A humidification care method and equipment

By using an image acquisition device in the steam care equipment to determine whether the exposed area of ​​the load has been humidified, and adjusting the load posture to achieve uniform steam spraying, the problem of uneven steam spraying in the prior art is solved, and the wrinkle removal and deodorization effects are improved.

CN123108913APending Publication Date: 2026-11-17GREE ELECTRIC APPLIANCE INC OF ZHUHAI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611586956.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing steam care equipment cannot guarantee the uniformity of steam spray during garment treatment, resulting in poor wrinkle removal and odor removal effects.

Method used

The image acquisition device acquires real-time image information of the load inside the containment cavity, determines whether the exposed area has been humidified, and adjusts the distribution or posture of the load according to the image information to expose the untreated area and achieve uniform steam spraying.

Benefits of technology

It achieves uniform steam spraying across the load, improves wrinkle removal and deodorization effects, avoids repeated humidification of the same area, and improves humidification accuracy and control robustness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123108913A_ABST
    Figure CN123108913A_ABST
Patent Text Reader

Abstract

This invention belongs to the technical field of nursing equipment, and particularly relates to a humidification nursing method and equipment. The nursing method includes: S1: controlling an image acquisition device to acquire image information of the load within a receiving cavity; S2: determining, based on the image information, whether the currently exposed area of ​​the load within the receiving cavity has been humidified by the humidification device; S3: if the currently exposed area has not been humidified, controlling the humidification device to humidify the currently exposed area; if the currently exposed area has been humidified, changing the distribution state or posture of the load to expose another area of ​​the load; S4: repeating S1 to S3 until a preset condition is met and the program ends. This invention, by determining whether the exposed area has been humidified through image analysis, can accurately analyze the steam adhesion of the load itself, avoiding repeated humidification of the same exposed area, thereby achieving uniform steam spraying of the entire load, improving the uniformity of steam nursing and the wrinkle-removing and deodorizing effects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of nursing equipment technology, and in particular relates to a humidification nursing method and equipment. Background Technology

[0002] Clothing care equipment, such as washing machines and garment dryers, typically includes a steam care function. Steam care works by delivering high-temperature steam into the drum, allowing the steam to penetrate the clothing fibers, thus smoothing out wrinkles and quickly removing odors such as hot pot smells and cooking fumes. Users can restore overnight clothes to a smooth state without washing them with water, making steam care widely popular.

[0003] In existing technologies, steam treatments typically control the atomizing device to spray mist into the roller according to a preset time, or optimize the treatment effect by controlling the airflow direction, flow rate, and treatment time of the auxiliary fan. However, these solutions do not consider the steam adhesion of the load itself. Since the load may be folded, tangled, or attached to the wall inside the roller, the exposed area of ​​the load may not change after the roller rotates, resulting in the same side being sprayed with steam repeatedly, while the side close to the bottom of the roller or the side folded inside is not sprayed with steam. This leads to uneven steam spraying, affecting the wrinkle removal and odor removal effects on clothing.

[0004] Therefore, existing technologies cannot guarantee that the load is uniformly adhered to by steam. Summary of the Invention

[0005] The purpose of this invention is to address the aforementioned technical problems by providing a humidification care method and device, which achieves the desired effect.

[0006] In view of this, the present invention provides a humidification care method for a care device, the care device comprising a receiving cavity, a humidification device, and an image acquisition device, the method comprising: S1: Control the image acquisition device to acquire image information of the load inside the cavity; S2: Determine, based on the image information, whether the currently exposed area of ​​the load within the containment cavity has been humidified by the humidification device; S3: If the current exposed area has not been humidified, control the humidification device to humidify the current exposed area; if the current exposed area has been humidified, change the distribution state or orientation of the load to expose another area of ​​the load. S4: Repeat S1 to S3 until the preset conditions are met and the program ends.

[0007] In this technical solution, during the humidification process, the image acquisition device continuously or periodically acquires image information of the load within the containment cavity. The controller analyzes the exposed area of ​​the load currently facing the humidification device based on the image information and determines whether this exposed area has been previously humidified. If it has not been treated, it indicates that the area is a newly exposed surface, and the humidification device is controlled to humidify that area. If it has already been treated, it means that the area has already been sprayed with steam; repeated humidification would lead to localized over-humidification while other areas remain uncovered. Therefore, humidification is not performed; instead, the distribution or orientation of the load is changed, causing the load to flip, move, or unfold, thereby exposing another area. This process of acquiring images, making judgments, humidifying, or changing the state is repeated, and the cycle continues until the humidification program ends.

[0008] Its beneficial effects are: by judging whether the exposed area has been humidified by the image, it can accurately analyze the steam adhesion of the load itself, avoid the same exposed area being repeatedly humidified, and promote different areas of the load to be exposed one after the other and humidified in sequence, so as to achieve uniform steam spraying of the entire load, and improve the uniformity of steam care and the wrinkle removal and deodorization effects.

[0009] Furthermore, determining whether the exposed area of ​​the load has been humidified based on the image information includes: The current load area image is determined based on the image information; The current load area image is compared with the historical load area images. The similarity comparison is to compare the current load area image with the load area images of exposed areas that have been previously humidified and stored in the historical image library. If the similarity between the current load area image and all historical load area images is less than a preset similarity threshold A, then it is determined that the current exposed area has not been humidified. If the similarity between the current load area image and any historical load area image is greater than or equal to the preset similarity threshold A, then it is determined that the current exposed area has been humidified.

[0010] In this technical solution, the working process is as follows: After each photo is taken, the current load area image is first determined from the image information, that is, the area of ​​the current load in the image. Then, the current load area image is compared one by one with each historical load area image stored in the historical image library. The historical image library stores images of exposed areas that have been previously humidified. If the current image is dissimilar to all historical images (similarity less than A), it means that the current exposed area is a new area that has not been processed before; if the current image is similar to a historical image (similarity greater than or equal to A), it means that the current exposed area has been humidified before. By comparing similarity, it is possible to quantitatively determine whether the current exposed area has been processed, avoiding the subjectivity and uncertainty of manual judgment, improving the accuracy of judgment, and thus ensuring that each humidification is applied to a new, unprocessed area, achieving uniform humidification.

[0011] Furthermore, controlling the humidification device to humidify the currently exposed area includes: The humidification device is controlled to perform initial humidification on the currently exposed area; After the initial humidification, a moisture content assessment was conducted. Based on the moisture content assessment results, it is determined whether the moisture content meets the preset requirements. If the preset requirements are not met, the humidification device is controlled to supplement humidification until the moisture content assessment results meet the preset requirements. If the moisture content meets the preset requirements, the process of controlling the humidification device to humidify the current exposed area ends.

[0012] In this technical solution, when it is determined that the exposed area has not been humidified, the humidification device is controlled to perform initial humidification on that area. After the initial humidification, the moisture content of the load is assessed to determine if it meets the preset requirements. If it does not, it indicates insufficient humidification, and humidification continues; after supplemental humidification, the assessment is repeated, and if it is still insufficient, supplemental humidification continues until the moisture content reaches the preset requirements. If the moisture content has reached the preset requirements after the initial humidification, the humidification process ends, and the area is considered to have been humidified. Through closed-loop control of initial humidification, moisture content assessment, and supplemental humidification, it is ensured that the moisture content of the load reaches the preset target each time humidification is performed. This avoids both insufficient humidification leading to poor wrinkle removal effects and excessive humidification leading to over-wetting of the load, thus improving humidification accuracy and treatment effectiveness.

[0013] Furthermore, the moisture content assessment includes: Before the first execution of S1, the near-infrared camera is controlled to acquire a dry near-infrared image of the load, and the average gray level G0 of the dry load region when the load is in a dry state is obtained based on the dry near-infrared image. After each humidification, the near-infrared camera is controlled to acquire the current near-infrared image and the average gray level Gt of the current load area is acquired; Calculate the grayscale difference GΔ = G0 - Gt; If GΔ is greater than the moisture content analysis threshold G, then the moisture content meets the preset requirements; If GΔ is less than or equal to the moisture content analysis threshold G, then the moisture content has not met the preset requirements.

[0014] In this technical solution, before humidification begins, the load is in a dry state. A dry near-infrared image is captured using a near-infrared camera, and the average grayscale value G0 of the load area is extracted as the dry state baseline. After each humidification cycle, a new near-infrared image is captured, and the average grayscale value Gt of the current load area is extracted. Since water molecules absorb near-infrared light, the wetter the load, the lower the average grayscale value of the load area in the near-infrared image, thus resulting in a smaller Gt and a larger G0-Gt. GΔ = G0-Gt is calculated. If GΔ is greater than the threshold G, the moisture content has reached the preset requirement; if GΔ is less than or equal to G, the moisture content is insufficient, and further humidification is needed. Utilizing the characteristic absorption principle of water molecules to near-infrared light, the difference between the dry state baseline grayscale value G0 and the current grayscale value Gt characterizes the moisture content. This eliminates the influence of individual differences in load material, color, etc., on the grayscale value, improving the accuracy of moisture content judgment and ensuring precise and reliable humidification control.

[0015] Furthermore, in S4, the completion of the humidification program is determined by the number of humidification cycles, including: Set the effective humidification count N, and initialize the effective humidification count n=0; After the humidification treatment is performed and the moisture content assessment shows that the moisture content meets the preset requirements, n is incremented by 1. The humidification process ends when n is greater than or equal to N.

[0016] In this technical solution, before the humidification program begins, a preset effective humidification count N is set, and the current effective humidification count n is initialized to 0. Each time a humidification process is completed (i.e., the moisture content assessment reaches the preset requirement), n is incremented by 1, indicating that one effective humidification cycle has been completed. When n reaches or exceeds N, it indicates that the preset number of effective humidification cycles has been completed, and the humidification program ends; otherwise, the cycle continues from S1 to S3. Controlling the end of the humidification program by the effective humidification count ensures that the load undergoes a sufficient number of effective humidification cycles, allowing all areas to be humidified, preventing premature termination that leaves some areas uncovered, and also avoiding infinite loops, thus improving control efficiency.

[0017] Furthermore, if the current exposed area has been humidified, changing the distribution state or orientation of the load includes: Set an upper limit B for the number of consecutive similarities and a current number b for the number of consecutive similarities, where the initial value of b is 0; When it is determined that the current exposed area has been humidified, increment b by 1; If b is less than B, then return to the step of changing the distribution state or orientation of the load; If b is greater than or equal to B, perform a moisture content assessment on the current load. If the moisture content does not meet the preset requirements, supplement humidification until the moisture content assessment result meets the preset requirements. Then, increment n by 1 and set b to zero.

[0018] In this technical solution, when it is determined that the currently exposed area has been humidified, it means that no new area has been exposed after the roller rotates. At this time, the number of consecutive similarity counts, b, is incremented by 1. If b is less than B, it means that only once was a new area not exposed. The load state is changed again, and the image is taken again for judgment, in an attempt to expose a new area. If b is greater than or equal to B, it means that no new area has been exposed after multiple rotations, and the load may have been folded and tangled, making it impossible to unfold further. At this time, the moisture content of the current load is assessed. If the moisture content is insufficient, humidification is added until the preset requirements are met. Then, n is incremented by 1, b is set to zero, completing one effective humidification cycle, and the counting restarts. By using the upper limit B of the number of consecutive similarity counts, timely supplemental humidification and counting can be performed when the load cannot expose a new area, avoiding dead loops, while ensuring that the overall moisture content of the load meets the standard, improving control robustness and care effect.

[0019] Furthermore, the method also includes: when the current exposed area has not been humidified and humidification is performed, saving the current load area image to the historical image library to update the historical image library.

[0020] In this technical solution, after determining that the current exposed area has not been humidified and performing humidification, the image of the current loaded area is saved to the historical image library. Thus, the historical image library records images of all exposed areas that have been humidified. During the next image capture and assessment, the current image is compared with the updated historical image library to determine whether the current area has been processed. By continuously updating the historical image library, ensuring it always contains all processed areas, the accuracy of subsequent similarity comparisons is guaranteed, thereby achieving a complete record and accurate determination of the processing status of each area throughout the entire humidification process.

[0021] Further, obtaining the average grayscale G0 of the dry load region when the load is in a dry state includes: At the start of humidification, acquire RGB and near-infrared images of the load; The load contour is segmented using the RGB image to obtain a load region mask; The load area mask is applied to the near-infrared image, near-infrared pixels within the load area are extracted, and the average gray value is calculated as the average gray value G0 of the dry load area.

[0022] In this technical solution, the load is in a dry state at the start of humidification, and RGB and near-infrared images are captured simultaneously. Image segmentation is performed using the RGB images to identify the load's outline and generate a load area mask. This mask is applied to the near-infrared image to extract only the near-infrared pixels within the load area, eliminating background interference. The average grayscale value of these pixels is calculated as the dry-state baseline grayscale G0. Through RGB image segmentation and near-infrared image mask mapping, the near-infrared grayscale of the load area can be accurately extracted, eliminating interference from the drum background and improving the accuracy of G0, providing a reliable baseline for subsequent moisture content assessment.

[0023] Furthermore, changing the distribution state or orientation of the load includes: The bearing of the receiving cavity is controlled to rotate, thereby causing the load to flip and exposing another area of ​​the load.

[0024] In this technical solution, when the load state needs to be changed, the carrier component of the receiving cavity is controlled to rotate. The carrier component drives the load to flip, lift, and drop together, causing the load to change its position and orientation within the receiving cavity. This causes the surface that was originally facing the humidification device to turn away, exposing the previously blocked surface. The load flipping is achieved by rotating the carrier component, requiring no additional mechanisms, resulting in a simple structure, high reliability, and effective modification of the exposed load area, ensuring that each surface is humidified sequentially.

[0025] Furthermore, determining the current load region image based on the image information includes: The load contour is segmented from the image information using an image segmentation algorithm; Set the background area outside the load contour to zero and crop it to the bounding rectangle of the load contour to obtain the current load area image.

[0026] In this technical solution, the image information acquired by the image acquisition device is used to identify the outline of the load using an image segmentation algorithm. The pixel values ​​of the background area outside the outline are set to zero to eliminate background interference. Then, the circumscribed rectangle of the load outline is calculated, and the image is cropped within this rectangle to obtain the current load area image containing only the load area. Through image segmentation and cropping, interference from the drum background, inner cylinder wall, etc., is removed, allowing subsequent similarity comparisons to focus on the features of the load itself, thus improving the accuracy of the judgment.

[0027] Furthermore, the step of comparing the current load region image with historical load region images for similarity includes: The current load area image and the historical load area image are uniformly scaled to a preset resolution; The feature vectors of the current load region image and the historical load region image are extracted using a lightweight convolutional neural network. Calculate the cosine similarity between the current feature vector and the historical feature vectors to perform the similarity comparison.

[0028] In this technical solution, the current load area image and the historical load area image are first scaled to a preset resolution to ensure consistent size. Then, a lightweight convolutional neural network is used to extract features from both images, outputting their respective feature vectors. These feature vectors integrate global and local features such as color, texture, decoration, collar, and cuffs. Finally, the cosine similarity between the two feature vectors is calculated; a higher cosine similarity indicates greater similarity between the two images. By using a lightweight convolutional neural network and cosine similarity, it is possible to quickly and accurately determine whether two load area images are similar, thereby determining whether exposed areas have been processed. This approach requires minimal computation and is suitable for real-time operation in embedded controllers.

[0029] Furthermore, the response band of the near-infrared camera is around 1450nm or around 970nm; the nursing device also includes an active near-infrared light source and a narrow-band filter disposed in front of the lens of the near-infrared camera, the narrow-band filter being configured to correspond to the response band of the near-infrared camera.

[0030] In this technical solution, the response band of the near-infrared camera is selected near 1450nm or 970nm, which are sensitive to water molecule absorption. An active near-infrared light source illuminates the load with near-infrared light, improving image brightness. A narrow-band filter allows only near-infrared light of the corresponding band to pass through, filtering out ambient light interference. By selecting a band sensitive to water molecule absorption and combining it with an active light source and a narrow-band filter, the sensitivity and stability of water content detection can be improved, ambient light interference reduced, and reliable detection results ensured.

[0031] Furthermore, the image acquisition device includes an RGB camera, which has the same focal length, horizontal and vertical field of view, and imaging size as the near-infrared camera, and the installation distance between the RGB camera and the near-infrared camera is less than a preset distance.

[0032] In this technical solution, an RGB camera and a near-infrared camera are installed side-by-side, with identical focal lengths, horizontal and vertical field of view, and imaging sizes. The installation distance between them is less than a preset distance, ensuring that the images captured by the two cameras are essentially corresponding in coordinates. This allows the load region mask segmented from the RGB image to be directly mapped onto the near-infrared image, enabling the extraction of the near-infrared load region. By ensuring consistent parameters and close installation distance between the RGB and near-infrared cameras, the difference in coordinate correspondence between their images is reduced, improving mask mapping accuracy and thus guaranteeing the accuracy of water content analysis.

[0033] Furthermore, a garment processing device is characterized by comprising a receiving cavity, a humidifying device, an image acquisition device, and a controller, the controller being configured to perform the humidifying care method.

[0034] The beneficial effects of this invention are: 1. By judging whether the exposed area has been humidified through images, it can accurately analyze the steam adhesion of the load itself, avoid the same exposed area being repeatedly humidified, and promote different areas of the load to be exposed one after the other and humidified in sequence, so as to achieve uniform steam spraying of the entire load, improving the uniformity of steam care and the wrinkle removal and deodorization effects.

[0035] 2. Through a closed-loop control system of initial humidification, moisture content assessment, and supplementary humidification, the moisture content of the load is ensured to reach the preset target with each humidification. This avoids both insufficient humidification leading to poor wrinkle removal and excessive humidification leading to over-wetting of the load, thereby improving humidification accuracy and treatment effect.

[0036] 3. Utilizing the characteristic absorption principle of water molecules to near-infrared light, the moisture content is characterized by the difference between the dry reference gray level G0 and the current gray level Gt. This can eliminate the influence of individual differences in load material, color, etc. on gray level, improve the accuracy of moisture content judgment, and ensure precise and reliable humidification control.

[0037] 4. By using the upper limit B of consecutive similar counts, timely humidification and counting can be performed when the load cannot be further expanded to a new area, avoiding dead loops, while ensuring that the overall moisture content of the load meets the standard, thus improving control robustness and care effect.

[0038] 5. By using RGB image segmentation and near-infrared image masking, the near-infrared grayscale of the load area can be accurately extracted, eliminating interference from the drum background, improving the accuracy of G0, and providing a reliable benchmark for subsequent moisture content assessment. Attached Figure Description

[0039] Figure 1 This is a flowchart of the humidification care method. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0041] In the description of this application, it should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. For ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0042] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and are not limited in number; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0043] It should be noted that in the description of this application, the directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this application. The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.

[0044] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0045] A humidification care method for a care device, the care device including a receiving cavity, a humidification device, and an image acquisition device, the method comprising: S1: Control the image acquisition device to acquire image information of the load inside the cavity; S2: Determine, based on the image information, whether the currently exposed area of ​​the load within the containment cavity has been humidified by the humidification device; S3: If the current exposed area has not been humidified, control the humidification device to humidify the current exposed area; if the current exposed area has been humidified, change the distribution state or orientation of the load to expose another area of ​​the load. S4: Repeat S1 to S3 until the preset conditions are met and the program ends.

[0046] In this technical solution, during the humidification process, the image acquisition device continuously or periodically acquires image information of the load within the containment cavity. In this application, the "load" refers to fabrics such as clothing and bed sheets. The controller analyzes the exposed area of ​​the load currently facing the humidification device based on the image information and determines whether this exposed area has been previously humidified. If it has not been treated, it indicates that the area is a newly exposed surface, and the humidification device is controlled to humidify that area. If it has been treated, it indicates that the area has already been sprayed with steam; repeated humidification would lead to localized over-humidification while other areas remain uncovered. Therefore, humidification is not performed; instead, the distribution or orientation of the load is changed, causing the load to flip, move, or unfold, thereby exposing another area. The process of acquiring images, making judgments, humidifying, or changing the state is repeated, and this cycle continues until the humidification program ends.

[0047] Its beneficial effects are as follows: by judging whether the exposed area has been humidified by the image, it can accurately analyze the steam adhesion of the load itself, avoid the same exposed area being repeatedly humidified, and promote different areas of the load to be exposed and humidified in sequence, thereby achieving uniform steam spraying of the entire load, improving the uniformity of steam care and the wrinkle removal and deodorization effects.

[0048] Furthermore, determining whether the exposed area of ​​the load has been humidified based on the image information includes: The current load area image is determined based on the image information; The current load area image is compared with the historical load area image, wherein the historical load area image is a load area image of the exposed area that has been humidified and stored in the historical image library; If the similarity between the current load area image and all historical load area images is less than a preset similarity threshold A, then it is determined that the current exposed area has not been humidified. If the similarity between the current load area image and any historical load area image is greater than or equal to the preset similarity threshold A, then it is determined that the current exposed area has been humidified.

[0049] In this technical solution, the working process is as follows: After each photo is taken, the current load area image is first determined from the image information, that is, the area of ​​the current load in the image. Then, the current load area image is compared one by one with each historical load area image stored in the historical image library. The historical image library stores images of exposed areas that have been previously humidified. If the current image is dissimilar to all historical images (similarity less than A), it means that the current exposed area is a new area that has not been processed before; if the current image is similar to a historical image (similarity greater than or equal to A), it means that the current exposed area has been humidified before. By comparing similarity, it is possible to quantitatively determine whether the current exposed area has been processed, avoiding the subjectivity and uncertainty of manual judgment, improving the accuracy of judgment, and thus ensuring that each humidification is applied to a new, unprocessed area, achieving uniform humidification.

[0050] Furthermore, controlling the humidification device to humidify the currently exposed area includes: The humidification device is controlled to perform initial humidification on the currently exposed area; After the initial humidification, a moisture content assessment was conducted. Based on the moisture content assessment results, determine whether the moisture content meets the preset requirements. If the moisture content does not meet the preset requirements, supplemental humidification is performed until the moisture content assessment results meet the preset requirements. If the moisture content meets the preset requirements, the process of controlling the humidification device to humidify the current exposed area ends.

[0051] In this technical solution, when it is determined that the exposed area has not been humidified, the humidification device is controlled to perform initial humidification on that area. After the initial humidification, the moisture content of the load is assessed to determine if it meets the preset requirements. If it does not, it indicates insufficient humidification, and humidification continues; after supplemental humidification, the assessment is repeated, and if it is still insufficient, supplemental humidification continues until the moisture content reaches the preset requirements. If the moisture content has reached the preset requirements after the initial humidification, the humidification process ends, and the area is considered to have been humidified. Through closed-loop control of initial humidification, moisture content assessment, and supplemental humidification, it is ensured that the moisture content of the load reaches the preset target each time humidification is performed. This avoids both insufficient humidification leading to poor wrinkle removal effects and excessive humidification leading to over-wetting of the load, thus improving humidification accuracy and treatment effectiveness.

[0052] Furthermore, the moisture content assessment includes: Before the first execution of S1, the near-infrared camera is controlled to acquire a dry near-infrared image of the load, and the average gray level G0 of the dry load region when the load is in a dry state is obtained based on the dry near-infrared image. After each humidification, the near-infrared camera is controlled to acquire the current near-infrared image and the average gray level Gt of the current load area is acquired; Calculate the grayscale difference GΔ = G0 - Gt; If GΔ is greater than the moisture content analysis threshold G, then the moisture content meets the preset requirements; If GΔ is less than or equal to the moisture content analysis threshold G, then the moisture content has not met the preset requirements.

[0053] In this technical solution, before humidification begins, the load is in a dry state. A dry near-infrared image is captured using a near-infrared camera, and the average grayscale value G0 of the load area is extracted as the dry state baseline. After each humidification cycle, a new near-infrared image is captured, and the average grayscale value Gt of the current load area is extracted. Since water molecules absorb near-infrared light, the wetter the load, the lower the average grayscale value of the load area in the near-infrared image, thus resulting in a smaller Gt and a larger G0-Gt. GΔ = G0-Gt is calculated. If GΔ is greater than the threshold G, the moisture content has reached the preset requirement; if GΔ is less than or equal to G, the moisture content is insufficient, and further humidification is needed. Utilizing the characteristic absorption principle of water molecules to near-infrared light, the difference between the dry state baseline grayscale value G0 and the current grayscale value Gt characterizes the moisture content. This eliminates the influence of individual differences in load material, color, etc., on the grayscale value, improving the accuracy of moisture content judgment and ensuring precise and reliable humidification control.

[0054] Furthermore, in S4, the completion of the humidification program is determined by the number of humidification cycles, including: Set the effective humidification count N, and initialize the effective humidification count n=0; After the humidification treatment is performed and the moisture content assessment shows that the moisture content meets the preset requirements, n is incremented by 1. The humidification process ends when n is greater than or equal to N.

[0055] In this technical solution, before the humidification program begins, a preset effective humidification count N is set, and the current effective humidification count n is initialized to 0. Each time a humidification process is completed (i.e., the moisture content assessment reaches the preset requirement), n is incremented by 1, indicating that one effective humidification cycle has been completed. When n reaches or exceeds N, it indicates that the preset number of effective humidification cycles has been completed, and the humidification program ends; otherwise, the cycle continues from S1 to S3. Controlling the end of the humidification program by the effective humidification count ensures that the load undergoes a sufficient number of effective humidification cycles, allowing all areas to be humidified, preventing premature termination that leaves some areas uncovered, and also avoiding infinite loops, thus improving control efficiency.

[0056] Furthermore, if the current exposed area has been humidified, changing the distribution state or orientation of the load includes: Set an upper limit B for the number of consecutive similarities and a current number b for the number of consecutive similarities, where the initial value of b is 0; When it is determined that the current exposed area has been humidified, increment b by 1; If b is less than B, then return to the step of changing the distribution state or orientation of the load; If b is greater than or equal to B, perform a moisture content assessment on the current load. If the moisture content does not meet the preset requirements, supplement humidification until the moisture content assessment result meets the preset requirements. Then, increment n by 1 and set b to zero.

[0057] In this technical solution, when it is determined that the currently exposed area has been humidified, it means that no new area has been exposed after the roller rotates. At this time, the number of consecutive similarity counts, b, is incremented by 1. If b is less than B, it means that only once was a new area not exposed. The load state is changed again, and the image is taken again for judgment, in an attempt to expose a new area. If b is greater than or equal to B, it means that no new area has been exposed after multiple rotations, and the load may have been folded and tangled, making it impossible to unfold further. At this time, the moisture content of the current load is assessed. If the moisture content is insufficient, humidification is added until the preset requirements are met. Then, n is incremented by 1, b is set to zero, completing one effective humidification cycle, and the counting restarts. By using the upper limit B of the number of consecutive similarity counts, timely supplemental humidification and counting can be performed when the load cannot expose a new area, avoiding dead loops, while ensuring that the overall moisture content of the load meets the standard, improving control robustness and care effect.

[0058] Furthermore, the method also includes: when the current exposed area has not been humidified and humidification is performed, saving the current load area image to the historical image library to update the historical image library.

[0059] In this technical solution, after determining that the current exposed area has not been humidified and performing humidification, the image of the current loaded area is saved to the historical image library. Thus, the historical image library records images of all exposed areas that have been humidified. During the next image capture and assessment, the current image is compared with the updated historical image library to determine whether the current area has been processed. By continuously updating the historical image library, ensuring it always contains all processed areas, the accuracy of subsequent similarity comparisons is guaranteed, thereby achieving a complete record and accurate determination of the processing status of each area throughout the entire humidification process.

[0060] Further, obtaining the average grayscale G0 of the dry load region when the load is in a dry state includes: At the start of humidification, acquire RGB and near-infrared images of the load; The load contour is segmented using the RGB image to obtain a load region mask; The load area mask is applied to the near-infrared image, near-infrared pixels within the load area are extracted, and the average gray value is calculated as the average gray value G0 of the dry load area.

[0061] In this technical solution, the load is in a dry state at the start of humidification, and RGB and near-infrared images are captured simultaneously. Image segmentation is performed using the RGB images to identify the load's outline and generate a load area mask. This mask is applied to the near-infrared image to extract only the near-infrared pixels within the load area, eliminating background interference. The average grayscale value of these pixels is calculated as the dry-state baseline grayscale G0. Through RGB image segmentation and near-infrared image mask mapping, the near-infrared grayscale of the load area can be accurately extracted, eliminating interference from the drum background and improving the accuracy of G0, providing a reliable baseline for subsequent moisture content assessment.

[0062] Furthermore, changing the distribution state or orientation of the load includes: The bearing of the receiving cavity is controlled to rotate, thereby causing the load to flip and exposing another area of ​​the load.

[0063] In this technical solution, when the load state needs to be changed, the carrier component of the receiving cavity is controlled to rotate. The carrier component drives the load to flip, lift, and drop together, causing the load to change its position and orientation within the receiving cavity. This causes the surface that was originally facing the humidification device to turn away, exposing the previously blocked surface. The load flipping is achieved by rotating the carrier component, requiring no additional mechanisms, resulting in a simple structure, high reliability, and effective modification of the exposed load area, ensuring that each surface is humidified sequentially.

[0064] Furthermore, determining the current load region image based on the image information includes: The load contour is segmented from the image information using an image segmentation algorithm; Set the background area outside the load contour to zero and crop it to the bounding rectangle of the load contour to obtain the current load area image.

[0065] In this technical solution, the image information acquired by the image acquisition device is used to identify the outline of the load using an image segmentation algorithm. The pixel values ​​of the background area outside the outline are set to zero to eliminate background interference. Then, the circumscribed rectangle of the load outline is calculated, and the image is cropped within this rectangle to obtain the current load area image containing only the load area. Through image segmentation and cropping, interference from the drum background, inner cylinder wall, etc., is removed, allowing subsequent similarity comparisons to focus on the features of the load itself, thus improving the accuracy of the judgment.

[0066] Furthermore, the step of comparing the current load region image with historical load region images for similarity includes: The current load area image and the historical load area image are uniformly scaled to a preset resolution; The feature vectors of the current load region image and the historical load region image are extracted using a lightweight convolutional neural network. Calculate the cosine similarity between the current feature vector and the historical feature vectors to perform the similarity comparison.

[0067] In this technical solution, the current load area image and the historical load area image are first scaled to a preset resolution to ensure consistent size. Then, a lightweight convolutional neural network is used to extract features from both images, outputting their respective feature vectors. These feature vectors integrate global and local features such as color, texture, decoration, collar, and cuffs. Finally, the cosine similarity between the two feature vectors is calculated; a higher cosine similarity indicates greater similarity between the two images. By using a lightweight convolutional neural network and cosine similarity, it is possible to quickly and accurately determine whether two load area images are similar, thereby determining whether exposed areas have been processed. This approach requires minimal computation and is suitable for real-time operation in embedded controllers.

[0068] Furthermore, the response band of the near-infrared camera is around 1450nm or around 970nm; the nursing device also includes an active near-infrared light source and a narrow-band filter disposed in front of the lens of the near-infrared camera, the narrow-band filter being configured to correspond to the response band of the near-infrared camera.

[0069] In this technical solution, the response band of the near-infrared camera is selected near 1450nm or 970nm, which are sensitive to water molecule absorption. An active near-infrared light source illuminates the load with near-infrared light, improving image brightness. A narrow-band filter allows only near-infrared light of the corresponding band to pass through, filtering out ambient light interference. By selecting a band sensitive to water molecule absorption and combining it with an active light source and a narrow-band filter, the sensitivity and stability of water content detection can be improved, ambient light interference reduced, and reliable detection results ensured.

[0070] Furthermore, the image acquisition device includes an RGB camera, which has the same focal length, horizontal and vertical field of view, and imaging size as the near-infrared camera, and the installation distance between the RGB camera and the near-infrared camera is less than a preset distance.

[0071] In this technical solution, an RGB camera and a near-infrared camera are installed side-by-side, with identical focal lengths, horizontal and vertical field of view, and imaging sizes. The installation distance between them is less than a preset distance, ensuring that the images captured by the two cameras are essentially corresponding in coordinates. This allows the load region mask segmented from the RGB image to be directly mapped onto the near-infrared image, enabling the extraction of the near-infrared load region. By ensuring consistent parameters and close installation distance between the RGB and near-infrared cameras, the difference in coordinate correspondence between their images is reduced, improving mask mapping accuracy and thus guaranteeing the accuracy of water content analysis.

[0072] Furthermore, a garment processing device is characterized by comprising a receiving cavity, a humidifying device, an image acquisition device, and a controller, the controller being configured to perform the humidifying care method.

[0073] The specific work process is as follows: Symbol explanation: T1 is the preset single spray time, T2 is the supplementary spray time, G is the moisture content analysis threshold, A is the similarity threshold, B is the upper limit of consecutive similarity, b is the current consecutive similarity, N is the preset total number of effective sprays, and n is the current number of effective sprays.

[0074] Figure 1 The humidification process proposed in this patent is illustrated in the following flowchart, with detailed implementation steps: S1. Initialization An RGB camera and a near-infrared camera are arranged in the inner drum of the washing machine, and it is necessary to ensure that the parameters of the two cameras are as consistent as possible, such as focal length, horizontal and vertical field of view angles, imaging size, etc. The installation distance should also be as close as possible to reduce the coordinate correspondence difference between the images captured by the two cameras. When the care process starts, set the effective spray times n=1 and the consecutive similarity times b=0. Synchronously capture the first frame of RGB image and near-infrared image; segment the load contour based on the RGB image using image recognition technology, set the background outside the contour to 0 and crop it to the circumscribed rectangle to obtain the RGB load image and store it in the RGB image library; process the near-infrared image using the same contour to obtain the dry-state near-infrared reference image of the load, and calculate the average gray value G0 of the dry-state load area.

[0075] S2. First spray and water content analysis closed loop Directly perform the first steam spray for a preset time T1. After the spray ends, synchronously capture RGB and near-infrared images, extract the current near-infrared load area using the current RGB load contour, and calculate the current gray index Gt; judge whether the water content meets the preset condition according to the change G△=G0-Gt of Gt relative to G0, where it is determined to be satisfied when G△>G, otherwise it is not satisfied. If not satisfied, add additional spray for T2 and re-detect until the condition is met.

[0076] S3. Change the load state Control the drum to rotate for a preset angle or time to change the load distribution; synchronously capture RGB and near-infrared images, and perform segmentation to obtain the current RGB load image.

[0077] S4. Similarity judgment Compare the similarity of the current RGB load image with all images in the RGB image library one by one. If the similarity between the current image and all images in the image library is less than the threshold A, it is determined as a new load state, and the process proceeds to S5; if any similarity is greater than or equal to A, it is determined as a covered state, and the process proceeds to S6.

[0078] S5. New state processing Add the current RGB load image to the RGB image library; perform steam spray for T1 and water content closed-loop detection, and add additional spray for T2 when the preset condition is not met until the condition is satisfied; after completion, set n=n+1 and b=0. If n≥N, the humidification process ends; otherwise, return to S3.

[0079] S6. Covered state processing Set b=b+1. If b<B, return to S3 to continue changing the load state; if b≥B, detect the water content of the current state, and if the preset condition is not met, add additional spray for T2 until the condition is satisfied; then set n=n+1 and b=0. If n≥N, the humidification process ends; otherwise, return to S3.

[0080] Method description: Image segmentation: Load segmentation can be achieved based on algorithms such as YOLO26-SEG. This can be achieved by taking a large number of images with a set-up RGB camera, creating labels, building a dataset, and completing the training.

[0081] Similarity comparison: All load RGB images to be compared are uniformly scaled to a resolution of 224*224. A lightweight convolutional neural network, such as MobileNetV3, is used for feature extraction. The neural network can integrate global and local features such as color, texture, decoration, collar, and cuffs to output a feature vector. The similarity of the load RGB images is compared by calculating the cosine similarity of the two feature vectors.

[0082] Moisture content analysis: A near-infrared camera is used, with its response band selected to be sensitive to water molecule absorption, such as around 1450 nm or 970 nm. An active near-infrared light source is used for illumination, and a narrow-band filter corresponding to the corresponding wavelength is added in front of the lens to reduce ambient light interference and improve detection stability. At the start of the treatment, the load is in a dry state, and the first frame of RGB and near-infrared images are captured simultaneously. The load contour is segmented using the RGB image to obtain a load area mask; this mask is applied to the near-infrared image, and only the near-infrared pixels within the load area are extracted. The average gray value is calculated and denoted as the dry-state baseline gray value G0. After each steam spray (or when judgment is required), RGB and near-infrared images are captured simultaneously. The load area is segmented using the current RGB image to obtain the current load mask; this mask is mapped to the near-infrared image, and the near-infrared pixels of the current load area are extracted. The average gray value Gt is calculated; the moisture content is characterized using GΔ = G0 - Gt.

[0083] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A humidification care method, characterized in that... For use in nursing devices, the nursing devices including a receiving cavity, a humidifying device, and an image acquisition device, the method includes: S1: Control the image acquisition device to acquire image information of the load inside the cavity; S2: Determine, based on the image information, whether the currently exposed area of ​​the load within the containment cavity has been humidified by the humidification device; S3: If the current exposed area has not been humidified, control the humidification device to humidify the current exposed area; if the current exposed area has been humidified, change the distribution state or orientation of the load to expose another area of ​​the load. S4: Repeat S1 to S3 until the preset conditions are met and the program ends.

2. The humidification care method according to claim 1, characterized in that, The step of determining whether the exposed area of ​​the load has been humidified based on the image information includes: The current load area image is determined based on the image information; The current load area image is compared with the historical load area images. The similarity comparison is to compare the current load area image with the load area images of exposed areas that have been previously humidified and stored in the historical image library. If the similarity between the current load area image and all historical load area images is less than a preset similarity threshold A, then it is determined that the current exposed area has not been humidified. If the similarity between the current load area image and any historical load area image is greater than or equal to the preset similarity threshold A, then it is determined that the current exposed area has been humidified.

3. The humidification care method according to claim 1, characterized in that, The control of the humidification device to humidify the currently exposed area includes: The humidification device is controlled to perform initial humidification on the currently exposed area; After the initial humidification, a moisture content assessment was conducted. Based on the moisture content assessment results, it is determined whether the moisture content meets the preset requirements. If the preset requirements are not met, the humidification device is controlled to supplement humidification until the moisture content assessment results meet the preset requirements. If the moisture content meets the preset requirements, the process of controlling the humidification device to humidify the current exposed area ends.

4. The humidification care method according to claim 3, characterized in that, The moisture content assessment includes: Before the first execution of S1, the near-infrared camera is controlled to acquire a dry near-infrared image of the load, and the average gray level G0 of the dry load region when the load is in a dry state is obtained based on the dry near-infrared image. After each humidification, the near-infrared camera is controlled to acquire the current near-infrared image and the average gray level Gt of the current load area is acquired; Calculate the grayscale difference GΔ = G0 - Gt; If GΔ is greater than the moisture content analysis threshold G, then the moisture content meets the preset requirements; If GΔ is less than or equal to the moisture content analysis threshold G, then the moisture content has not met the preset requirements.

5. The humidification care method according to claim 4, characterized in that, In S4, the completion of the humidification program is determined by the number of humidification cycles, including: Set the effective humidification count N, and initialize the effective humidification count n=0; After each moisture content assessment confirms that the moisture content meets the preset requirements, n is incremented by 1. The humidification process ends when n is greater than or equal to N.

6. The humidification care method according to claim 1, characterized in that, If the current exposed area has been humidified, then changing the distribution state or orientation of the load includes: Set an upper limit B for the number of consecutive similarities and a current number b for the number of consecutive similarities, where the initial value of b is 0; When it is determined that the current exposed area has been humidified, increment b by 1; If b is less than B, then return to the step of changing the distribution state or orientation of the load; If b is greater than or equal to B, perform a moisture content assessment on the current load. If the moisture content does not meet the preset requirements, add humidification until the moisture content assessment result meets the preset requirements, and then set b to zero.

7. The humidification care method according to claim 2, characterized in that, The method further includes: when the current exposed area has not been humidified and humidification is performed, saving the current load area image to the historical image library to update the historical image library.

8. The humidification care method according to claim 4, characterized in that, The step of obtaining the average gray level G0 of the dry load region when the load is in a dry state includes: At the start of humidification, acquire RGB and near-infrared images of the load; The load contour is segmented using the RGB image to obtain a load region mask; The load area mask is applied to the near-infrared image, near-infrared pixels within the load area are extracted, and the average gray value is calculated as the average gray value G0 of the dry load area.

9. A humidification care method according to claim 1, characterized in that, The change in the distribution state or orientation of the load includes: The bearing of the receiving cavity is controlled to rotate, thereby causing the load to flip and exposing another area of ​​the load.

10. A garment processing device, characterized in that, It includes a receiving cavity, a humidification device, an image acquisition device, and a controller, the controller being configured to perform the humidification care method according to any one of claims 1-9.