image data of new household appliances to be identified are input into the trained initial cleaning level recognition model to obtain the predicted cleaning level. The model's performance is evaluated using metrics such as accuracy, recall, F1 score, and mean squared error by comparing it with manually labeled real results.

[0062] For example, if identifying the cleanliness level of a home appliance is a classification problem, and the model predicts the cleanliness level of the appliance to be identified as moderate, and the true label is also moderate, then the model's prediction is correct. Accuracy is the proportion of all correctly predicted samples out of the total number of samples.

[0063] For example, if the result of identifying the cleanliness level of home appliances is a regression problem, assuming the model predicts a cleanliness level of 0.7 and the actual cleanliness level is 0.8, the accuracy of the model is measured by calculating the mean square error between the predicted value and the actual value.

[0064] Optionally, the initial cleaning degree recognition model can be iteratively trained using the target loss function until the training stopping condition is met, thus obtaining the target cleaning degree recognition model. This can be achieved through the following steps:

[0065] Step A: For the current training iteration, identify the model parameters based on the current initial cleaning level, and calculate the gradient corresponding to the target loss function;

[0066] Step B: Update the model parameters based on the gradient;

[0067] Step C: Iterate through steps A through B until the training stop condition is met, and obtain the trained target cleanliness recognition model.

[0068] In this embodiment, the computational logic of the model is optimized by continuously adjusting the model's structure, parameters, and training data, thereby improving the accuracy and reliability of the cleaning degree recognition model in assessing the cleaning degree of home appliances.

[0069] Alternatively, the target loss function is expressed by the following formula (1):

[0070]

[0071] Where N represents the number of image samples; Y ijP represents the true label of the i-th image sample belonging to category j; ij W represents the predicted probability that the i-th image sample belongs to class j; γ represents the regularization parameter; l These are the weight parameters of the l-th layer of the initial cleaning degree recognition model; This represents the square of the L2 norm of the weighting parameter.

[0072] In the above formula (1), N represents the number of image samples, and the input image corresponding to each image sample i is X. i The actual label information is Y. i ∈{0, 1, 2}, representing mild, moderate, and severe cleaning levels, respectively. The cleaning level identification model's prediction output for sample i is a probability vector p. i =[p i0 p i1 p i2 ].

[0073] Optionally, the gradient is calculated using the following formulas (2) and (3):

[0074]

[0075] in, This represents the gradient corresponding to the cross-entropy loss function in the objective loss function. This represents the gradient corresponding to the L2 regularized loss function in the target loss function.

[0076] Optionally, the model parameters can be updated based on the gradient, which can be achieved through the following steps:

[0077] The model parameters are updated based on the following formula (4);

[0078]

[0079] Where, θ new Represents the updated model parameters; θ old The model parameters before the update are represented by α; the learning rate is represented by μ; and the momentum parameter is represented by μ. This represents the gradient of the target loss function with respect to the parameter θ.

[0080] Figure 2 This is a schematic diagram of the training process of a target cleanliness recognition model provided in an embodiment of this application. See also... Figure 2 As shown, it includes the following steps:

[0081] Step 1: Begin.

[0082] Step 2: Input the training data into the initial cleaning degree recognition model.

[0083] Specifically, the training data includes image samples x i and real labels y i .

[0084] Step 3: Initialize the model parameters of the initial cleaning degree recognition model.

[0085] Specifically, model parameters include weights, biases, etc.

[0086] Step 4: Calculate the prediction probability p using forward propagation. i .

[0087] Step 5: Calculate the loss function L.

[0088] Specifically, the loss function L includes cross-entropy loss and L2 regularization term.

[0089] Step 6: Calculate the gradient of the loss function with respect to the model parameters.

[0090] Step 7: Update the model parameters using an optimization algorithm.

[0091] Specifically, optimization algorithms include, for example, stochastic gradient descent.

[0092] Step 8: Determine whether the training stop condition has been met.

[0093] Specifically, training stopping conditions include, for example, reaching a specified number of iterations or when the loss function converges.

[0094] If yes, then end the model training process; otherwise, return to step 4.

[0095] Figure 3 This is a schematic diagram of the structure of a household appliance cleaning degree identification device provided in an embodiment of this application.

[0096] like Figure 3 As shown, the appliance cleaning level identification device may include:

[0097] The acquisition module is used to acquire image data of the home appliance to be identified;

[0098] The recognition module is used to input image data into the target cleaning degree recognition model and obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model;

[0099] The target cleaning degree recognition model is obtained by training the initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include the first image sample corresponding to the home appliance before cleaning and the second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

[0100] This application proposes a device for identifying the cleaning level of home appliances. By using a target cleaning level identification model to identify image data, the device can accurately identify the cleaning level of home appliances, avoiding manual observation and experience-based judgment, improving the accuracy of home appliance cleaning level assessment, and reducing labor costs. At the same time, it provides users and maintenance personnel with scientific and reasonable cleaning suggestions and decision-making basis, thereby improving the user experience.

[0101] Optionally, the target cleaning level recognition model determines the cleaning level recognition result of the home appliances in the following ways:

[0102] Feature extraction is performed on image data to obtain the image features corresponding to the image data; the image features include at least one of the following: color features, texture features, and shape features;

[0103] The degree of cleaning of home appliances is determined based on at least one of the following features: color, texture, and shape.

[0104] Optionally, the device further includes a training module for:

[0105] Input the image sample pairs and label information into the initial cleaning degree recognition model;

[0106] The initial cleaning degree recognition model is iteratively trained using the target loss function until the training stopping condition is met, thus obtaining the target cleaning degree recognition model.

[0107] The training stopping conditions include: the target loss function reaching a preset threshold, or the number of training iterations reaching a preset value; the target loss function includes the cross-entropy loss function and the L2 regularization loss function.

[0108] Optionally, the training module is further used for:

[0109] Step A: For the current training iteration, identify the model parameters based on the current initial cleaning level, and calculate the gradient corresponding to the target loss function;

[0110] Step B: Update the model parameters based on the gradient;

[0111] Step C: Iterate through steps A through B until the training stop condition is met, and obtain the trained target cleanliness recognition model.

[0112] Alternatively, the target loss function is expressed by the following formula (1):

[0113]

[0114] Where N represents the number of image samples; Y ijP represents the true label of the i-th image sample belonging to category j; ij W represents the predicted probability that the i-th image sample belongs to class j; γ represents the regularization parameter; l These are the weight parameters of the l-th layer of the initial cleaning degree recognition model; This represents the square of the L2 norm of the weighting parameter.

[0115] Optionally, the gradient is calculated using the following formulas (2) and (3):

[0116]

[0117] in, This represents the gradient corresponding to the cross-entropy loss function. This represents the gradient corresponding to the L2 regularization loss function.

[0118] Optionally, the model parameters are updated based on the gradient, including:

[0119] The model parameters are updated based on the following formula (4);

[0120]

[0121] Where, θ new Represents the updated model parameters; θ old The model parameters before the update are represented by α; the learning rate is represented by μ; and the momentum parameter is represented by μ. This represents the gradient of the target loss function with respect to the parameter θ.

[0122] The aforementioned household appliance cleaning level identification device can execute the method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the household appliance cleaning level identification method provided in any embodiment of this application.

[0123] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement embodiments of the present invention is shown. For example... Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0124] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0125] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for identifying the degree of cleaning of household appliances.

[0126] In some embodiments, the appliance cleaning level identification method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the appliance cleaning level identification method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the appliance cleaning level identification method by any other suitable means (e.g., by means of firmware).

[0127] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0129] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0130] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 10, which includes: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device 10. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0131] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0132] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0133] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0134] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for identifying the cleaning level of household appliances, characterized in that, The method includes: Acquire image data of the home appliance to be identified; The image data is input into the target cleaning degree recognition model to obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model; The target cleaning degree recognition model is obtained by training an initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include a first image sample corresponding to the home appliance before cleaning and a second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

2. The method according to claim 1, characterized in that, The target cleaning degree recognition model determines the cleaning degree recognition result of the home appliance in the following way: Feature extraction is performed on the image data to obtain the image features corresponding to the image data; the image features include at least one of the following: color features, texture features, and shape features; The degree of cleaning of the home appliance is determined based on at least one of the color feature, the texture feature, and the shape feature.

3. The method according to claim 1 or 2, characterized in that, The target cleaning degree recognition model is trained in the following way: The image sample pairs and the label information are input into the initial cleaning degree recognition model; The initial cleaning degree recognition model is iteratively trained using the target loss function until the training stopping condition is met, thus obtaining the target cleaning degree recognition model. The training stopping conditions include: the target loss function reaching a preset threshold, or the number of training iterations reaching a preset value; the target loss function includes the cross-entropy loss function and the L2 regularization loss function.

4. The method according to claim 3, characterized in that, The step of iteratively training the initial cleaning degree recognition model using a target loss function until the training stopping condition is met, to obtain the target cleaning degree recognition model, includes: Step A: For the current iteration training number, calculate the gradient corresponding to the target loss function based on the model parameters of the identification model at the current initial cleaning level; Step B: Update the model parameters based on the gradient; Step C: Iterate through steps A through B until the training stop condition is met to obtain the trained target cleaning degree recognition model.

5. The method according to claim 3, characterized in that, The target loss function is expressed by the following formula (1): Where N represents the number of image samples; Y ij P represents the true label of the i-th image sample belonging to category j; ij W represents the predicted probability that the i-th image sample belongs to class j; γ represents the regularization parameter; l These are the weight parameters of the l-th layer of the initial cleaning degree recognition model; This represents the square of the L2 norm of the weighting parameter.

6. The method according to claim 4, characterized in that, The gradient is calculated using the following formulas (2) and (3): in, This represents the gradient corresponding to the cross-entropy loss function. This represents the gradient corresponding to the L2 regularization loss function.

7. The method according to claim 4, characterized in that, The step of updating the model parameters based on the gradient includes: The model parameters are updated based on the following formula (4); Where, θ new Represents the updated model parameters; θ old The model parameters before the update are represented by α; the learning rate is represented by μ; and the momentum parameter is represented by μ. This represents the gradient of the target loss function with respect to the parameter θ.

8. A device for identifying the cleaning level of household appliances, characterized in that, The device includes: The acquisition module is used to acquire image data of the home appliance to be identified; The recognition module is used to input the image data into the target cleaning degree recognition model to obtain the cleaning degree recognition result of the home appliance output by the target cleaning degree recognition model; The target cleaning degree recognition model is obtained by training an initial cleaning degree recognition model based on image sample pairs and label information. The image sample pairs include a first image sample corresponding to the home appliance before cleaning and a second image sample corresponding to the home appliance after cleaning. The label information is used to characterize the cleaning degree of the second image sample.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for identifying the degree of cleaning of household appliances as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method for identifying the cleaning level of household appliances as described in any one of claims 1 to 7.

Images