Laundry treating apparatus and dehydration control method, device, and storage medium thereof

Abstract

This invention provides a garment processing device and its dehydration control method, equipment, and storage medium. The garment processing device is equipped with a camera module for acquiring images of garments inside the washing drum. The method includes the following steps: before dehydration begins, acquiring images of the garments inside the washing drum captured by the camera module; extracting features from the garment images to obtain garment features; generating optimal dehydration parameters based on the garment features; during dehydration according to the optimal dehydration parameters, tracking the movement trajectory of the garments through real-time images and fitting the centrifugal force distribution; identifying garment adhesion and determining whether to trigger a shaking program based on the centrifugal force distribution to shake the garments apart. This invention achieves precise and personalized dehydration parameters, effectively preventing and intervening in real-time garment adhesion, significantly improving rinsing efficiency and final washing effect, and realizing intelligent closed-loop control of perception-decision-execution.

Description

Technical Field

[0001] This invention relates to the field of clothing processing technology, and in particular to a clothing processing device, its dehydration control method, equipment, and storage medium. Background Technology

[0002] In existing washing machine spin-drying controls, especially for cotton clothing (such as T-shirts and bed sheets), there is a common phenomenon where clothes adhere tightly to the inner drum wall during the high-speed spin-drying stage. Specifically, in the initial stage of spin-drying, when the speed is below 800 rpm, the clothes oscillate with the water flow; however, when the speed exceeds a critical value (usually ≥1000 rpm), due to the significant increase in centrifugal force, the cotton fibers, after absorbing water, have an increased density (up to more than 3 times their dry weight), resulting in an increase in overall weight. As the centrifugal force in the inner drum is much greater than gravity, the clothes gradually form a uniform ring-shaped adhesion layer, tightly adhering to the inner drum wall.

[0003] The dynamic process of this phenomenon involves the combined effects of physical mechanisms and structural factors: cotton fibers increase in density after absorbing water, generating a strong centrifugal force during high-speed rotation, forcing the garments to adhere completely to the drum wall. At the same time, the design of the dehydration holes in the inner drum (including the size and distribution of the holes) has a significant impact on the efficiency of water drainage; if the hole size is too small, it will further hinder the rapid drainage of water and exacerbate the adhesion of the garments.

[0004] However, this close contact between the clothes and the inner drum leads to a series of problems. For example, after the spin-drying process, the clothes often fail to disperse automatically, reducing the efficiency of subsequent rinsing and affecting the overall washing effect, making it difficult for the clothes to achieve the desired loose state. While some existing technologies attempt to alleviate this problem by optimizing the spin-drying program or adjusting the inner drum structure, they still lack a fundamental solution to the problem of cotton clothes sticking together during high-speed spin-drying, especially in terms of balancing centrifugal force with the dynamic distribution of the clothes.

[0005] Therefore, there is an urgent need for a solution that can effectively overcome the problem of cotton clothing sticking during the dehydration process and improve dehydration efficiency and clothing treatment effect. Summary of the Invention

[0006] To achieve the above-mentioned objectives and other advantages of the present invention, a first objective of the present invention is to provide a dehydration control method for a garment processing device, wherein the garment processing device is equipped with a camera module for acquiring images of garments inside the washing drum of the garment processing device, comprising the following steps: The camera module captures images of the clothes inside the washing drum before the spin cycle begins. Feature extraction is performed on the clothing image to obtain clothing features; Optimal dehydration parameters are generated based on the characteristics of the clothing. During the dehydration process according to the optimal dehydration parameters, the movement trajectory of the clothing is tracked by real-time images of the clothing, and the centrifugal force distribution is fitted. The system identifies the adhesion of clothing and, based on the centrifugal force distribution, determines whether to trigger a shaking procedure to disperse the clothing.

[0007] Further, the step of extracting features from the clothing image to obtain clothing features includes: The thickness of the clothing layers is calculated using a contour detection algorithm to obtain the stacking height feature; Based on the HSV color space, clothing and background are separated, and the pixel ratio is statistically analyzed to obtain the unfolded area features; The texture consistency of the area adhering to the wall of the garment was analyzed using a local binary model to obtain the distribution uniformity characteristics.

[0008] Furthermore, the step of generating optimal dehydration parameters based on the clothing characteristics includes: The stacking height feature, the unfolded area feature, and the distribution uniformity feature are input into a pre-trained neural network model to output the optimal dehydration parameters.

[0009] Furthermore, the optimal dehydration parameters include the optimal dehydration speed, or any combination of the optimal dehydration speed with dehydration time, eccentricity compensation, and acceleration curve.

[0010] Furthermore, the step of tracking the movement trajectory of clothing through real-time clothing images and fitting the centrifugal force distribution includes: Histogram equalization was used to enhance the contrast of the clothing image, and median filtering was used to remove reflective noise from the dehydration holes. The movement trajectory of clothing was tracked using optical flow, and a centrifugal force distribution model was fitted.

[0011] Furthermore, the step of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model includes: At the start of dehydration, key feature points are detected in the clothing area of ​​the first frame image; In subsequent frames, optical flow is used to trace the motion trajectory of the feature points to obtain the displacement vector of the feature points. Calculate the expected circular motion of each feature point relative to the cylinder wall and compare it with the actual motion trajectory to identify abnormal trajectory points; Correlate the motion acceleration of each feature point with its estimated mass; Using the data from the aforementioned feature points, a continuous centrifugal force distribution cloud map is fitted using an interpolation algorithm.

[0012] Furthermore, the step of identifying clothing adhesion includes: Convert the clothing image to the HSV color space; The clothes are segmented from the converted clothes image by using a preset HSV threshold range of the washing drum itself, and a color-based binary mask is generated. A LBP map is calculated on the converted clothing image. The clothing is segmented from the LBP map by the LBP value distribution of the washing drum itself, and a texture-based binary mask is generated. Perform a logical AND operation between the color-based binary mask and the texture-based binary mask to obtain the unattached area; Perform mathematical morphological operations on the unattached regions, mark each connected background region, and return statistical information for each region; The total area of ​​the unattached region is calculated by summing the number of pixels in all connected background regions and taking the total number of pixels in the clothing image as the total area of ​​the cylinder wall. The percentage of the unattached region to the total area of ​​the cylinder wall is then calculated to obtain the unattached region percentage.

[0013] Furthermore, the step of identifying the clothing adhesion and determining whether to trigger a shaking procedure to shake the clothing, in conjunction with the centrifugal force distribution, includes: If the proportion of the unattached area is greater than a first threshold, or if the proportion of the unattached area is greater than a second threshold and the spatial overlap between the distribution of the abnormal trajectory points and the unattached area reaches a threshold, then the jitter procedure is triggered; wherein, the first threshold is greater than the second threshold.

[0014] Furthermore, the step of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model also includes: The proportion of the abnormal trajectory points to the total feature points, the centroid shift rate of the abnormal trajectory points, and the maximum stress value and its location extracted from the centrifugal force distribution cloud map are input into the stability risk prediction model to obtain the real-time instability risk probability.

[0015] Furthermore, the step of identifying the clothing adhesion and determining whether to trigger a shaking procedure to shake the clothing, in conjunction with the centrifugal force distribution, further includes: When the probability of instability reaches the threshold, the shakeout procedure is triggered.

[0016] Furthermore, the shaking process includes at least one of the following methods: automatically replenishing water into the washing drum, adjusting the rotation speed of the washing drum, and adjusting the rotation-stop ratio of the washing drum.

[0017] Furthermore, the automatic water replenishment to the washing drum includes the following steps: Brake the rotation speed of the garment processing device to a preset safe speed; Based on the location of the unattached area and the distribution area of ​​the abnormal trajectory points, control the water inlet valve to spray water at the corresponding angle to soften and impact the clothing; The washing drum is controlled to rotate at a low speed, alternating between forward and reverse directions. The amplitude, frequency, and duration of this rotation are jointly determined by the total pixel area of ​​the unattached area and the probability of instability.

[0018] Furthermore, during the dehydration process according to the optimal dehydration parameters, a stepped speed increase is adopted, with the optimal dehydration speed as the target endpoint of the stepped speed increase.

[0019] A second objective of this invention is to provide a garment processing device that utilizes the aforementioned method and includes a camera module for capturing images of garments inside the washing drum of the garment processing device.

[0020] Furthermore, the camera module is mounted on the door glass of the clothing processing device.

[0021] A third objective of the present invention is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0022] A fourth objective of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0023] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a garment processing device and its dehydration control method, equipment and storage medium. By introducing a camera module for image recognition, this invention can intelligently sense the key features of the garment that affect dynamic balance before physical dehydration, thereby generating the optimal dehydration parameters that match them, realizing a technological leap from traditional fixed dehydration programs to intelligent decision-making based on visual perception.

[0024] This invention dynamically and accurately monitors changes in the state of clothing by tracking the movement trajectory of clothing in real time and fitting the centrifugal force distribution during the dehydration process. Once image recognition determines that the clothing is tending to adhere tightly to the drum wall, a shaking-off program can be triggered in a timely manner by combining the centrifugal force distribution model. This program breaks the mechanical balance between the clothing and the drum wall by injecting a small amount of water, thereby quickly shaking it off in the early stages or after adhesion has formed, fundamentally solving the problem of clothing tangling and adhesion.

[0025] Because this invention effectively prevents clothes from clumping together during the dehydration stage, it ensures that the clothes can fully contact the water flow in the subsequent rinsing process, greatly improving the uniformity and efficiency of rinsing. At the same time, it avoids problems such as uneven dehydration and increased wrinkles caused by clothes sticking together, resulting in better looseness and smoothness of the clothes after the program is completed, thus improving the overall user experience.

[0026] This invention deeply integrates machine vision and motion control to form a complete intelligent control closed loop. By sensing the state of the clothing through a camera, determining the optimal parameters and identifying abnormal situations through algorithms, and finally executing dehydration or rehydration shaking actions through an actuator, this intelligent closed-loop control makes the clothing processing process more adaptive, efficient, and reliable.

[0027] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Specific embodiments of the present invention are given in detail in the following embodiments and their accompanying drawings. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 Flowchart of the dehydration control method for a garment processing device; Figure 2 A schematic diagram of the garment processing device; Figure 3 This is a cross-sectional view of the garment processing device; Figure 4 Flowchart for clothing feature extraction; Figure 5 To fit the flow chart of centrifugal force distribution; Figure 6 Flowchart for fitting a centrifugal force distribution model to track the movement trajectory of clothing using optical flow; Figure 7 Flowchart for identifying clothing adhesion; Figure 8 To identify the adhesion of clothing and determine whether to trigger the shaking procedure based on the centrifugal force distribution, the clothing shaking process is described. Figure 1 ; Figure 9 To identify the adhesion of clothing and determine whether to trigger the shaking procedure based on the centrifugal force distribution, the clothing shaking process is described. Figure 2 ; Figure 10 Flowchart for the automatic water replenishment process into the washing drum; Figure 11 This is a schematic diagram of a computer device. Figure 12 This is a schematic diagram of a computer-readable storage medium.

[0029] In the picture: 1. Camera module; 2. Door glass; 3. Washing drum; 4. Clothes. Detailed Implementation

[0030] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0031] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0032] The drawing numbers in this application are only used to distinguish the steps in the scheme and are not used to limit the execution order of the steps. The specific execution order is as described in the specification.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0034] Traditional washing machines rely on a single weight sensor for spin-drying control. However, cotton garments inevitably form a tightly adhered layer during high-speed spin-drying. While this is beneficial for dehydration, it also has side effects (such as difficulty in shaking out, tangling, and wrinkling). To upgrade spin-drying control from traditional blind decision-making (based on weight) to visual decision-making (based on image analysis), this invention deeply integrates image recognition with a physical model of the spin-drying process. This allows the washing machine's spin-drying process to not only remove water but also create optimal conditions for the next steps of processing the clothes (rinsing, drying, air-drying, etc.).

[0035] The clothing processing equipment can be configured as a washing machine, a washer-dryer combo, etc. For ease of description, this invention uses a drum washing machine as an example for illustration, and should not be construed as a limitation on the type of clothing processing equipment.

[0036] This method can be executed by the main control unit of the garment processing device. The main control unit can be implemented in the form of software and / or hardware, and is generally integrated into any electronic device with network communication capabilities, such as a mobile terminal, PC, or server.

[0037] Example 1 A method for controlling the dehydration of a garment processing device, such as Figures 2-3 As shown, a camera module 1 is installed inside the garment processing device. The camera module 1 is used to capture images of the garments inside the washing drum 3 of the garment processing device, such as... Figure 1 As shown, the method includes the following steps: S100: Obtain images of clothes inside the washing drum captured by the camera module before the spin-drying process begins; In some preferred embodiments, the camera module 1 is mounted on the door glass 2 of the clothing handling device, which can directly cover the entire axial field of view of the washing drum, realize global and unobstructed real-time monitoring of the distribution of clothes 4 in the drum, and effectively solve the problem of blind spots that may exist when installed on the side wall or top.

[0038] Since the clothes are mainly distributed circumferentially along the drum wall under the action of centrifugal force during the dehydration process, the camera at the door glass axis position is exactly perpendicular to this distribution plane, which can obtain the clearest radial projection image, avoid image distortion caused by tilted viewing angle, and significantly improve the accuracy of clothing feature extraction and motion trajectory tracking.

[0039] This embodiment integrates the camera into the door glass, rather than placing the sensor on the high-speed rotating inner cylinder or the strongly vibrating outer cylinder. This avoids stability interference and durability damage to the camera module caused by high-speed rotation and mechanical vibration, while simplifying wiring and improving the overall reliability of the system.

[0040] The camera is installed on the door glass of the garment processing device, so that the acquisition frequency of the camera is naturally synchronized with the rotation cycle of the inner drum. This makes it easy to directly analyze the circumferential motion state of the garment through image sequences, providing a visual data foundation with better time-series matching for real-time fitting of centrifugal force distribution.

[0041] This layout aligns with the user's habitual perspective of observing the inside of the cylinder through doors and windows. It not only facilitates technical implementation but also makes the visual monitoring of the dehydration process more intuitive for users, thus enhancing the user-friendliness of the product interaction.

[0042] Optionally, the camera module uses a 2-megapixel waterproof CMOS camera that supports a capture frame rate of ≥30fps and a wide dynamic range, ensuring that high-quality images can still be acquired in the internal high-reflectivity environment.

[0043] Optionally, an embedded edge computing chip with an NPU (such as Rockchip RK3588) can be used, which is optimized for visual algorithms to ensure that the latency of the entire image processing and decision-making process is less than 200ms, meeting the requirements of real-time control.

[0044] Since the garment processing usually involves more than just dehydration, after the preceding garment processing steps are completed, drainage is typically performed first, followed by the dehydration process. During this static window period after drainage and before dehydration begins, images of the garments inside the washing drum 3 can be captured by a camera module 1 installed in the door glass 2 (e.g., ...). Figure 3 The images of clothes captured by the camera module can avoid motion blur. Alternatively, a 120° wide-angle camera and an infrared fill light module can be used to capture images of clothes inside the washing drum. For example, a pulsed infrared LED can be used, with strict synchronization between the exposure and the camera module, shortening the exposure time to 0.1-0.5ms, completely freezing motion and ensuring clear images without motion blur.

[0045] This method is also applicable to scenarios where users only select the spin-drying function for clothing processing. During the static window period before spin-drying begins, images of the clothes inside the washing drum 3 can be captured by a camera module 1 installed in the door glass 2 (e.g., ...). Figure 3 (Image of clothing captured by the camera module).

[0046] S200: Extract features from the clothing image to obtain clothing features; In order to accurately describe the key factors affecting dynamic equilibrium from different dimensions, such as Figure 4 As shown, the step of extracting features from the clothing image to obtain clothing features includes: S210. Identify the surface contour of the clothing using a contour detection algorithm (such as OpenCV's findContours) and calculate its stacking thickness characteristics in the washing drum. S220. Perform image segmentation in the HSV color space to distinguish clothing from the background, and calculate the effective unfolded area features of clothing through pixel statistics. S230. Local binary model (LBP) is used to analyze the texture features of the area attached to the wall of the garment and to evaluate its distribution uniformity.

[0047] This embodiment uses three features—stack height, unfolded area, and distribution uniformity—to accurately describe the key factors affecting dynamic balance from different dimensions.

[0048] S300: Generate optimal dehydration parameters based on the characteristics of the clothing; To address the complex nonlinear relationship between clothing characteristics and optimal dehydration parameters, the step of generating optimal dehydration parameters based on the clothing characteristics includes: The stacking height feature, the unfolded area feature, and the distribution uniformity feature are input into a pre-trained neural network model (such as a ResNet-50 variant) to output the optimal dehydration parameters.

[0049] Furthermore, the optimal dehydration parameters include the optimal dehydration speed, or any combination of the optimal dehydration speed with dehydration time, eccentricity compensation, and acceleration curve.

[0050] The optimal spin speed can be understood as the model recommending the highest possible speed without triggering the risk of drum collision, so as to achieve the ideal dryness in a shorter time and save time and energy for subsequent air drying or tumble drying.

[0051] Optionally, the neural network model estimates the time required to dehydrate the garment to the target humidity based on the unfolded area and texture complexity of the garment (related to moisture content). Instead of using a fixed 8-minute or 10-minute dehydration time, the garment processing method dynamically adjusts. When the model predicts that the garment is essentially dehydrated, the dehydration process can be stopped early, achieving energy savings and time efficiency.

[0052] The neural network model calculates the initial imbalance and orientation by analyzing the centroid coordinates of the clothing distribution and the local unattached areas, and can then make targeted corrections.

[0053] The neural network model recommends acceleration curves or rotational speeds and durations for each stage to match different acceleration strategies to different clothing conditions. For example, for large, easily tangled garments (such as bed sheets): it outputs a "slow linear acceleration" curve to avoid instantaneous centrifugal force causing the garments to clump together and trigger violent vibrations. For small, uniform garments (such as T-shirts): it outputs a "rapid step-like acceleration" curve to reach the target rotational speed as quickly as possible, saving total time.

[0054] This embodiment uses the optimal dehydration parameter, which only includes the optimal dehydration speed, as an example and should not be construed as a specific limitation on the content of the optimal dehydration parameter. The extracted multidimensional features are input into a pre-trained neural network model, which directly outputs a recommended optimal dehydration speed (e.g., 1200±50 rpm), thereby achieving precise adaptive control of the clothing load and fundamentally preventing eccentric drum collisions.

[0055] To further reduce the impact of clothes instantly sticking to the surface, a stepped speed increase is adopted during the dehydration process according to the optimal dehydration parameters, with the optimal dehydration speed as the target endpoint of the stepped speed increase. For example, if the optimal dehydration speed is obtained after the above steps is 1400 rpm, the main control unit of the garment processing device can control the motor to increase the speed in steps of 400->800->1200->1400 rpm.

[0056] S400. During the dehydration process according to the optimal dehydration parameters, the movement trajectory of the clothing is tracked by real-time images of the clothing, and the centrifugal force distribution is fitted. To achieve the combination of static feature analysis and dynamic motion prediction, such as Figure 5 As shown, the step of tracking the movement trajectory of clothing through real-time clothing images and fitting the centrifugal force distribution includes: S410. Histogram equalization is applied to the clothing image to enhance image contrast and improve feature visibility. Median filtering is used to remove reflective noise from the washing drum's spin-drying holes. In this embodiment, the median filter effectively suppresses light spot noise caused by the washing drum's spin-drying holes.

[0057] S420: Track the movement trajectory of clothing using optical flow and fit a centrifugal force distribution model.

[0058] Optionally, during the low-speed rotation phase of the inner drum, the Lucas-Kanade optical flow method is used to track the microscopic motion trajectory of the clothing. Based on the motion data, a dynamic model is constructed to fit and predict the load stability and centrifugal force distribution under high centrifugal force.

[0059] This embodiment combines static feature analysis with dynamic motion prediction, which not only analyzes the current state of the clothes, but also predicts the dynamic stability during high-speed dehydration. Furthermore, the dehydration parameters are dynamically generated according to the actual situation of each batch of clothes, enabling personalized control. Corrections can be made before potential imbalances occur, greatly improving the reliability of the equipment and the user experience.

[0060] To extract valuable motion information from moving images, such as Figure 6 As shown, the steps of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model include: S421. At the start of dehydration, detect key feature points in the clothing area of ​​the first frame image, for example, using FAST corner points or Shi-Tomasi corner points for key feature point detection. S422. In subsequent frames, optical flow is used to trace the motion trajectory of the feature point to obtain the displacement vector (dx, dy) of the feature point. S423. Calculate the expected circular motion of each feature point relative to the cylinder wall, and compare it with the actual motion trajectory to identify abnormal trajectory points. Points where the expected circular motion closely matches the actual trajectory are considered stable attachment points; random points with irregular actual trajectories are considered splashing water droplets / noise points, which can be filtered out by trajectory length and consistency; and points where the trajectory is violent, non-circular, swaying, jumping, or lagging behind the cylinder wall are considered loose clothing points. These are the key points of concern.

[0061] S424. Associate the motion acceleration of each feature point (obtained by second differentiation of the displacement vector) with its estimated mass (which can be roughly estimated based on the pixel density of clothing in the area where the point is located); S425. Using the data from the aforementioned feature points, a continuous centrifugal force distribution cloud map is fitted using an interpolation algorithm (such as radial basis function interpolation). This cloud map visually displays which areas on the inner cylinder wall experience normal stress and which areas exhibit abnormal stress concentration.

[0062] S500: Identify the clothing adhesion and determine whether to trigger the shaking procedure based on the centrifugal force distribution to shake the clothing.

[0063] To stably and accurately quantify the distribution of clothing within the inner drum, such as... Figure 7 As shown, the step of identifying clothing adhesion includes: S510. Convert the clothing image to the HSV color space; Because the RGB color space is sensitive to changes in lighting, while the HSV color space separates the hue (H), saturation (S), and lightness (V) of a color, it is closer to the human eye's perception of color and can better separate clothing from the background. Therefore, this embodiment converts the clothing image from the RGB color space to the HSV color space.

[0064] S520: Segment the clothing from the converted clothing image using the preset HSV threshold range of the washing drum itself, and generate a color-based binary mask. Because the color and texture of the washing machine drum wall (usually stainless steel) differ significantly from that of clothing, it is worthwhile to pre-sample to obtain the H, S, and V value ranges of the "pure background" (clean drum wall) in HSV space. For example, a stainless steel drum wall may have low saturation (S) and high brightness (V) characteristics. An HSV threshold range for the background is set, and a binary mask is created using the cv2.inRange() function: pixels within this threshold range are labeled as background (white, value 255), and pixels outside are labeled as foreground / clothing (black, value 0).

[0065] S530. Calculate the LBP map on the converted clothing image, segment the clothing from the LBP map by the LBP value distribution of the washing drum itself, and generate a texture-based binary mask. To address the shortcomings of color segmentation, consider a scenario where a white T-shirt and a stainless steel cylinder are similar in color, but the T-shirt has a textile texture while the cylinder is smooth. Local Binary Pattern (LBP) can be used: The LBP map is calculated across the entire image. Then, through pre-sampling, the LBP value distribution in the "pure background" region (where the background texture is usually very uniform) is statistically analyzed, while the LBP values ​​in the clothing region are more complex and varied. By comparing these distributions, a texture-based binary mask can be generated.

[0066] S540. Perform a logical AND operation between the color-based binary mask and the texture-based binary mask. A pixel is only identified as the final "background" (i.e., the unattached area) if it is judged as the background by both masks simultaneously. This greatly improves accuracy and reduces false positives. In this embodiment, color and texture information are combined to obtain the final segmentation result, i.e., the unattached area. S550. Perform mathematical morphological operations on the unattached regions, mark each connected background region, and return the statistical information of each region; Since the original segmentation results often contain noise and small holes, optimization is needed. Specifically, opening operations (erosion followed by dilation) can effectively remove isolated, small foreground elements in the background region (such as noise misidentified as clothing). Closing operations (dilation followed by erosion) can fill small, isolated background holes in the clothing region (such as dehydration holes or reflective points misidentified as background elements). To identify independent unattached regions and analyze their position and size, the `cv2.connectedComponentsWithStats()` function can be used to mark each connected background region, returning statistical information for each region, including pixel area, bounding rectangle, and centroid coordinates.

[0067] S560. The sum of the number of pixels in all connected background regions is taken as the total area of ​​the unattached region, and the total number of pixels in the clothing image is taken as the total area of ​​the cylinder wall. The percentage of the total area of ​​the unattached region to the total area of ​​the cylinder wall is calculated to obtain the percentage of the unattached region.

[0068] This embodiment can stably and accurately quantify the distribution of clothes in the inner drum, which is the sensing foundation of the entire intelligent dehydration control system.

[0069] To achieve real-time visual monitoring and dynamic intervention during high-speed dehydration, such as Figure 8 As shown, the step of identifying the clothing adhesion and determining whether to trigger the shaking procedure based on the centrifugal force distribution to shake the clothing includes: S570. If the proportion of the unattached area is greater than a first threshold, or if the proportion of the unattached area is greater than a second threshold and the spatial overlap between the distribution of abnormal trajectory points and the unattached area reaches a threshold, then the jitter procedure is triggered; wherein, the first threshold is greater than the second threshold.

[0070] S571. If the conditions set in S570 are not met, there is no need to trigger the shaking procedure, and the subsequent process will proceed normally.

[0071] This embodiment compares the percentage of unattached areas with a first threshold to determine whether the current state of the garment processing device is severely unbalanced. Optionally, when the area of ​​a partially unattached area exceeds 15% of the total area of ​​the drum wall, an automatic water replenishment and re-shaking procedure is immediately triggered to prevent problems before they occur.

[0072] Optionally, when the area of ​​a partially unattached region exceeds 10% of the total area of ​​the cylinder wall, and the abnormal centrifugal force distribution overlaps with the spatial height of the unattached region*, the root cause and severity of the risk are confirmed, and an automatic water replenishment and re-dispersal procedure is immediately triggered to prevent problems before they occur.

[0073] If any of the above conditions are met, it indicates that the garment handling device is in or about to be in a state of serious imbalance, and immediate intervention is necessary.

[0074] Based on the above embodiments, in order to comprehensively reflect the possibility of severe vibration occurring in the next few seconds, such as Figure 9 As shown, the step of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model further includes: S426. Input the proportion of the abnormal trajectory points to the total feature points, the centroid offset change rate of the abnormal trajectory points, and the maximum stress value and its location extracted from the centrifugal force distribution cloud map into the stability risk prediction model to obtain the real-time instability risk probability Prisk (between 0 and 1). This probability comprehensively reflects the possibility of severe vibration occurring in the next few seconds.

[0075] Among them, the centroid offset change rate of abnormal trajectory points is the centroid calculated based on loose points, and whether its offset and direction are changing rapidly.

[0076] Accordingly, the step of identifying the clothing adhesion and determining whether to trigger the shaking procedure to shake the clothing, in conjunction with the centrifugal force distribution, further includes: S580. When the probability of instability reaches the threshold, the shakeout procedure is triggered.

[0077] For example, when the probability of instability risk Prisk > 0.7, a severe instability is predicted to occur, and an automatic water replenishment and re-shaking procedure is immediately triggered.

[0078] S581. If the conditions set in S580 are not met, there is no need to trigger the jitter procedure, and the subsequent process will proceed normally.

[0079] The above embodiments provide a multi-threshold, multi-strategy decision-making scheme, which transforms point motion into a systematic mechanical distribution and risk prediction, and executes corresponding decisions based on the predicted risks.

[0080] In some embodiments, the shaking process includes at least one of the following methods: automatically replenishing water into the washing drum, adjusting the rotation speed of the washing drum, and adjusting the rotation-to-stop ratio of the washing drum.

[0081] Among them, the shaking program based on adjusting the rotation speed of the washing drum mainly breaks up the adhesion of clothes and promotes their redistribution by dynamically changing the rotation speed of the washing drum.

[0082] Optionally, when clothing is detected adhering to the drum wall and the shaking program is triggered, the main control unit first controls the drive motor to perform a rapid acceleration-deceleration cycle. For example, within 2 seconds, the rotation speed is rapidly increased from the current spin-drying speed (e.g., 800 rpm) to a preset shaking speed (e.g., 1200 rpm) higher than the optimal spin-drying speed, maintained for 1-2 seconds, and then rapidly reduced back to a lower speed (e.g., 400 rpm) within 2 seconds. This drastic speed change generates a significant difference in inertial force, causing the clothing adhering to the drum wall to slide relative to the drum wall due to inertia, thereby separating from the drum wall.

[0083] After the initial impact, the main control unit controls the washing drum speed to fluctuate periodically or non-periodically within a set range (e.g., 600 rpm to 1000 rpm). A preferred embodiment uses a sine wave or triangular wave to control the speed, with the fluctuation frequency adjustable between 0.2 Hz and 1 Hz. This continuous variable speed motion applies varying centrifugal and tangential forces to the clothes, preventing them from sticking together again and causing the loosened clothes to tumble and disperse within the drum.

[0084] The swaying program, which adjusts the rotation-to-stop ratio of the washing drum, precisely controls the alternating rhythm of the drum's rotation and stillness, using the alternating action of gravity and inertia to sway the clothes.

[0085] Optionally, the main control unit selects or dynamically calculates a set of turnaround / stop ratio parameters from a pre-stored strategy library based on the identified severity of adhesion (e.g., the proportion of unattached areas) and their distribution. For example: A short-turn, short-stop pattern can be used, such as: turn forward for 1.5 seconds, stop abruptly and remain still for 1 second, turn backward for 1.5 seconds, stop abruptly and remain still for 1 second, forming a cycle.

[0086] A short-duration rotation followed by a long standstill mode can also be used, such as: rapidly accelerating to 500 rpm and rotating for 0.5 seconds, then abruptly stopping and remaining still for 3 seconds, forming one cycle. The longer standstill time allows the clothing to fall fully and detach from the drum wall under the influence of gravity.

[0087] The main control unit drives the motor cyclically according to the selected rotation-stop ratio. During the rotation phase, the clothes are drawn against the drum wall by centrifugal force; during the stationary phase after a sudden stop, the centrifugal force suddenly disappears, and the clothes are mainly subject to gravity. The clothes that were originally attached will continue to move slightly due to inertia and then slide off the drum wall or fall directly, thus achieving a shaking effect. Alternating forward and reverse rotation can disrupt the balance of attachment from different directions.

[0088] In order to achieve targeted troubleshooting, such as Figure 10 As shown, the automatic water replenishment to the washing drum includes the following steps: S590, brake the rotation speed of the garment processing device to a preset safe speed (e.g., 200 rpm). S591. Based on the location of the unattached area and the distribution area of ​​the abnormal trajectory points, control the water inlet valve to spray water at a specific angle onto the problem area to improve the shaking efficiency, soften and impact the clothing. S592. Control the washing drum to rotate at a low speed, alternating between forward and reverse directions. The amplitude, frequency, and duration of this rotation are jointly determined by the total pixel area of ​​the unattached area and the probability of instability.

[0089] After the shaking process ends, the system re-enters a low-speed rotation state and performs visual analysis again. If the new analysis results indicate that the risk has been eliminated, the dehydration parameters are re-planned (potentially reducing the target rotation speed) and dehydration continues. If the risk still exists, a second round of shaking is performed or a final downgrade is implemented (extremely low-speed dehydration and the user is notified).

[0090] This embodiment provides a garment processing device and its dehydration control method, achieving precise and personalized customization of dehydration parameters: by introducing a camera module for image recognition, it can intelligently sense key features affecting the dynamic balance of the garment before physical dehydration, thereby generating optimal dehydration parameters that match these features. This completely changes the traditional one-size-fits-all, crude mode of dehydration programs, achieving a leap from fixed programs to intelligent decision-making based on visual perception.

[0091] This embodiment effectively prevents and intervenes in real-time in the phenomenon of clothing adhesion: by tracking the movement trajectory of the clothing in real time during the dehydration process and fitting the centrifugal force distribution, the changes in the state of the clothing can be dynamically and accurately monitored. Once the clothing processing device determines through image recognition that the clothing is tending to adhere tightly to the drum wall, it can trigger a shaking-off program in a timely manner by combining the centrifugal force distribution model. This program breaks the mechanical balance between the clothing and the drum wall by injecting a small amount of water, thereby quickly shaking it off in the early stage or after adhesion formation, fundamentally solving the problem of clothing entanglement and adhesion.

[0092] This embodiment significantly improves rinsing efficiency and final washing results: by effectively preventing clothes from clumping together during the spin-drying stage, it ensures that the clothes can fully contact the water flow in the subsequent rinsing process, greatly improving the uniformity and efficiency of rinsing. At the same time, it avoids problems such as uneven dehydration and increased wrinkles caused by clothes sticking together, resulting in better looseness and smoothness of the clothes after the program ends, thus improving the overall user experience.

[0093] This invention achieves intelligent closed-loop control of perception-decision-execution: it deeply integrates machine vision and motion control to form a complete intelligent control closed loop. The system senses the state of the clothing through a camera, determines the optimal parameters and identifies abnormal situations through algorithms, and finally executes dehydration or rehydration shaking actions through an actuator. This intelligent closed-loop control makes the clothing handling process more adaptive, efficient, and reliable, representing an advanced direction in the development of smart home appliances.

[0094] Example 2 Based on the same concept, this embodiment also provides a clothing processing device that applies the dehydration control method provided in Embodiment 1. A detailed description of the dehydration control method provided in Embodiment 1 can be found in the corresponding descriptions in the above method embodiments, and will not be repeated here. In some embodiments, the clothing processing device can be applied as a washing machine, and in other embodiments as a washer-dryer combo.

[0095] It is understood that the garment processing device provided in this embodiment includes hardware structures and / or software modules corresponding to each function in order to achieve the above-mentioned functions. In conjunction with the units and algorithm steps of the various examples disclosed in this embodiment, this embodiment can be implemented in hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the technical solution of this embodiment.

[0096] A garment processing device, such as Figures 2-3 As shown, it includes a camera module 1, which is used to capture images of clothes inside the washing drum 3 of the clothing processing device.

[0097] In some preferred embodiments, the camera module 1 is mounted on the door glass 2 of the clothing handling device, which can directly cover the entire axial field of view of the washing drum, realize global and unobstructed real-time monitoring of the distribution of clothes 4 in the drum, and effectively solve the problem of blind spots that may exist when installed on the side wall or top.

[0098] Since the clothes are mainly distributed circumferentially along the drum wall under the action of centrifugal force during the dehydration process, the camera at the door glass axis position is exactly perpendicular to this distribution plane, which can obtain the clearest radial projection image, avoid image distortion caused by tilted viewing angle, and significantly improve the accuracy of clothing feature extraction and motion trajectory tracking.

[0099] This embodiment integrates the camera into the door glass, rather than placing the sensor on the high-speed rotating inner cylinder or the strongly vibrating outer cylinder. This avoids stability interference and durability damage to the camera module caused by high-speed rotation and mechanical vibration, while simplifying wiring and improving the overall reliability of the system.

[0100] The camera is installed on the door glass of the garment processing device, so that the acquisition frequency of the camera is naturally synchronized with the rotation cycle of the inner drum. This makes it easy to directly analyze the circumferential motion state of the garment through image sequences, providing a visual data foundation with better time-series matching for real-time fitting of centrifugal force distribution.

[0101] This layout aligns with the user's habitual perspective of observing the inside of the cylinder through doors and windows. It not only facilitates technical implementation but also makes the visual monitoring of the dehydration process more intuitive for users, thus enhancing the user-friendliness of the product interaction.

[0102] Optionally, the camera module uses a 2-megapixel waterproof CMOS camera that supports a capture frame rate of ≥30fps and a wide dynamic range, ensuring that high-quality images can still be acquired in the internal high-reflectivity environment.

[0103] Optionally, an embedded edge computing chip with an NPU (such as Rockchip RK3588) can be used, which is optimized for visual algorithms to ensure that the latency of the entire image processing and decision-making process is less than 200ms, meeting the requirements of real-time control.

[0104] In some embodiments, by optimizing the inner cylinder structure (micro-protrusions, distribution of dehydration holes), the physical adhesion of clothing and the water flow state are improved, creating more favorable and predictable initial conditions for the image recognition algorithm, which is equivalent to reducing the decision difficulty of the algorithm.

[0105] The clothing processing device of this embodiment may include the following process: The washing drum image acquisition module is used to acquire images of clothes inside the washing drum captured by the camera module before the spin-drying begins. The clothing feature extraction module is used to extract features from the clothing image to obtain clothing features; An optimal dehydration parameter generation module is used to generate optimal dehydration parameters based on the characteristics of the clothing. The centrifugal force distribution fitting module is used to track the movement trajectory of clothing through real-time images of clothing and fit the centrifugal force distribution during the dehydration process according to the optimal dehydration parameters. The shaking program trigger module is used to identify the adhesion of clothing and determine whether to trigger the shaking program in order to shake the clothing based on the centrifugal force distribution.

[0106] Based on the technical solution of the above embodiments, optionally, the step of extracting features from the clothing image to obtain clothing features includes: The thickness of the clothing layers is calculated using a contour detection algorithm to obtain the stacking height feature; Based on the HSV color space, clothing and background are separated, and the pixel ratio is statistically analyzed to obtain the unfolded area features; The texture consistency of the area adhering to the wall of the garment was analyzed using a local binary model to obtain the distribution uniformity characteristics.

[0107] Based on the technical solution of the above embodiments, optionally, the step of generating optimal dehydration parameters based on the clothing characteristics includes: The stacking height feature, the unfolded area feature, and the distribution uniformity feature are input into a pre-trained neural network model to output the optimal dehydration parameters.

[0108] Based on the technical solutions of the above embodiments, optionally, the optimal dehydration parameters include the optimal dehydration speed, or any combination of the optimal dehydration speed with dehydration time, eccentricity compensation amount, and acceleration curve.

[0109] Based on the technical solution of the above embodiments, optionally, the step of tracking the movement trajectory of clothing through real-time clothing images and fitting the centrifugal force distribution includes: Histogram equalization was used to enhance the contrast of the clothing image, and median filtering was used to remove reflective noise from the dehydration holes. The movement trajectory of clothing was tracked using optical flow, and a centrifugal force distribution model was fitted.

[0110] Based on the technical solution of the above embodiments, optionally, the step of tracking the movement trajectory of clothing by optical flow and fitting a centrifugal force distribution model includes: At the start of dehydration, key feature points are detected in the clothing area of ​​the first frame image; In subsequent frames, optical flow is used to trace the motion trajectory of the feature points to obtain the displacement vector of the feature points. Calculate the expected circular motion of each feature point relative to the cylinder wall and compare it with the actual motion trajectory to identify abnormal trajectory points; Correlate the motion acceleration of each feature point with its estimated mass; Using the data from the aforementioned feature points, a continuous centrifugal force distribution cloud map is fitted using an interpolation algorithm.

[0111] Based on the technical solutions of the above embodiments, optionally, the step of identifying clothing adhesion includes: Convert the clothing image to the HSV color space; The clothes are segmented from the converted clothes image by using a preset HSV threshold range of the washing drum itself, and a color-based binary mask is generated. A LBP map is calculated on the converted clothing image. The clothing is segmented from the LBP map by the LBP value distribution of the washing drum itself, and a texture-based binary mask is generated. Perform a logical AND operation between the color-based binary mask and the texture-based binary mask to obtain the unattached area; Perform mathematical morphological operations on the unattached regions, mark each connected background region, and return statistical information for each region; The total area of ​​the unattached region is calculated by summing the number of pixels in all connected background regions and taking the total number of pixels in the clothing image as the total area of ​​the cylinder wall. The percentage of the unattached region to the total area of ​​the cylinder wall is then calculated to obtain the unattached region percentage.

[0112] Based on the technical solution of the above embodiments, optionally, the step of identifying the clothing adhesion and determining whether to trigger the shaking procedure in combination with the centrifugal force distribution to shake the clothing includes: If the proportion of the unattached area is greater than a first threshold, or if the proportion of the unattached area is greater than a second threshold and the spatial overlap between the distribution of the abnormal trajectory points and the unattached area reaches a threshold, then the jitter procedure is triggered; wherein, the first threshold is greater than the second threshold.

[0113] Based on the technical solution of the above embodiments, optionally, the step of tracking the movement trajectory of clothing by optical flow and fitting a centrifugal force distribution model further includes: The proportion of the abnormal trajectory points to the total feature points, the centroid shift rate of the abnormal trajectory points, and the maximum stress value and its location extracted from the centrifugal force distribution cloud map are input into the stability risk prediction model to obtain the real-time instability risk probability.

[0114] Based on the technical solution of the above embodiments, optionally, the step of identifying the clothing adhesion and determining whether to trigger the shaking procedure in combination with the centrifugal force distribution to shake the clothing further includes: When the probability of instability reaches the threshold, the shakeout procedure is triggered.

[0115] Based on the technical solutions of the above embodiments, the shaking process may optionally include at least one of the following methods: automatically replenishing water into the washing drum, adjusting the rotation speed of the washing drum, and adjusting the rotation-stop ratio of the washing drum.

[0116] Based on the technical solution of the above embodiments, optionally, the automatic water replenishment to the washing drum includes the following steps: Brake the rotation speed of the garment processing device to a preset safe speed; Based on the location of the unattached area and the distribution area of ​​the abnormal trajectory points, control the water inlet valve to spray water at the corresponding angle to soften and impact the clothing; The washing drum is controlled to rotate at a low speed, alternating between forward and reverse directions. The amplitude, frequency, and duration of this rotation are jointly determined by the total pixel area of ​​the unattached area and the probability of instability.

[0117] Based on the technical solution of the above embodiments, optionally, during the dehydration process according to the optimal dehydration parameters, a stepped speed increase is adopted, and the optimal dehydration speed is used as the target endpoint of the stepped speed increase.

[0118] This embodiment provides a garment processing device that achieves precise and personalized customization of dehydration parameters: by introducing a camera module for image recognition, it can intelligently sense key features affecting the dynamic balance of the garment before physical dehydration, thereby generating optimal dehydration parameters that match these features. This completely changes the traditional one-size-fits-all, crude mode of dehydration programs, achieving a leap from fixed programs to intelligent decision-making based on visual perception.

[0119] This embodiment effectively prevents and intervenes in real-time in the phenomenon of clothing adhesion: by tracking the movement trajectory of the clothing in real time during the dehydration process and fitting the centrifugal force distribution, the changes in the state of the clothing can be dynamically and accurately monitored. Once the clothing processing device determines through image recognition that the clothing is tending to adhere tightly to the drum wall, it can trigger a shaking-off program in a timely manner by combining the centrifugal force distribution model. This program breaks the mechanical balance between the clothing and the drum wall by injecting a small amount of water, thereby quickly shaking it off in the early stage or after adhesion formation, fundamentally solving the problem of clothing entanglement and adhesion.

[0120] This embodiment significantly improves rinsing efficiency and final washing results: by effectively preventing clothes from clumping together during the spin-drying stage, it ensures that the clothes can fully contact the water flow in the subsequent rinsing process, greatly improving the uniformity and efficiency of rinsing. At the same time, it avoids problems such as uneven dehydration and increased wrinkles caused by clothes sticking together, resulting in better looseness and smoothness of the clothes after the program ends, thus improving the overall user experience.

[0121] This invention achieves intelligent closed-loop control of perception-decision-execution: it deeply integrates machine vision and motion control to form a complete intelligent control closed loop. The system senses the state of the clothing through a camera, determines the optimal parameters and identifies abnormal situations through algorithms, and finally executes dehydration or rehydration shaking actions through an actuator. This intelligent closed-loop control makes the clothing handling process more adaptive, efficient, and reliable, representing an advanced direction in the development of smart home appliances.

[0122] Example 3 A computer device 600, such as Figure 11 As shown, the device includes a memory 610, a processor 620, and a computer program 630 stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a dehydration control method for a garment processing apparatus. For a detailed description of the method, please refer to the corresponding description in the above method embodiments; it will not be repeated here.

[0123] Example 4 A computer-readable storage medium, such as Figure 12 As shown, a computer program is stored thereon, which, when executed by a processor, implements the steps of a dehydration control method for a garment processing device. For a detailed description of the method, please refer to the corresponding description in the above method embodiments, and will not be repeated here.

[0124] The number of devices and processing scale described herein are for the purpose of simplifying the description of the invention. Applications, modifications, and variations of the invention will be readily apparent to those skilled in the art.

[0125] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

[0126] The apparatus, computer device, and non-volatile computer storage medium and method provided in the embodiments of this specification are corresponding. Therefore, the apparatus, computer device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device, and non-volatile computer storage medium will not be repeated here.

[0127] Those skilled in the art will also know that, besides implementing the controller in the form of purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller take the form of logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices included within it for implementing various functions can also be considered structures within that hardware component. Alternatively, the devices for implementing various functions can be considered as both software units implementing the method and structures within a hardware component.

[0128] The systems, apparatuses, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above apparatuses are described separately as various units based on their functions. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0129] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0131] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0132] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0133] It should also be noted that 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 limitation, 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 said element.

[0134] This specification may be described in the general context of computer-executable instructions, such as program units, that are executed by a computer. Generally, program units include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification may also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program units may reside in local and remote computer storage media, including storage devices.

[0135] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0136] The above description is merely an embodiment of this specification and is not intended to limit the scope of one or more embodiments of this specification. Various modifications and variations can be made to one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of one or more embodiments of this specification.

Claims

1. A method for controlling the dehydration of a garment processing device, characterized in that, The garment processing device is equipped with a camera module, which is used to capture images of the garments inside the washing drum of the garment processing device, including the following steps: The camera module captures images of the clothes inside the washing drum before the spin cycle begins. Feature extraction is performed on the clothing image to obtain clothing features; Optimal dehydration parameters are generated based on the characteristics of the clothing. During the dehydration process according to the optimal dehydration parameters, the movement trajectory of the clothing is tracked by real-time images of the clothing, and the centrifugal force distribution is fitted. The system identifies the adhesion of clothing and, based on the centrifugal force distribution, determines whether to trigger a shaking procedure to disperse the clothing.

2. The dehydration control method of a garment processing device as described in claim 1, characterized in that, The step of extracting features from the clothing image to obtain clothing features includes: The thickness of the clothing layers is calculated using a contour detection algorithm to obtain the stacking height feature; Based on the HSV color space, clothing and background are separated, and the pixel ratio is statistically analyzed to obtain the unfolded area features; The texture consistency of the area adhering to the wall of the garment was analyzed using a local binary model to obtain the distribution uniformity characteristics.

3. The dehydration control method of a garment processing device as described in claim 2, characterized in that, The step of generating optimal dehydration parameters based on the clothing characteristics includes: The stacking height feature, the unfolded area feature, and the distribution uniformity feature are input into a pre-trained neural network model to output the optimal dehydration parameters.

4. The dehydration control method of the garment processing device as described in claim 1, characterized in that, The optimal dehydration parameters include the optimal dehydration speed, or any combination of the optimal dehydration speed with dehydration time, eccentricity compensation, and acceleration curve.

5. The dehydration control method of a garment processing device as described in claim 1, characterized in that, The step of tracking the movement trajectory of clothing through real-time clothing images and fitting the centrifugal force distribution includes: Histogram equalization was used to enhance the contrast of the clothing image, and median filtering was used to remove reflective noise from the dehydration holes. The movement trajectory of clothing was tracked using optical flow, and a centrifugal force distribution model was fitted.

6. The dehydration control method of a garment processing device as described in claim 5, characterized in that, The steps of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model include: At the start of dehydration, key feature points are detected in the clothing area of ​​the first frame image; In subsequent frames, optical flow is used to trace the motion trajectory of the feature points to obtain the displacement vector of the feature points. Calculate the expected circular motion of each feature point relative to the cylinder wall and compare it with the actual motion trajectory to identify abnormal trajectory points; Correlate the motion acceleration of each feature point with its estimated mass; Using the data from the aforementioned feature points, a continuous centrifugal force distribution cloud map is fitted using an interpolation algorithm.

7. The dehydration control method of a garment processing device as described in claim 6, characterized in that, The steps for identifying clothing adhesion include: Convert the clothing image to the HSV color space; The clothes are segmented from the converted clothes image by using a preset HSV threshold range of the washing drum itself, and a color-based binary mask is generated. A LBP map is calculated on the converted clothing image. The clothing is segmented from the LBP map by the LBP value distribution of the washing drum itself, and a texture-based binary mask is generated. Perform a logical AND operation between the color-based binary mask and the texture-based binary mask to obtain the unattached area; Perform mathematical morphological operations on the unattached regions, mark each connected background region, and return statistical information for each region; The total area of ​​the unattached region is calculated by summing the number of pixels in all connected background regions and taking the total number of pixels in the clothing image as the total area of ​​the cylinder wall. The percentage of the unattached region to the total area of ​​the cylinder wall is then calculated to obtain the unattached region percentage.

8. The dehydration control method of a garment processing device as described in claim 7, characterized in that, The step of identifying the clothing adhesion and determining whether to trigger the shaking procedure based on the centrifugal force distribution to shake the clothing includes: If the proportion of the unattached area is greater than a first threshold, or if the proportion of the unattached area is greater than a second threshold and the spatial overlap between the distribution of the abnormal trajectory points and the unattached area reaches a threshold, then the jitter procedure is triggered; wherein, the first threshold is greater than the second threshold.

9. The dehydration control method of a garment processing device as described in claim 8, characterized in that, The step of tracking the movement trajectory of clothing using optical flow and fitting a centrifugal force distribution model further includes: The proportion of the abnormal trajectory points to the total feature points, the centroid shift rate of the abnormal trajectory points, and the maximum stress value and its location extracted from the centrifugal force distribution cloud map are input into the stability risk prediction model to obtain the real-time instability risk probability.

10. The dehydration control method of a garment processing device as described in claim 9, characterized in that, The step of identifying the clothing adhesion and determining whether to trigger the shaking procedure based on the centrifugal force distribution to shake the clothing further includes: When the probability of instability reaches the threshold, the shakeout procedure is triggered.

11. The dehydration control method of a garment processing device as described in claim 9, characterized in that: The shaking process includes at least one of the following methods: automatically replenishing water into the washing drum, adjusting the rotation speed of the washing drum, and adjusting the rotation-stop ratio of the washing drum.

12. The dehydration control method of a garment processing device as described in claim 11, characterized in that, The automatic water replenishment to the washing drum includes the following steps: Brake the rotation speed of the garment processing device to a preset safe speed; Based on the location of the unattached area and the distribution area of ​​the abnormal trajectory points, control the water inlet valve to spray water at the corresponding angle to soften and impact the clothing; The washing drum is controlled to rotate at a low speed, alternating between forward and reverse directions. The amplitude, frequency, and duration of this rotation are jointly determined by the total pixel area of ​​the unattached area and the probability of instability.

13. The dehydration control method of a garment processing device as described in claim 4, characterized in that: During the dehydration process according to the optimal dehydration parameters, a stepped speed increase is adopted, with the optimal dehydration speed as the target endpoint of the stepped speed increase.

14. A garment processing apparatus, employing the method as described in any one of claims 1 to 13, characterized in that: It includes a camera module, which is used to capture images of clothes inside the washing drum of the clothing processing device.

15. The garment processing apparatus as described in claim 14, characterized in that: The camera module is mounted on the door glass of the clothing processing device.

16. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 13.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 13.

Images