Washing machine dehydration control method, control device and washing machine based on image recognition

Abstract

The present invention relates to a washing machine dehydration control method, control device, and washing machine based on image recognition. The dehydration control method comprises: after drainage is completed and before dehydration begins, acquiring multiple clothing images from different perspectives within a washing drum; processing the multiple clothing images using a pre-trained recognition model to identify the stacking height, expanded area, and distribution state parameters of the clothing within the washing drum; inputting the identified stacking height, expanded area, and distribution state parameters into a pre-trained neural network model to determine the maximum non-oscillation speed at which dehydration begins under the current load of the clothing treatment drum; and determining the dehydration progress of the washing drum based on the maximum non-oscillation speed. The technical solution of the present invention predicts the maximum non-oscillation speed corresponding to different amounts of clothing and clothing states, allowing the washing machine to quickly reach the appropriate dehydration speed, significantly shortening the dehydration time, improving dehydration efficiency, and enhancing the dehydration effect.

Description

Technical Field

[0001] The present invention relates to the field of washing machines, and in particular to a washing machine dehydration control method and control device based on image recognition, and a washing machine. Background Art

[0002] The most common dehydration process for washing machines currently involves determining the preset spin speed level for the current load before dehydration begins after draining is complete. The speed is then gradually increased to the maximum speed using an eccentricity detection algorithm. This process fails to consider that, given a certain amount of laundry and its distribution, the washing machine will not oscillate before reaching a certain speed. This is especially true when the amount of laundry is very small, as the spin speed can be increased directly to the set maximum speed regardless of the laundry's current state in the drum, without oscillation or drum collision. Consequently, it takes a long time for the spin speed to reach the maximum speed during dehydration. Summary of the Invention

[0003] The present invention provides a washing machine dehydration control method, a control device and a washing machine based on image recognition, so as to at least solve the technical problem that it takes a long time for the dehydration speed of the washing machine to increase to the maximum speed during the dehydration process.

[0004] A first aspect of the present invention provides a washing machine dehydration control method based on image recognition, the dehydration control method comprising:

[0005] After draining is completed and before dehydration begins, multiple images of clothes in the washing drum are obtained from different perspectives;

[0006] Processing the plurality of clothing images using a pre-trained recognition model to identify stacking height, spread area, and distribution parameters of the clothing in the washing tub;

[0007] Inputting the identified stack height, the expanded area, and the distribution state parameters into a pre-trained neural network model to determine a maximum non-oscillating rotational speed for the laundry treatment drum to start dehydration with the current load;

[0008] The dehydration process of the washing tub is determined according to the maximum non-oscillation speed.

[0009] In some embodiments, determining the actual spin speed of the washing tub according to the maximum non-oscillation speed and a set maximum spin speed of the washing machine includes:

[0010] Determining whether the maximum non-oscillation speed reaches the set maximum dehydration speed;

[0011] When the maximum non-oscillation speed reaches the set maximum dehydration speed, controlling the washing tub to dehydrate at the set maximum dehydration speed;

[0012] When the highest non-oscillation speed does not reach the set highest dehydration speed, the washing tub is controlled to be dehydrated at the highest non-oscillation speed and enters a dehydration speed increasing stage.

[0013] In some embodiments, the dehydration process is controlled according to the eccentricity value of the washing tub during the step-by-step increase in the dehydration speed.

[0014] In some embodiments, controlling the dehydration process according to the eccentricity value of the washing tub during the step-by-step increase in the dehydration speed includes:

[0015] Control the dehydration speed of the washing drum to increase the preset speed;

[0016] Obtaining the eccentricity value of the washing drum, and determining the size of the eccentricity value and the set eccentricity value;

[0017] When the eccentricity value is greater than a set eccentricity value, controlling the washing machine to enter a shaking program and / or an unwinding program;

[0018] When the eccentricity value is less than or equal to the set eccentricity value, the step of controlling the dehydration speed of the washing tub to increase the preset speed is entered again until the set maximum dehydration speed is reached.

[0019] In some embodiments, after the shaking process and / or the unwinding process, the control method further includes:

[0020] Cumulative number of dehydration failures;

[0021] When the accumulated number of dehydration failures reaches the set number, the current dehydration process is skipped and the subsequent clothing processing program is entered;

[0022] When the accumulated number of dehydration failures does not reach the set number, the process returns to the step of acquiring a plurality of images of clothes in the washing tub at different viewing angles.

[0023] In some embodiments, the multiple clothing images from different perspectives include at least:

[0024] A first image of clothes taken from a perspective along the axial direction of the washing tub, and a second image of clothes taken from a perspective from an upper end on one side to a lower end on the opposite side in the axial direction of the washing tub.

[0025] In some embodiments, the processing of the plurality of clothing images using a pre-trained recognition model to determine the stacking height, spread area, and distribution parameters of the clothing in the washing tub includes:

[0026] performing image segmentation processing on the first clothing image using a pre-trained recognition model to identify the stacking height of the clothing in the washing tub;

[0027] The second clothing image is subjected to image segmentation processing using a pre-trained recognition model to identify the expanded area and distribution state parameters of the clothing in the washing tub.

[0028] In some embodiments, the training process of the recognition model includes:

[0029] Obtaining a training data set, wherein the training data set includes multiple clothing image samples from different perspectives in a washing drum, each clothing image sample having a status label for marking status information of the clothing in the washing drum, the status information including stacking height, spread area, and distribution parameters of the clothing in the washing drum;

[0030] The initial recognition model is trained a preset number of times using the training data set, and after each training, the training results are compared with the state labels;

[0031] If the comparison result does not meet the preset training conditions, the model parameters of the initial recognition model are adjusted to obtain an updated initial recognition model, and the initial recognition model is trained again for a preset number of times using the training data set;

[0032] When the comparison result meets the preset training conditions or the number of training times reaches the preset number, the training of the initial recognition model is completed, and a recognition model is obtained that can output the status information of the clothing in the input image after learning based on the input image.

[0033] The second aspect of the present invention proposes a control device, which includes one or more processors and a non-transitory computer-readable storage medium storing program instructions. When the one or more processors execute the program instructions, the one or more processors are used to implement any image recognition-based washing machine dehydration control method proposed in the first aspect of the present invention.

[0034] A third aspect of the present invention provides a washing machine, which operates according to any one of the image recognition-based washing machine dehydration control methods provided in the first aspect of the present invention, or includes the control device provided in the second aspect of the present invention.

[0035] The technical solution of the present invention can achieve the following beneficial effects: by acquiring images of the laundry inside the washing drum from different perspectives, the present invention identifies the stacking height, spread area, and distribution parameters of the laundry within the washing drum. Then, through machine learning, a neural network model is trained to predict the maximum non-oscillating speed corresponding to different laundry loads and clothing states, enabling the washing machine to quickly reach the appropriate spin speed. When the laundry load is minimal, the machine can directly spin at the set maximum spin speed. When the laundry load is large, the machine can first spin at the predicted, preset maximum non-oscillating speed, and then gradually increase the speed through eccentricity detection to quickly reach the maximum speed. This significantly shortens spin time, improves dehydration efficiency, and enhances dehydration results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0037] Figure 1 This is one of the dehydration flow charts of a washing machine according to an exemplary embodiment.

[0038] Figure 2 This is a second dehydration flow chart of a washing machine according to an exemplary embodiment. DETAILED DESCRIPTION

[0039] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0040] During the dehydration process, when the amount of laundry and the distribution of the laundry are constant, the washing machine will not oscillate before reaching a certain speed. In particular, when the amount of laundry is very small, the dehydration speed can be directly increased to the set maximum dehydration speed regardless of the current state of the laundry in the drum, without oscillation or drum collision. Secondly, if the current amount of laundry and its distribution state may cause oscillation, when an appropriate speed is given, the eccentricity of the drum can still be guaranteed not to exceed the specified range at that speed, allowing the washing drum to be directly increased to that speed and enter a speed-increasing state until the maximum speed is reached. In addition, when the laundry in the washing drum has the same stacking height h and spread area s but different distribution states in the drum, the corresponding maximum non-oscillation speed will also be different. Therefore, this embodiment proposes a washing machine dehydration control method, control device, and washing machine based on image recognition to predict the maximum non-oscillation dehydration speed of the washing drum based on the current load and clothing distribution state, thereby significantly shortening the time it takes for the dehydration speed to increase to the maximum speed, thereby improving dehydration efficiency and dehydration effect.

[0041] Figure 1 A washing machine dehydration control method based on image recognition is shown according to an exemplary embodiment. Figure 1 The dehydration control method of this embodiment includes the following steps:

[0042] S11 . After the drainage is completed and before the dehydration begins, a plurality of images of the clothes in the washing tub are acquired from different perspectives.

[0043] In this embodiment, the washing machine includes multiple camera devices, each of which is configured to capture images of the laundry within the washing drum from different perspectives. In one example, the multiple camera devices include a first camera and a second camera. The first camera is positioned at the door of the washing drum, and its viewing angle is along the axial direction of the washing drum. The first camera captures a first image of the laundry within the washing drum. To ensure the accuracy of the captured images, multiple images of the first laundry are continuously captured after draining is completed and before dehydration begins. The second camera is positioned at the upper portion of the rear wall of the washing drum, and its viewing angle is from the upper end of one side to the lower end of the opposite side along the axial direction of the washing drum. The second camera captures a second image of the laundry within the washing drum. To ensure the accuracy of the captured images, multiple images of the second laundry are continuously captured after draining is completed and before dehydration begins. In subsequent steps, the stacking height of the laundry within the washing drum can be determined based on the first image of the laundry, and the expanded area and distribution parameters of the laundry within the washing drum can be determined based on the second image of the laundry.

[0044] In other feasible implementations, clothing images from different perspectives are not limited to the shooting perspectives of the first camera and the second camera. As long as the recognition of clothing stacking height, unfolded area and distribution status parameters can be achieved under other different shooting perspectives, they are all included in the solution of this embodiment.

[0045] S12. Processing multiple clothing images using a pre-trained recognition model to identify the stacking height, spread area, and distribution state parameters of the clothing in the washing tub.

[0046] In this embodiment, a recognition model is used to identify the state of clothing within a washing drum based on multiple clothing images. The recognition model can be a Mobile-Unet recognition model, or other recognition models. After obtaining multiple clothing images from different perspectives, each of the clothing images is trained using a pre-trained recognition model. In one example, when the multiple clothing images include a first clothing image viewed along the axial direction of the washing drum and a second clothing image viewed from the upper end of one side of the washing drum toward the lower end of the opposite side, the pre-trained recognition model performs image segmentation processing on the first clothing image to identify the respective segments between the clothing and the washing drum, thereby determining the stacking height of the clothing within the washing drum. The pre-trained recognition model also performs image segmentation processing on the second clothing image to identify the respective segments between the clothing and the washing drum, thereby determining the expanded area and distribution state parameters of the clothing within the washing drum. The distribution state parameters refer to the distribution state of the clothing at the same stacking height and expanded area. Distribution states θ include θ1, θ2, θ3, etc. For example, θ1 represents a balanced distribution of clothing, θ2 represents a distribution with larger clothing on the left and smaller clothing on the right, and θ3 represents a distribution with larger clothing on the top and smaller clothing on the bottom.

[0047] S13. Input the identified stack height, expansion area, and distribution state parameters into a pre-trained neural network model to determine the maximum non-oscillation rotation speed for the laundry treatment drum to start dehydration under the current load.

[0048] In this embodiment, after the clothing image is calculated through the recognition model to identify the different stacking heights h, expansion areas s and distribution state parameters θ of the clothing occupying the inner drum, the h, s, θ values ​​obtained by the model calculation of a large number of clothing images and the highest non-oscillation speed obtained by the experiment can be used as a data set to input the network to train the convolutional neural network. Then, the h, s, θ values ​​of multiple clothing images from different perspectives obtained after the actual drainage is completed but before the dehydration begins are input into the neural network for prediction through the pre-trained recognition model to obtain the highest non-oscillation speed corresponding to the current amount of clothing and clothing state. The unit of the stacking height h is CM or percentage, and the unit of the expansion area s is CM 2 Or percentage, distribution state parameter θ value, for example, is balanced, larger on the left and smaller on the right, larger on the top and smaller on the bottom, etc.

[0049] S14. Determine the dehydration process of the washing tub according to the highest non-oscillation speed.

[0050] In this embodiment, after determining the maximum non-oscillation speed of the washing machine, the actual spin speed of the washing drum can be determined based on the maximum non-oscillation speed and the set maximum spin speed allowed by the washing machine. Specifically, it is first determined whether the maximum non-oscillation speed has reached the set maximum spin speed. If the maximum non-oscillation speed has reached the set maximum spin speed, the washing drum is controlled to be dehydrated at the set maximum spin speed. If the maximum non-oscillation speed has not reached the set maximum spin speed, the washing drum is controlled to be dehydrated at the maximum non-oscillation speed. This embodiment allows dehydration to be performed directly at the set maximum spin speed when the amount of clothes is very small. When the amount of clothes is large, dehydration can be performed first according to the predicted maximum non-oscillation speed as the initial speed, and then enter the stage of gradually increasing the spin speed. By gradually increasing the speed to quickly reach the maximum speed, the dehydration time is greatly shortened, the dehydration efficiency is improved, and the dehydration effect is improved.

[0051] In some embodiments, the dehydration process is controlled according to the eccentricity value of the washing tub during the stage of gradually increasing the dehydration speed, so as to avoid oscillation and collision with the tub during the process of gradually increasing the speed.

[0052] In some embodiments, the dehydration process is controlled according to the eccentricity value of the washing drum during the stage of gradually increasing the dehydration speed, including: controlling the dehydration speed of the washing drum to increase to a preset speed; obtaining the eccentricity value of the washing drum, and determining the size of the eccentricity value and the set eccentricity value; when the eccentricity value is greater than the set eccentricity value, controlling the washing machine to enter the shaking program and / or the unwinding program; when the eccentricity value is less than or equal to the set eccentricity value, again entering the step of controlling the dehydration speed of the washing drum to increase to a preset speed until the set maximum dehydration speed is reached.

[0053] Specifically, when the predicted maximum non-oscillation speed is less than the set maximum dehydration speed, the washing drum is first controlled to run directly at the predicted maximum non-oscillation speed, and then the preset speed is increased based on the maximum non-oscillation speed. After the speed of the washing drum is increased, the eccentricity value of the washing drum is obtained, and the eccentricity value is compared with the set eccentricity value corresponding to the current weight of the clothes. If the detected eccentricity value is greater than the set eccentricity value, the dehydration speed cannot be further increased, and the clothes need to be shaken and / or unwound to avoid collision or vibration. Among them, the shaking program is to control the washing drum to rotate alternately forward and reverse at a first speed and for a first time period, and the unwinding program is to control the washing drum to decelerate to 0, and then control the washing drum to rotate in the opposite direction at a second speed and for a second time period, and relax the set eccentricity value range of the eccentricity detection. In this embodiment, when the detected eccentricity value is greater than the set eccentricity value, the shaking program and the unwinding program can be executed simultaneously, or only one of the shaking program and the unwinding program can be executed. After completing the shaking and unwinding programs, the maximum non-oscillation speed will also change due to changes in the stacking height, unfolding area and distribution state parameters of the clothes in the washing drum. At this time, it is necessary to re-predict the maximum non-oscillation speed of the washing drum and determine the dehydration process of the washing drum according to the newly predicted maximum non-oscillation speed.

[0054] In some embodiments, after the shaking program and / or the unwinding program, the control method further includes: accumulating the number of dehydration failures; when the accumulated number of dehydration failures reaches a set number, skipping the current dehydration process and entering the subsequent clothing processing program; when the accumulated number of dehydration failures does not reach the set number, returning to the step of obtaining multiple clothing images from different perspectives in the washing drum.

[0055] Specifically, after the washing drum speed is increased to a set value, if the detected washing drum eccentricity value is greater than the set eccentricity value, the spin speed cannot be increased further, indicating that the spin cycle has failed. The previous spin cycle is repeated after shaking and / or untangling the clothes. To avoid endless repetition of the spin cycle after a spin cycle failure, which would extend the spin cycle time, a limit is set, for example, 3 or 5 spin cycle failures. When the number of spin cycle failures reaches the set number, the previous spin cycle is not repeated, but the current spin cycle is skipped and the subsequent laundry treatment program is run. If the current spin cycle is a main wash or rinse spin cycle, the main wash or rinse spin cycle is skipped and the subsequent program is run. If the current spin cycle is a final spin cycle, if the washing machine is only washing clothes, the final spin cycle is skipped, the laundry cycle ends, and a notification message indicating that the spin cycle failed is sent to the user. If the washing machine also includes a drying cycle or a laundry care program, the final spin cycle is skipped and the drying cycle or laundry treatment program is entered. After all laundry treatment programs are completed, a notification message indicating that the spin cycle failed is sent to the user.

[0056] In some embodiments, the training process of the recognition model includes: obtaining a training data set, wherein the training data set includes multiple clothing image samples from different perspectives in a washing drum, each clothing image sample has a status label for marking the status information of the clothing in the washing drum, and the status information includes the stacking height, unfolded area and distribution status parameters of the clothing in the washing drum; training the initial recognition model a preset number of times using the training data set, and after each training, comparing the training result with the status label; if the comparison result does not meet the preset training conditions, adjusting the model parameters of the initial recognition model to obtain an updated initial recognition model, and training the initial recognition model again a preset number of times using the training data set; when the comparison result meets the preset training conditions or the number of training times reaches the preset number, completing the training of the initial recognition model, and obtaining a recognition model that can output the status information of the clothing in the input image after learning based on the input image.

[0057] Specifically, in the training process of the recognition model, a large number of clothing images from different perspectives of the clothes in the washing drum are first obtained, and then the clothing status information in each clothing image is annotated using software such as Labelme to obtain a large number of annotated images with status labels. Then, a large number of annotated images are used to train a recognition model such as MobileU-net. The trained model can calculate the given actual clothing images taken from different perspectives and identify the different stacking heights h, expanded areas s and distribution state parameters θ of the clothes occupying the inner drum.

[0058] After determining the stack height h, spread area s, and distribution parameter θ of the clothes in the washing drum, the identified stack height h, spread area s, and distribution parameter θ are input into a pre-trained neural network model, and the pre-trained neural network model can output the highest non-oscillation speed.

[0059] During the training phase of the convolutional neural network, the stack height h, spread area s, distribution parameter θ of the laundry in the washing tub, and the optimal preset speed obtained from the experiment, identified by the semantic segmentation model, are first input into the network to train the convolutional neural network. The parameters for the training phase of the convolutional neural network are shown in Table 1:

[0060] Table 1:

[0061]

[0062]

[0063] In Table 1, h1, h2, h3... are the clothing stacking heights h corresponding to the clothing in different clothing images, s1, s2, s3... are the clothing unfolding areas s corresponding to the clothing in different clothing images, θ1, θ2, θ2... are the distribution state parameters θ corresponding to the clothing in different clothing images, and V1, V2, V3... are the maximum non-oscillation speeds V corresponding to the stacking height, unfolding area and distribution state parameters.

[0064] During the actual dehydration process, the measured h, s, and θ values ​​corresponding to the current load are input into the neural network for prediction, resulting in the maximum non-oscillation speed corresponding to the current clothing stack height, spread area, and clothing state. Table 2 shows the maximum non-oscillation speed predicted by the neural network model for different stack height, spread area, and distribution state parameters, according to an exemplary embodiment.

[0065] Table 2:

[0066]

[0067] The above examples are for demonstration purposes only, and the experimental results during specific implementation are irrelevant to the results shown in this example.

[0068] Figure 2 A washing machine dehydration control method based on image recognition is shown according to an exemplary embodiment. Figure 2 The dehydration control method of this embodiment includes the following steps:

[0069] S201, acquiring multiple images of clothes in a washing drum from different perspectives;

[0070] S202, performing semantic segmentation and recognition on multiple clothing images using a pre-trained recognition model to obtain the stacking height h, spread area s, and distribution state parameter θ of the clothing in the washing drum;

[0071] S203, inputting h, s, and θ into a pre-trained convolutional neural network to determine the highest non-oscillation speed;

[0072] S204, determining whether the maximum non-oscillation speed is greater than or equal to the set maximum spin speed. If so, the washing drum is controlled to rotate at the set maximum spin speed until the spin is completed. If not, proceed to S205;

[0073] S205, controlling the washing drum to rotate at the highest non-oscillating speed;

[0074] S206, increasing the preset speed Vt based on the current speed;

[0075] S207, obtaining the eccentricity value of the washing drum;

[0076] S208, determine whether the eccentricity value is greater than the set eccentricity value, if the judgment result is yes, go to S209, if the judgment result is no, go to S210;

[0077] S209, enter the shaking program and the unwinding program, and then enter S211;

[0078] S210, determining whether the current speed reaches the set maximum spin speed. If yes, run at the set maximum spin speed until the spin is finished. If no, return to S206;

[0079] S211, cumulative number of dehydration failures n;

[0080] S212, determining whether the number of dehydration failures reaches the set number; if yes, skip the current dehydration process and enter the subsequent clothing processing program; if no, return to S201.

[0081] According to an exemplary embodiment, this embodiment proposes a control device, which includes one or more processors and a non-transitory computer-readable storage medium storing program instructions. When the one or more processors execute the program instructions, the one or more processors are used to implement any of the image recognition-based washing machine dehydration control methods proposed in the above embodiments.

[0082] According to an exemplary embodiment, this embodiment proposes a washing machine, which operates according to any of the washing machine dehydration control methods based on image recognition proposed in the above embodiments, or includes the control device proposed in the above embodiments.

[0083] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0084] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0085] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0086] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0087] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0088] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0089] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0090] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A washing machine dehydration control method based on image recognition, characterized in that: The dehydration control method comprises: After draining is completed and before dehydration begins, multiple images of clothes in the washing drum are obtained from different perspectives; Processing the plurality of clothing images using a pre-trained recognition model to identify stacking height, spread area, and distribution parameters of the clothing in the washing tub; Inputting the identified stack height, the expanded area, and the distribution state parameters into a pre-trained neural network model to determine a maximum non-oscillating rotational speed for the laundry treatment drum to start dehydration with the current load; The dehydration process of the washing tub is determined according to the maximum non-oscillation speed.

2. The washing machine dehydration control method based on image recognition according to claim 1, characterized in that: The step of determining the dehydration process of the washing tub according to the maximum non-oscillation speed includes: Determining whether the maximum non-oscillation speed reaches the set maximum dehydration speed; When the maximum non-oscillation speed reaches the set maximum dehydration speed, controlling the washing tub to dehydrate at the set maximum dehydration speed; When the highest non-oscillation speed does not reach the set highest dehydration speed, the washing tub is controlled to be dehydrated at the highest non-oscillation speed and enters a dehydration speed increasing stage.

3. The washing machine dehydration control method based on image recognition according to claim 2, characterized in that: During the step-by-step increase in the spin speed, the spin process is controlled according to the eccentricity value of the washing tub.

4. The washing machine dehydration control method based on image recognition according to claim 3, characterized in that: The step of controlling the dehydration process according to the eccentricity value of the washing tub during the step-by-step increase in the dehydration speed comprises: Control the dehydration speed of the washing drum to increase the preset speed; Obtaining the eccentricity value of the washing drum, and determining the size of the eccentricity value and the set eccentricity value; When the eccentricity value is greater than a set eccentricity value, controlling the washing machine to enter a shaking program and / or an unwinding program; When the eccentricity value is less than or equal to the set eccentricity value, the step of controlling the dehydration speed of the washing tub to increase the preset speed is entered again until the set maximum dehydration speed is reached.

5. The washing machine dehydration control method based on image recognition according to claim 4, characterized in that: After the shaking process and / or the unwinding process, the control method further includes: Cumulative number of dehydration failures; When the accumulated number of dehydration failures reaches the set number, the current dehydration process is skipped and the subsequent clothing processing program is entered; When the accumulated number of dehydration failures does not reach the set number, the process returns to the step of acquiring a plurality of images of clothes in the washing tub at different viewing angles.

6. The washing machine dehydration control method based on image recognition according to claim 1, characterized in that: The multiple clothing images from different perspectives include at least: A first image of clothes taken from a perspective along the axial direction of the washing tub, and a second image of clothes taken from a perspective from an upper end on one side to a lower end on the opposite side in the axial direction of the washing tub.

7. The washing machine dehydration control method based on image recognition according to claim 6, characterized in that: The processing of the plurality of clothing images by a pre-trained recognition model to determine the stacking height, spread area, and distribution state parameters of the clothing in the washing tub includes: performing image segmentation processing on the first clothing image using a pre-trained recognition model to identify the stacking height of the clothing in the washing tub; The second clothing image is subjected to image segmentation processing using a pre-trained recognition model to identify the expanded area and distribution state parameters of the clothing in the washing tub.

8. The washing machine dehydration control method based on image recognition according to claim 1, characterized in that: The training process of the recognition model includes: Obtaining a training data set, wherein the training data set includes multiple clothing image samples from different perspectives in a washing drum, each clothing image sample having a status label for marking status information of the clothing in the washing drum, the status information including stacking height, spread area, and distribution parameters of the clothing in the washing drum; The initial recognition model is trained a preset number of times using the training data set, and after each training, the training results are compared with the state labels; If the comparison result does not meet the preset training conditions, the model parameters of the initial recognition model are adjusted to obtain an updated initial recognition model, and the initial recognition model is trained again for a preset number of times using the training data set; When the comparison result meets the preset training conditions or the number of training times reaches the preset number, the training of the initial recognition model is completed, and a recognition model is obtained that can output the status information of the clothing in the input image after learning based on the input image.

9. A control device, characterized in that: It includes one or more processors and a non-transitory computer-readable storage medium storing program instructions. When the one or more processors execute the program instructions, the one or more processors are used to implement the washing machine dehydration control method based on image recognition as described in any one of claims 1 to 8.

10. A washing machine, characterized in that: The washing machine operates according to the washing machine dehydration control method based on image recognition according to any one of claims 1 to 8, or includes the control device according to claim 9.

Images