Methods, devices, and systems for identifying the color of clothing in laundry and care equipment.

Abstract

This application provides a method, apparatus, and system for identifying the color of clothing in a laundry appliance. The method includes: acquiring a first image of the clothing in the laundry appliance; performing a single-step dyeing analysis based on the first image without selecting a washing program to obtain a first identification result, wherein the first identification result is used to characterize whether there is a risk of dyeing in the clothing identified in a single step; acquiring a second image of the clothing in the laundry appliance if the first identification result indicates no risk of dyeing; performing multiple dyeing analyses based on the second image if a washing program has been selected to obtain a second identification result, wherein the second identification result is used to characterize whether there is a risk of dyeing in the clothing identified in multiple steps; determining that the clothing is at risk of dyeing if either the first identification result indicates a risk of dyeing or the second identification result indicates a risk of dyeing, and generating an alarm message. This solution addresses the problem of dyeing easily occurring in washing machines in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of washing machines, in particular to a method for identifying the color of clothes of a washing and caring device, an apparatus for identifying the color of clothes of a washing and caring device, a computer program product and a washing and caring device system. BACKGROUND

[0002] At present, the washing machine market is in the initial stage in terms of color recognition of clothes, and some high-end products can only distinguish basic colors, and are difficult to deal with complex patterns and special process clothes. In household washing, consumers often cause dyeing problems due to improper color classification of clothes; manual sorting in commercial laundry is labor-intensive and prone to errors, and accidents may face complaints and economic losses. For silk, wool and other special material clothes, color gradient and delicate changes increase the difficulty of manual classification, and traditional methods have been difficult to meet the demand. SUMMARY

[0003] The main purpose of the present application is to provide a method for identifying the color of clothes of a washing and caring device, an apparatus for identifying the color of clothes of a washing and caring device, a computer program product and a washing and caring device system, to at least solve the problem that the washing machine is prone to dyeing in the prior art.

[0004] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for identifying the color of clothes of a washing and caring device is provided, comprising: acquiring a first image of clothes of a washing and caring device; in the case where no washing program is selected, performing single dyeing analysis according to the first image to obtain a first identification result, wherein the first identification result is used to represent whether the single-identified clothes have a dyeing risk; in the case where the first identification result represents no dyeing risk, acquiring a second image of the clothes of the washing and caring device; in the case where the washing program is selected, performing multiple dyeing analysis according to the second image to obtain a second identification result, wherein the second identification result is used to represent whether the multiple-identified clothes have a dyeing risk; in the case where the first identification result represents a dyeing risk or the second identification result represents a dyeing risk, it is determined that the clothes have a dyeing risk, and an alarm information is generated.

[0005] Optionally, acquiring the first image of the clothes of the washing and caring device comprises: determining whether the door body of the washing and caring device is closed; in the case where the door body of the washing and caring device is closed, acquiring the image of the clothes of the washing and caring device to obtain the first image.

[0006] Optionally, a single coloring analysis is performed on the first image to obtain a first identification result, including: extracting colors from the first image and converting the colors to a color space; using a clustering algorithm in the color space to cluster the colors to obtain at least one color cluster; determining that the first identification result has a coloring risk if the number of color clusters is greater than or equal to a preset number, the area of ​​each color cluster is greater than or equal to a preset area, and the brightness difference between any two color clusters is greater than or equal to a preset brightness difference; and determining that the first identification result has no coloring risk if one or both of the following conditions are met: the number of color clusters is greater than or equal to the preset number, the area of ​​each color cluster is greater than or equal to the preset area, and the brightness difference between any two color clusters is greater than or equal to the preset brightness difference.

[0007] Optionally, if the first identification result indicates no risk of staining, acquiring a second image of the clothing in the washing and care device includes: determining whether the washing program has been selected if the first identification result indicates no risk of staining; if the washing program has been selected, controlling the drum of the washing and care device to rotate, and acquiring an image of the clothing in the washing and care device after each drum rotation to obtain multiple second images.

[0008] Optionally, performing multiple staining analyses on the second image to obtain a second identification result includes: performing staining analysis on each second image obtained after each drum flip to obtain multiple preliminary identification results, wherein each preliminary identification result corresponds one-to-one with the second image obtained after each drum flip; determining the second identification result as having staining risk if the number of preliminary identification results indicating staining risk is greater than or equal to a preset number threshold; and determining the second identification result as having no staining risk if the number of preliminary identification results indicating staining risk is less than the preset number threshold.

[0009] Optionally, before acquiring a first image of the clothes in the washing and care device and before acquiring a second image of the clothes in the washing and care device, the method further includes: acquiring preset shooting parameters, wherein the preset shooting parameters include one or more of metering mode, exposure parameters, glare suppression, ISO sensitivity and white balance; and locking the shooting parameters of the image acquisition device to the preset shooting parameters.

[0010] Optionally, after performing multiple dyeing analyses based on the second image to obtain a second identification result, the method further includes: determining that the clothing has no dyeing risk when both the first identification result and the second identification result indicate no dyeing risk, and controlling the washing and care equipment to operate according to the washing program.

[0011] According to another aspect of this application, a garment color recognition device for a laundry appliance is provided, comprising: a first acquisition unit for acquiring a first image of the garment in the laundry appliance; a first analysis unit for performing a single dyeing analysis based on the first image without selecting a washing program to obtain a first recognition result, wherein the first recognition result is used to characterize whether the garment has a dyeing risk in a single identification; a second acquisition unit for acquiring a second image of the garment in the laundry appliance if the first recognition result indicates no dyeing risk; a second analysis unit for performing multiple dyeing analyses based on the second image if the washing program has been selected to obtain a second recognition result, wherein the second recognition result is used to characterize whether the garment has a dyeing risk in multiple identifications; and a first processing unit for determining that the garment has a dyeing risk and generating an alarm message if the first recognition result indicates a dyeing risk or the second recognition result indicates a dyeing risk.

[0012] According to another aspect of this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of a method for recognizing the color of clothing in any of the washing and care devices.

[0013] According to another aspect of this application, a laundry care device system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including a method for performing a clothing color recognition method for any of the laundry care devices described above.

[0014] By applying the technical solution of this application, the washing machine first performs a single identification of the clothes. If a risk of staining is detected, the washing machine immediately pauses the washing program to prevent staining accidents. At the same time, an alarm message is sent to the user, allowing the user to understand the situation in a timely manner and make the right decision, thus preventing the need for secondary operations. If no staining risk is detected in the single identification, further analysis is performed. By identifying the clothes in the washing machine multiple times, omissions in the single identification are avoided, ensuring the comprehensiveness and reliability of staining risk identification. Then, if a staining risk is detected, the washing machine immediately pauses the washing program to prevent staining accidents. At the same time, an alarm message is sent to the user, allowing the user to understand the situation in a timely manner and make the right decision, thereby avoiding the problem of staining easily occurring in the washing machine. Attached Figure Description

[0015] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0016] Figure 1 A hardware structure block diagram of a mobile terminal for performing a method for identifying the color of clothing using a laundry device, according to an embodiment of this application, is shown.

[0017] Figure 2 A flowchart illustrating a method for identifying the color of clothing using a laundry device according to an embodiment of this application is shown.

[0018] Figure 3 A flowchart illustrating the multicolor detection process is shown.

[0019] Figure 4 A flowchart illustrating the intelligent clustering algorithm is shown.

[0020] Figure 5 A schematic diagram of the risk circuit breaker mechanism is shown;

[0021] Figure 6 A structural block diagram of a clothing color recognition device for a laundry appliance provided according to an embodiment of this application is shown.

[0022] The above figures include the following reference numerals:

[0023] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] As described in the background section, washing machines in the prior art are prone to staining. To solve the above problem, embodiments of this application provide a method for identifying the color of clothes in a washing and care device, a device for identifying the color of clothes in a washing and care device, a computer program product, and a washing and care device system.

[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0029] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of recognizing the color of clothing in a laundry and care device according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1The different configurations shown.

[0030] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the clothing color recognition method of the washing and care device in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-described networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the above-described networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0031] This embodiment provides a method for identifying the color of clothing in a laundry device that runs on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] Figure 2 This is a schematic flowchart illustrating a method for identifying the color of clothing using a laundry care device according to an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0033] Step S201: Obtain the first image of the clothes in the washing and care equipment;

[0034] Specifically, the process of acquiring the first image begins the instant the user loads the clothes and closes the washing machine door; this moment is called door-closing pre-inspection. A highly sensitive built-in camera captures images of the clothes inside the washing machine with a millisecond-level response time, ensuring that the system quickly scans and identifies the initial color state of the clothes before the washing program starts. This immediate response mechanism ensures that potential staining risks are detected before washing begins, allowing users to receive warning information before the wash starts and avoid unnecessary damage. This is because by starting detection the instant the washing machine door closes, the system can capture the color information of the clothes immediately and perform timely staining risk analysis.

[0035] Step S202: Without selecting a washing program, perform a single dyeing analysis based on the first image to obtain a first identification result, wherein the first identification result is used to characterize whether there is a risk of dyeing in the garment in a single identification.

[0036] Specifically, based on the first image, an advanced self-supervised learning adaptive clustering algorithm is used to perform clustering operations within the CIE Lab color space to analyze the color distribution of the clothing. If the algorithm determines the existence of at least two significant color clusters, and the area ratio of any one cluster exceeds a set threshold (e.g., 5%), then a staining risk is identified, and the first identification result is considered to be a staining risk. Conversely, if no color clusters meeting the above conditions are identified, then no staining risk is identified. Accurate initial screening and efficient filtering: Through accurate initial color screening, most clothing without staining risk can be effectively filtered out, avoiding resource waste. This is because the initial identification uses strict standards; a staining risk alarm is only triggered when the color distribution of the clothing is abnormal. This reduces invalid computations when the system processes clothing without staining risk, improving overall detection efficiency.

[0037] Step S203: If the first identification result indicates no risk of staining, obtain a second image of the clothing in the washing and care equipment.

[0038] Specifically, if the initial detection shows no risk of staining, the system will not immediately interrupt the process but will allow the user to select a washing program. After the user selects a program, the drum will perform one or more tumbling actions to change the position and stacking state of the clothes inside the drum. Subsequently, the system will capture a new image, the second image, to capture the new state of the clothes after the drum tumbling, ensuring that no potential sources of staining risk are missed. Dynamic tracking eliminates the influence of occlusion. Through the acquisition of the second image triggered by the drum tumbling, the system can dynamically track the changes in the clothes, eliminating recognition blind spots caused by clothes occlusion or stacking, ensuring that all clothing colors are fully checked. This is because even parts that were not detected in the initial detection, located under the clothes or obscured by other clothes, will be exposed after the drum tumbling, and the acquisition of the second image ensures the detection of these parts.

[0039] Step S204: With the above washing program selected, multiple dyeing analyses are performed based on the second image to obtain a second identification result, wherein the second identification result is used to characterize whether the repeatedly identified clothing has a risk of dyeing.

[0040] Specifically, after acquiring the second image, the dyeing analysis is repeated, a process known as multiple identification. The purpose is not only to verify the initial identification result but also to capture the full picture of color changes in the clothing after the drum tumbles. If the algorithm still determines there is no dyeing risk at this stage, the clothing combination is considered safe and can be washed normally; conversely, if a dyeing risk is detected, an alarm will be triggered. This double-layered approach improves detection accuracy. The combination of initial and multiple identification forms a double-layered protection mechanism. Even if the initial detection fails to accurately identify dyeing risk due to environmental factors (such as lighting or clothing stacking), multiple identification can compensate for this deficiency, significantly improving the accuracy and reliability of dyeing detection. This is because, by acquiring the second image after the drum tumbles, the system can perform color identification based on the new position and state of the clothing. Even if the initial detection is affected by occlusion, multiple identification has the opportunity to correct this error.

[0041] Step S205: If the first identification result indicates a risk of staining or the second identification result indicates a risk of staining, determine that the clothing has a risk of staining and generate an alarm message.

[0042] Specifically, if a staining risk is detected at any stage of the initial or multiple identification processes, the system immediately executes a risk circuit breaker mechanism, automatically pausing the washing program that is about to start or is currently running, and sending an alarm message to the user indicating the presence of a staining risk. This immediate intervention mitigates the risk. The instant risk feedback and intervention function ensures that the system can react quickly upon detecting any staining risk, stopping the washing program and preventing staining incidents. This reduces potential losses in home laundry and operational risks in commercial laundry services. This is because, whether a staining risk is detected for the first time or after multiple identifications, the system takes immediate action, effectively preventing any possible staining events, protecting clothing from damage, and avoiding the need for users to perform additional work, such as rewashing or attempting to remove stains, thus improving laundry safety and user satisfaction.

[0043] In this embodiment, the washing machine first performs a single identification of the clothes. If a risk of staining is detected, the washing machine immediately pauses the washing program, preventing staining accidents. Simultaneously, an alarm message is sent to the user, allowing them to understand the situation and make the correct decision, preventing the need for secondary operations. If no staining risk is detected in the single identification, further analysis is performed. By repeatedly identifying the clothes in the washing machine, omissions in the single identification are avoided, ensuring the comprehensiveness and reliability of staining risk identification. Again, if a staining risk is detected, the washing machine immediately pauses the washing program, preventing staining accidents. An alarm message is sent to the user, allowing them to understand the situation and make the correct decision, thus preventing the washing machine from easily staining the clothes.

[0044] like Figure 3 As shown, multicolor detection consists of two main stages: initial detection and multiple detections.

[0045] Initial detection: After the user closes the washing machine door (regardless of whether a program has been selected), the system immediately initiates a rapid detection with a millisecond-level response. If a risk of mixing multi-colored clothes is detected, the system will immediately issue an alarm; otherwise, the process proceeds to subsequent multiple detection stages.

[0046] Multiple checks: After the user selects a washing program, the system controls the drum to perform one or more tumbling operations. After each tumbling, the system immediately performs a multi-color check again. Similarly, if a multi-color risk is detected, an alarm is immediately triggered; if no risk is detected, the user-selected washing program is executed normally.

[0047] In the specific implementation process, obtaining the first image of the clothes in the washing and care equipment can be achieved through the following steps: determining whether the door of the washing and care equipment is closed; and obtaining the image of the clothes in the washing and care equipment when the door is closed, thus obtaining the first image.

[0048] This solution addresses the issue of color bleeding in clothing, which often goes undetected before washing, leading to reactive measures during or after the wash cycle, increasing user inconvenience and costs. Therefore, by acquiring the first image immediately after the door closes, a high-speed camera minimizes latency, ensuring preliminary color recognition before the wash cycle begins. This not only improves the timeliness of recognition but also effectively avoids recognition errors caused by changes in clothing position or lighting conditions, providing an accurate and timely data foundation for subsequent color bleeding risk assessment.

[0049] The image acquisition step is triggered by monitoring the closing status of the washing machine door. Once the door closes, the camera immediately activates, acquiring the first image of the clothes inside with a millisecond-level response time. This ensures color detection is completed before the washing program begins, avoiding any delays. Instant response eliminates latency. By rapidly responding and acquiring the first image of the clothes the moment the door closes, the system can instantly initiate the color recognition process, effectively avoiding color bleeding problems caused by detection delays. This is because timely image acquisition ensures that the system has sufficient time for color analysis before any washing program begins, accurately identifying the initial color state of the clothes even during high-speed drum tumbling, thus providing early warning of staining risks.

[0050] If the initial test shows no risk, the washing cycle or user-selected program cannot proceed directly. Imagine the garments are initially captured in a state where the user randomly tossed them into the washing machine; obstructions could lead to missed detections. The drum needs to be tumbled to capture complete garment information for multi-color detection. If a single test already indicates a risk of staining, a warning can be issued immediately, eliminating the need for multiple tests and reducing detection time.

[0051] In some embodiments, a first identification result is obtained by performing a single staining analysis based on the first image. This can be achieved through the following steps: extracting colors from the first image and converting them to a color space; using a clustering algorithm to cluster the colors in the color space to obtain at least one color cluster; determining that the first identification result has a staining risk if the number of color clusters is greater than or equal to a preset number, the area of ​​each color cluster is greater than or equal to a preset area, and the brightness difference between any two color clusters is greater than or equal to a preset brightness difference; and determining that the first identification result has no staining risk if the number of color clusters is greater than or equal to the preset number, the area of ​​each color cluster is greater than or equal to the preset area, and the brightness difference between any two color clusters is greater than or equal to one or both of the preset brightness differences.

[0052] This solution converts colors to the CIE Lab color space, using three dimensions—L (brightness), a (red to green), and b (blue to yellow)—to represent color. Compared to color spaces like RGB, CIE Lab is more accurate in measuring color differences and less affected by changes in lighting conditions. Combined with an adaptive clustering algorithm, it automatically learns the complexity of the clothing color distribution and dynamically determines the optimal number of clusters, avoiding over-segmentation or under-segmentation problems that might arise from a preset K value. Finally, by setting reasonable thresholds for the number, area ratio, and brightness difference of color clusters, the system can accurately determine the presence or absence of staining risks, effectively avoiding false positives and false negatives, and ensuring washing safety.

[0053] This application's solution fully leverages the advantages of the CIE Lab color space, converting color information from the acquired first image into Lab values ​​for easier subsequent clustering analysis. An adaptive clustering algorithm is used to cluster colors within the color space, resulting in a series of color clusters, each representing a group of similar colors. Subsequently, the system checks whether the number of color clusters reaches or exceeds a preset K value (e.g., a preset number of 2), whether the area of ​​each color cluster exceeds a preset area percentage threshold (e.g., the area of ​​each color cluster is greater than or equal to 5%), and whether the brightness difference between two color clusters exceeds a preset value (e.g., the brightness difference is greater than or equal to 10). If all three conditions are met, a coloring risk is identified; if only one or two conditions are met, a coloring risk is identified. Accurate identification and effective differentiation are achieved. By employing the CIE Lab color space and a self-supervised learning clustering algorithm, combined with preset thresholds, effective identification and differentiation of coloring risks are realized. This is because the CIE Lab color space is designed to better suit human visual perception, enabling more accurate differentiation and identification of color differences. The self-supervised learning clustering algorithm can automatically adapt to various clothing combinations, determining the optimal number of clusters without human intervention. Combined with preset thresholds, such as the K value of the number of color clusters, the area ratio threshold, and the brightness difference, it forms an efficient and accurate dyeing risk identification framework that can effectively prevent color bleeding during the washing process.

[0054] Color recognition is performed using the CIE Lab color space, which provides a uniform perception. For easily confused colors, such as white, the criteria are: L > 95 and |a| < 5 and |b| < 5 (chromaticity tolerance range); for black, the criteria are: L < 15 and chromaticity is unlimited. This method can effectively distinguish between true black and false black, demonstrating its feasibility.

[0055] Existing solutions suffer from color space theory distortion. Traditional models such as HSV are severely affected by ambient light, leading to distortion in the brightness channel (e.g., light gray in shadows → false black, dark gray in reflected light → false white), resulting in low recognition accuracy. This application's solution adopts the CIE Lab color space and scientific thresholds: accurately identifying colors and effectively overcoming misjudgments caused by changes in lighting (overexposure / underexposure).

[0056] The intelligent multi-color recognition algorithm is shown in the figure. Figure 4 As shown, the core of this scheme lies in using an adaptive clustering algorithm based on self-supervised learning for color analysis and multi-color determination:

[0057] 1. Color Space: The algorithm performs clustering operations within the CIE Lab color space. This space represents color as L (lightness), a (red-green axis), and b (yellow-blue axis), and its design closely matches the visual perception characteristics of the human eye, making it more accurate than spaces such as RGB in measuring color differences.

[0058] 2. Algorithm advantages:

[0059] Automatic cluster number selection: The algorithm does not require a preset fixed number of color categories. It can automatically learn and determine the optimal number of clusters (K) based on the complexity of the color distribution of the clothes in the current drum. This solves the problem of over-segmentation or under-segmentation caused by improper preset K value in traditional methods, and significantly improves the adaptability to different combinations of clothes.

[0060] Precise Dark Color Differentiation: It excels at distinguishing between "True Black" and "Near-Black / Dark Colors" (such as dark navy and dark brown) that appear visually similar but have different spectral characteristics. This is a key capability that prevents dark-colored clothing from being mistakenly treated as "black" and thus allowed to be washed with light-colored clothing.

[0061] 3. Judgment Logic: The algorithm ultimately outputs the clustering results of the main colors of the clothing. For example... Figure 5 As shown, risk decisions are made based on color recognition results. If there are ≥2 significant color clusters, and the area of ​​each cluster is greater than 5% of the total area, an early warning is issued and feedback is provided to the user. The determination of multicolor risk is based on preset safety threshold rules (e.g., whether there are two main clusters whose L value (brightness) difference exceeds the threshold ΔLmax and whose cluster center distance is greater than the threshold ΔEab_min).

[0062] The multi-color detection method in this solution is deployed at the edge of the washing machine, which can achieve instantaneous response at the millisecond level or even faster, without requiring secondary operation by the user, and can effectively improve the prevention rate of color mixing accidents.

[0063] This solution employs an adaptive clustering algorithm based on self-supervised learning. This type of algorithm can be selected according to product requirements. Instead of using neural networks for training, it uses traditional clustering methods such as k-means++, DBSCAN, OPTICS, BIRCH, hierarchical clustering, and GMM. This problem scenario does not require the introduction of complex neural networks.

[0064] Data source: This consists of a large number of photos of clothes before washing, taken during single and multiple detection processes, used for algorithm verification and testing. K-value: This refers to the dominant color clusters in the captured images, an output value of the self-supervised learning algorithm, representing the number of main colors of the clothes in the image. The optimal K-value can be determined through extensive algorithm verification and testing on specific washing machine products.

[0065] This solution incorporates several innovations to ensure a stable data source: 1. Hardware control: Creates a stable and consistent lighting environment from the source, forcibly locks core imaging parameters, and can use downlights that approximate sunlight for illumination. 2. After obtaining stable image data, cluster analysis is performed in the CIE Lab color space.

[0066] The algorithm cannot guarantee that clustering is unaffected by lighting conditions; this is primarily controlled by hardware. The CIE Lab color space was chosen because it can distinguish between true black and pseudo black.

[0067] According to the characteristics of the CIE Lab color space, "true black" is characterized by an extremely low L value accompanied by neutral a and b values; while "pseudo black" is characterized by a low L value accompanied by a significant shift in a and / or b values.

[0068] This solution uses the CIE Lab color space, which provides a uniform perception, for color recognition. For easily confused colors, such as white, the criteria are: L > 95 and |a| < 5 and |b| < 5 (chromaticity tolerance range); for black, L < 15 and chromaticity is unlimited. The above is just a general technical standard. Further fine-tuning using algorithms can be performed based on the lighting conditions and shooting effects of different products and models to improve robustness.

[0069] Color distribution is the output value of the adaptive clustering algorithm of self-supervised learning, that is, the main color clusters in the captured image. It is an output value of the self-supervised learning algorithm mentioned above, representing how many colors of clothing are present in the image. In specific washing machine products, the self-supervised learning method can determine the optimal K value through a large number of algorithm verifications and tests.

[0070] Mixing dark colors: Washing black, navy blue, dark brown, and dark red together carries extremely low risk. No warning.

[0071] Mixing light colors: such as white, off-white, light gray, and light pink, together carries a lower risk. No warning.

[0072] High-risk scenarios: Color bleeding is highly likely to occur when washing dark and light colors together (such as black and white, red and white), or bright and light colors together (such as orange and pink). Warning.

[0073] In the specific implementation process, when the first identification result indicates no risk of staining, the acquisition of the second image of the clothes in the washing and care equipment can be achieved through the following steps: when the first identification result indicates no risk of staining, determine whether the washing program has been selected; when the washing program has been selected, control the drum of the washing and care equipment to rotate, and acquire the image of the clothes in the washing and care equipment after each drum rotation to obtain multiple second images.

[0074] In this system, if no staining risk is detected during the initial inspection, the system enters a waiting state until the user selects a washing program. Once a program is selected, the drum tumbling action is activated, allowing each side of the garment to be exposed to the camera. In the second image acquired after the drum tumbles, the system performs color analysis on each image to ensure that no new color combinations reach the preset staining risk threshold (e.g., at least two color clusters exist, each occupying more than 5% of the image area). Through this series of dynamic detections, the system can effectively monitor color changes in all garments, even when garments are stacked or obscured, thus avoiding any potential staining risks.

[0075] After initial detection confirms no risk of staining, the system does not immediately start the washing program but waits for the user to select a specific program. Once a program is selected, the system triggers the drum to tumble, changing the position and stacking of the clothes inside the drum, increasing the detection angle and field of view. Each time the drum tumbles, the system acquires a new image of the clothes—a second image—and may acquire multiple second images for analysis throughout the washing preparation process. The purpose of this process is to confirm whether the color distribution of the clothes consistently meets the requirements for safe washing from different perspectives, ensuring that unobstructed colors are exposed after the drum tumbles, thus preventing staining risks. Dynamic detection and comprehensive monitoring: By acquiring multiple second images after the drum tumbles, the system can dynamically capture color information of the clothes inside the washing machine over time, ensuring comprehensive monitoring of the staining status of the clothes. This is because clothes move during drum tumbles, and previously obstructed areas may be exposed. The second images captured at this time can cover more details, helping to more accurately assess staining risks and avoiding detection blind spots caused by clothing obstruction.

[0076] Existing solutions suffer from blind spots in dynamic detection, failing to identify the true colors of stacked / obscured clothing in a single static shot (e.g., dark clothing covered by light-colored clothing), thus missing the risk of mixed washing of multiple colors. This application's solution combines drum tumbling-triggered multi-frame image acquisition with temporal fusion analysis: dynamically capturing clothing colors under different stacking conditions to solve the problem of missed detection caused by occlusion.

[0077] The detection method for each of the multiple detections is the same as that for a single detection, and will not be elaborated here. The stacking state of the clothing is changed by repeatedly tumbling the roller, and images are captured at multiple key time points (the frame rate needs to be adjusted according to the actual application scenario; it is not a fixed parameter, as long as key information is captured). Subsequently, these image sequences from different time points are fused and analyzed to construct a more comprehensive and accurate set of clothing colors, greatly reducing the probability of missing colors due to occlusion in a single image.

[0078] In some embodiments, a second identification result is obtained by performing multiple staining analyses on the second image. This can be achieved through the following steps: When the second image is obtained each time the drum flips, a staining analysis is performed on each obtained second image to obtain multiple preliminary identification results, wherein each preliminary identification result corresponds one-to-one with the second image obtained each time the drum flips; if the number of preliminary identification results indicating a staining risk is greater than or equal to a preset threshold, the second identification result is determined to have a staining risk; if the number of preliminary identification results indicating a staining risk is less than the preset threshold, the second identification result is determined to have no staining risk.

[0079] In this scheme, it is assumed that most of the color on the surface of the clothing is correctly identified in the initial detection, and is therefore deemed to have no risk of staining. However, due to the stacking or obstruction of clothing, some small areas of dark-colored clothing may not be fully observed in the initial detection. After the roller is flipped, the previously obscured dark areas become apparent. When the system performs staining analysis on the new second image, it detects significant color differences twice consecutively (the preset threshold for the number of detections is two), indicating a potential risk of staining. By accumulating the initial identification results and setting an appropriate threshold, the system can effectively filter out misjudgments caused by accidental factors, ensuring that the alarm mechanism is triggered only when multiple pieces of evidence support the existence of a staining risk, thus improving the robustness and reliability of the detection strategy.

[0080] During the multiple stages of drum tumbling, the system performs independent staining analysis on the second image acquired after each tumbling, obtaining multiple preliminary identification results. Each preliminary identification result is an analysis result specific to a particular second image, ensuring the targeting and detail of the analysis. Next, the system checks how many of these preliminary identification results indicate a staining risk. If the number reaches or exceeds a preset threshold (e.g., the preset threshold is two), the final second identification result is determined to have a staining risk; conversely, if the number is below the preset threshold, the second identification result is considered to have no staining risk. Multiple verifications ensure detection reliability. Through multiple staining analyses, the system can comprehensively assess staining risk based on second image data from multiple angles and time points, improving the accuracy and reliability of early warnings. This is because a single preliminary identification result may be affected by accidental factors (such as changes in local lighting or temporary folding of clothing), leading to false judgments. The results of multiple analyses can corroborate each other; only when signs of staining risk are continuously found in multiple identifications will the system ultimately determine that a staining risk exists, which greatly reduces the possibility of false alarms and false negatives.

[0081] Existing detection methods suffer from a lag effect. Current technologies (such as sensor-based monitoring during the washing process) cannot proactively identify the risk of mixed washing of multi-colored garments before they enter the water. This results in reactive intervention only after staining problems occur, forcing users to interrupt the process for secondary treatment and significantly reducing washing efficiency. This application's solution employs an adaptive clustering algorithm based on self-supervised learning for color analysis and multi-color determination. A two-stage mechanism of "door-closing pre-inspection + program-based tumbling re-inspection" is implemented: multi-color risk screening and verification are completed before washing begins, at the moment the user closes the door (millisecond-level response) and during the first tumbling of the drum after selecting the program.

[0082] This solution proposes a process where "the system controls the roller to perform one or more tumbling operations. After each tumbling, the system immediately performs multi-color detection again" to address general problems. In specific applications, adjustments need to be made according to different models, requiring numerous experiments to select the most suitable parameters. However, the core objective remains the same: to maximize the capture of color information that may not be visible due to clothing stacking and obstruction, while minimizing the number of tumbling operations and image acquisition frames, and reducing resource consumption.

[0083] In the specific implementation process, before acquiring the first image of the clothes in the washing and care device and before acquiring the second image of the clothes in the washing and care device, the above method further includes the following steps: acquiring preset shooting parameters, wherein the preset shooting parameters include one or more of metering mode, exposure parameters, glare suppression, ISO sensitivity and white balance; and locking the shooting parameters of the image acquisition device to the preset shooting parameters.

[0084] In this scheme, the locking of preset shooting parameters means that whether acquiring the first or second image, the system will shoot using a uniform metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance, avoiding image color deviations caused by parameter changes. This consistent image acquisition method ensures that the data obtained during color recognition and cluster analysis is comparable and reliable, thus enabling accurate assessment of staining risks. For example, even in environments with significant changes in lighting conditions, such as switching from daytime to nighttime when using the washing machine, the system can still acquire highly consistent images based on uniform shooting parameters. This allows the algorithm to reliably identify the true color of clothing under different conditions, avoiding misjudgments caused by changes in lighting.

[0085] Before acquiring the first and second images, the system first reads a preset set of shooting parameters, including metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance. These parameters are carefully designed and calibrated to ensure color reproduction and consistency of the images. Next, the system locks the shooting parameters of the image acquisition device to the preset parameters to prevent color deviations caused by changes in the external environment (such as fluctuations in light intensity) during the acquisition process. Parameter locking ensures image consistency. By locking the preset shooting parameters, the system can guarantee that images acquired at different times and under different conditions have highly consistent color representation, meeting the key indicator in industrial inspection—image repeatability—thereby improving the reliability and accuracy of dyeing risk assessment. This is because stable and consistent shooting parameters can eliminate the influence of variables such as ambient light, exposure, and white balance on the accuracy of image color, ensuring that image data accurately reflects the true color distribution of clothing in both initial and subsequent inspections, providing a solid foundation for subsequent dyeing risk analysis.

[0086] Before performing single or multiple color tests, core shooting parameters such as metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance are forcibly locked. This standardized control not only ensures that the color reproduction of a single image closely approximates the true color gamut of the object, but also guarantees color consistency of the same garment across multiple frames in a time sequence. This fully meets the stringent requirements of image repeatability, a core indicator of industrial testing, and provides a data foundation that meets metrological accuracy standards for color difference calculation.

[0087] Existing solutions suffer from the drawback of non-reproducible image acquisition. Traditional image recognition schemes do not lock core parameters such as metering, exposure, and white balance, resulting in significant color difference errors when photographing the same garment at different locations and times within the tube. This renders industrial repeatability indicators ineffective and causes a surge in misjudgment rates. The solution proposed in this application forcibly locks core imaging parameters (metering, exposure, ISO, white balance, and glare suppression): ensuring high color fidelity and temporal consistency across multiple frames, thus meeting industrial-grade repeatability requirements.

[0088] To ensure the reliability of test results, especially to meet the critical "image repeatability" metric in industrial testing, the image acquisition process is strictly regulated: before acquiring images for single or multiple color tests, the system forcibly locks core shooting parameters, including but not limited to metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance. This measure ensures that the shooting parameters of all images acquired within a single washing cycle remain absolutely constant.

[0089] The core value of parameter locking lies in ensuring that the color reproduction of a single frame image closely approximates the true color gamut of the clothing. This guarantees that the same garment exhibits consistent color reproduction across multiple time-series acquired images, providing a reliable foundation for subsequent analysis.

[0090] After the washing machine door is closed, a stable and controllable imaging environment is created inside. By locking all variable shooting parameters, color representation differences caused by ambient light fluctuations and camera auto-adjustment are eliminated, ensuring:

[0091] 1. True color reproduction in single images: Approaching the true color gamut of clothing under standard light sources.

[0092] 2. Consistent timing of multiple frames: The color of the same garment remains consistent across different images throughout the entire inspection cycle, meeting the most critical requirement of image repeatability in industrial inspection.

[0093] How to adjust these parameters at the hardware level or in the software algorithm varies from "model" to "model". Generally, technicians need to debug according to the product manual and toolchain. It is not a uniform method, so it is difficult to give the implementation details here, and it is not the focus of this patent.

[0094] The system does not "automatically" adjust to different lighting conditions and lock parameters. Before the product leaves the factory, technicians have locked the metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance based on the specific characteristics of the model, the lighting conditions of the downlight, and the camera, and have conducted multiple experiments to ensure the robustness of these parameters in application.

[0095] In some embodiments, after performing multiple dyeing analyses based on the second image to obtain a second identification result, the method further includes the following steps: if the first identification result indicates no dyeing risk and the second identification result indicates no dyeing risk, determine that the clothing has no dyeing risk, and control the washing and care equipment to operate according to the washing procedure.

[0096] In this scheme, during the initial detection phase, if the system determines that the clothes pose no risk of staining, this is only a preliminary conclusion drawn from the first image information acquired at the moment the door closes. However, to further verify this conclusion, the system controls the drum to rotate after the washing program is selected and acquires a second image for independent staining analysis. If the results of this series of second image analyses also indicate no risk of staining, the system will combine the results of the two detections to confirm that the clothes as a whole pose no risk of staining, thus safely starting the washing program. This process, through a double confirmation mechanism, ensures that washing will only begin if there is no risk of color bleeding, greatly reducing the probability of color bleeding accidents caused by inaccurate single detections, and providing a solid guarantee for the safety of intelligent laundry.

[0097] After the system completes the initial detection and subsequent multiple detections, it comprehensively evaluates the first and second identification results. Only when both detection results indicate no risk of staining will the system ultimately determine that the clothes are safe from staining and automatically control the washing equipment to start working according to the user-selected washing program. This strategy ensures the comprehensiveness and reliability of the risk assessment. Double confirmation, safe start. Through the double confirmation of the initial detection and multiple detections, the system can ensure the safe start of the washing program when there is no risk of staining, avoiding color bleeding problems and significantly improving the user experience of smart laundry. This is because the initial detection provides a preliminary screening for staining risks, while the multiple detections further verify the results of the initial detection through drum tumbling after the washing program is selected. Only when both detection results consistently indicate no risk of staining will the system determine that the clothes are safe and start the normal washing process. This greatly enhances the accuracy and reliability of the risk assessment, providing users with a more reassuring smart laundry environment.

[0098] Developing a method for detecting multiple colors in clothing is of great significance, as it can improve the intelligence level of washing machines, enhance user experience, and adapt to the needs of commercial laundry scenarios. To improve user experience, this application proposes a "method for identifying the color of clothing in laundry equipment." The detection process consists of initial detection and multiple detections. Initial detection begins as soon as the user closes the washing machine door (regardless of whether the user has selected a program), completing a single detection within a very short time (millisecond-level response time). If multiple colors are detected, a report is immediately issued; otherwise, further detection proceeds to subsequent steps. Multiple detections occur after the user selects a program, the drum tumbles once or multiple times, and multiple colors are detected again after each tumbling. Under this process, users can eliminate the risk of staining only during the pre-selection stage, without consuming excessive time, thus improving the intelligent laundry experience.

[0099] This solution employs an adaptive clustering algorithm based on self-supervised learning for color analysis and multi-color determination. Through a triple breakthrough of "dual-stage early warning mechanism + industrial-grade imaging control + CIE Lab color gamut reconstruction," it can effectively prevent color bleeding accidents during washing, reduce user losses (in home scenarios) and business operation risks (complaints and compensation), and significantly enhance the practical value and market competitiveness of smart washing machines.

[0100] In summary, existing technologies suffer from several drawbacks, including delayed detection (problems can only be detected during or after washing, requiring secondary user intervention), unreliable recognition results due to non-standardized image acquisition parameters (susceptible to ambient light, stacking, and occlusion, resulting in large color difference errors), and the failure of color recognition models (such as the HSV color space) under complex lighting conditions (inability to accurately distinguish between black and white and similar colors). This solution, through a well-designed color detection process, locks in key camera parameters, and employs an adaptive clustering algorithm based on self-supervised learning for color analysis and multi-color determination (the algorithm performs clustering operations within the CIE Lab color space). This significantly improves the accuracy, reliability, and timeliness of multi-colored clothing recognition, providing a practical and feasible technical solution for smart washing machines to avoid color bleeding risks, representing a significant advancement.

[0101] This application also provides a garment color recognition device for a laundry and care device. It should be noted that this garment color recognition device can be used to execute the garment color recognition method for a laundry and care device provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0102] The following describes the clothing color recognition device for the washing and care equipment provided in the embodiments of this application.

[0103] Figure 6 This is a structural block diagram of a clothing color recognition device for a laundry and care equipment according to an embodiment of this application. Figure 6 As shown, the device includes:

[0104] The first acquisition unit 10 is used to acquire a first image of the clothes in the washing and care equipment;

[0105] The first analysis unit 20 is used to perform a single dyeing analysis based on the first image without selecting a washing program, and obtain a first identification result, wherein the first identification result is used to characterize whether the clothing identified in a single dyeing operation has a risk of dyeing.

[0106] The second acquisition unit 30 is used to acquire a second image of the clothing in the washing and care equipment when the first identification result indicates no risk of staining.

[0107] The second analysis unit 40 is used to perform multiple dyeing analyses based on the second image when the above washing program has been selected, and to obtain a second identification result, wherein the second identification result is used to characterize whether the repeatedly identified clothing has a risk of dyeing.

[0108] The first processing unit 50 is used to determine that the clothing has a risk of staining when the first identification result indicates a risk of staining or the second identification result indicates a risk of staining, and to generate an alarm message.

[0109] In this embodiment, the washing machine first performs a single identification of the clothes. If a risk of staining is detected, the washing machine immediately pauses the washing program, preventing staining accidents. Simultaneously, an alarm message is sent to the user, allowing them to understand the situation and make the correct decision, preventing the need for secondary operations. If no staining risk is detected in the single identification, further analysis is performed. By repeatedly identifying the clothes in the washing machine, omissions in the single identification are avoided, ensuring the comprehensiveness and reliability of staining risk identification. Again, if a staining risk is detected, the washing machine immediately pauses the washing program, preventing staining accidents. An alarm message is sent to the user, allowing them to understand the situation and make the correct decision, thus preventing the washing machine from easily staining the clothes.

[0110] In the specific implementation process, the first acquisition unit includes a first determination module and a first acquisition module. The first determination module is used to determine whether the door of the washing and care equipment is closed. The first acquisition module is used to acquire an image of the clothes in the washing and care equipment when the door of the washing and care equipment is closed, and obtain the first image.

[0111] This solution addresses the issue of color bleeding in clothing, which often goes undetected before washing, leading to reactive measures during or after the wash cycle, increasing user inconvenience and costs. Therefore, by acquiring the first image immediately after the door closes, a high-speed camera minimizes latency, ensuring preliminary color recognition before the wash cycle begins. This not only improves the timeliness of recognition but also effectively avoids recognition errors caused by changes in clothing position or lighting conditions, providing an accurate and timely data foundation for subsequent color bleeding risk assessment.

[0112] In some embodiments, the first analysis unit includes an extraction module, a clustering module, a second determination module, and a third determination module. The extraction module is used to extract colors from the first image and convert the colors to a color space. The clustering module is used to cluster the colors in the color space using a clustering algorithm to obtain at least one color cluster. The second determination module is used to determine that the first identification result is a risk of staining if the number of color clusters is greater than or equal to a preset number, the area of ​​each color cluster is greater than or equal to a preset area, and the brightness difference between any two color clusters is greater than or equal to a preset brightness difference. The third determination module is used to determine that the first identification result is a risk of staining if the number of color clusters is greater than or equal to the preset number, the area of ​​each color cluster is greater than or equal to the preset area, and the brightness difference between any two color clusters is greater than or equal to one or both of the preset brightness differences.

[0113] This solution converts colors to the CIE Lab color space, using three dimensions—L (brightness), a (red to green), and b (blue to yellow)—to represent color. Compared to color spaces like RGB, CIE Lab is more accurate in measuring color differences and less affected by changes in lighting conditions. Combined with an adaptive clustering algorithm, it automatically learns the complexity of the clothing color distribution and dynamically determines the optimal number of clusters, avoiding over-segmentation or under-segmentation problems that might arise from a preset K value. Finally, by setting reasonable thresholds for the number, area ratio, and brightness difference of color clusters, the system can accurately determine the presence or absence of staining risks, effectively avoiding false positives and false negatives, and ensuring washing safety.

[0114] In the specific implementation process, the second acquisition unit includes a fourth determination module and a second acquisition module. The fourth determination module is used to determine whether the washing program has been selected if the first identification result indicates no risk of staining. The second acquisition module is used to control the drum of the washing and care equipment to rotate when the washing program has been selected, and to acquire an image of the clothes in the washing and care equipment after each drum rotation to obtain multiple second images.

[0115] In this system, if no staining risk is detected during the initial inspection, the system enters a waiting state until the user selects a washing program. Once a program is selected, the drum tumbling action is activated, allowing each side of the garment to be exposed to the camera. In the second image acquired after the drum tumbles, the system performs color analysis on each image to ensure that no new color combinations reach the preset staining risk threshold (e.g., at least two color clusters exist, each occupying more than 5% of the image area). Through this series of dynamic detections, the system can effectively monitor color changes in all garments, even when garments are stacked or obscured, thus avoiding any potential staining risks.

[0116] In some embodiments, the second analysis unit includes an analysis module, a fifth determination module, and a sixth determination module. The analysis module is used to perform staining analysis on each second image obtained after each drum tumble to obtain multiple preliminary identification results, wherein each preliminary identification result corresponds one-to-one with the second image obtained after each drum tumble. The fifth determination module is used to determine that the second identification result is stained if the number of preliminary identification results indicating staining risk is greater than or equal to a preset number threshold. The sixth determination module is used to determine that the second identification result is not stained if the number of preliminary identification results indicating staining risk is less than the preset number threshold.

[0117] In this scheme, it is assumed that most of the color on the surface of the clothing is correctly identified in the initial detection, and is therefore deemed to have no risk of staining. However, due to the stacking or obstruction of clothing, some small areas of dark-colored clothing may not be fully observed in the initial detection. After the roller is flipped, the previously obscured dark areas become apparent. When the system performs staining analysis on the new second image, it detects significant color differences twice consecutively (the preset threshold for the number of detections is two), indicating a potential risk of staining. By accumulating the initial identification results and setting an appropriate threshold, the system can effectively filter out misjudgments caused by accidental factors, ensuring that the alarm mechanism is triggered only when multiple pieces of evidence support the existence of a staining risk, thus improving the robustness and reliability of the detection strategy.

[0118] In the specific implementation process, the above-mentioned device also includes a third acquisition unit and a locking unit. The third acquisition unit is used to acquire preset shooting parameters before acquiring the first image of the clothes in the washing and care device and before acquiring the second image of the clothes in the washing and care device. The preset shooting parameters include one or more of metering mode, exposure parameters, glare suppression, ISO sensitivity and white balance. The locking unit is used to lock the shooting parameters of the image acquisition device to the preset shooting parameters.

[0119] In this scheme, the locking of preset shooting parameters means that whether acquiring the first or second image, the system will shoot using a uniform metering mode, exposure parameters, glare suppression, ISO sensitivity, and white balance, avoiding image color deviations caused by parameter changes. This consistent image acquisition method ensures that the data obtained during color recognition and cluster analysis is comparable and reliable, thus enabling accurate assessment of staining risks. For example, even in environments with significant changes in lighting conditions, such as switching from daytime to nighttime when using the washing machine, the system can still acquire highly consistent images based on uniform shooting parameters. This allows the algorithm to reliably identify the true color of clothing under different conditions, avoiding misjudgments caused by changes in lighting.

[0120] In some embodiments, the above-described apparatus further includes a second processing unit, which is used to determine that the clothing has no risk of staining after performing multiple dyeing analyses based on the second image to obtain a second identification result, and when the first identification result indicates no risk of staining and the second identification result indicates no risk of staining, and control the washing and care equipment to operate according to the washing procedure.

[0121] In this scheme, during the initial detection phase, if the system determines that the clothes pose no risk of staining, this is only a preliminary conclusion drawn from the first image information acquired at the moment the door closes. However, to further verify this conclusion, the system controls the drum to rotate after the washing program is selected and acquires a second image for independent staining analysis. If the results of this series of second image analyses also indicate no risk of staining, the system will combine the results of the two detections to confirm that the clothes as a whole pose no risk of staining, thus safely starting the washing program. This process, through a double confirmation mechanism, ensures that washing will only begin if there is no risk of color bleeding, greatly reducing the probability of color bleeding accidents caused by inaccurate single detections, and providing a solid guarantee for the safety of intelligent laundry.

[0122] The clothing color recognition device of the aforementioned laundry care equipment includes a processor and a memory. The first acquisition unit, first analysis unit, second acquisition unit, second analysis unit, and first processing unit are all stored as program units in the memory. The processor executes the program units stored in the memory to achieve the corresponding functions. All of the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0123] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can address the problem of color bleeding in existing washing machines.

[0124] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0125] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform a method for recognizing the color of clothing in the washing and care device.

[0126] This invention provides a processor for running a program, wherein the program executes a method for recognizing the color of clothing in the washing and care device.

[0127] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements at least the method steps for recognizing the color of clothing in a laundry device. The device described herein can be a server, PC, tablet, mobile phone, etc.

[0128] This application also provides a computer program product that, when executed on a data processing device, is adapted to perform a program that initializes a method for identifying the color of clothing using at least the following washing and care equipment.

[0129] This application also provides a laundry care device system, including one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a method for recognizing the color of clothing in any of the above-described laundry care devices.

[0130] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0131] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0132] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. 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... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0133] 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.

[0134] 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.

[0135] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0136] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0137] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0138] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0139] 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 process, method, article, or apparatus. Unless otherwise specified, 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.

[0140] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method of identifying a color of laundry of a washing and caring apparatus, characterized by, The method comprises: obtaining a first image of clothes in a washing and protecting device; in the case where no washing program is selected, performing single-staining analysis according to the first image to obtain a first identification result, wherein the first identification result is used to represent whether the clothes have a staining risk; the first identification result is obtained by performing single-staining analysis according to the first image, comprising: extracting colors in the first image and converting the colors to a color space; clustering the colors in the color space using a clustering algorithm to obtain at least one color cluster; in the case where the number of color clusters is greater than or equal to a preset number, the area of each color cluster is greater than or equal to a preset area, and the brightness difference between any two color clusters is greater than or equal to a preset brightness difference, it is determined that the first identification result is a staining risk; in the case where the number of color clusters is greater than or equal to the preset number, the area of each color cluster is greater than or equal to the preset area, and the brightness difference between any two color clusters is greater than or equal to the preset brightness difference, it is determined that the first identification result is no staining risk; in the case where the first identification result represents no staining risk, obtaining a second image of the clothes in the washing and protecting device; in the case where the washing program has been selected, performing multi-staining analysis according to the second image to obtain a second identification result, wherein the second identification result is used to represent whether the clothes have a staining risk; in the case where the first identification result represents a staining risk or the second identification result represents a staining risk, it is determined that the clothes have a staining risk, and an alarm information is generated.

2. The method of claim 1, wherein, The method comprises: determining whether the door body of the washing and protecting device is closed; in the case where the door body of the washing and protecting device is closed, obtaining an image of the clothes in the washing and protecting device to obtain the first image.

3. The method of claim 1, wherein, in the case where the first identification result represents no staining risk, obtaining a second image of the clothes in the washing and protecting device, comprising: in the case where the first identification result represents no staining risk, determining whether the washing program has been selected; in the case where the washing program has been selected, controlling the drum of the washing and protecting device to reverse, and obtaining an image of the clothes in the washing and protecting device after each drum reversal to obtain a plurality of second images.

4. The method of claim 3, wherein, performing multi-staining analysis according to the second image to obtain a second identification result, comprising: in the case where the second image is obtained after each drum reversal, performing staining analysis on each obtained second image to obtain a plurality of preliminary identification results, wherein the preliminary identification result and the second image obtained after each drum reversal correspond one by one; in the case where the number of preliminary identification results representing a staining risk is greater than or equal to a preset number threshold, it is determined that the second identification result is a staining risk; in the case where the number of preliminary identification results representing a staining risk is less than the preset number threshold, it is determined that the second identification result is no staining risk.

5. The method according to any one of claims 1 to 4, characterized in that, Before acquiring the first image of the laundry of the washing and treating apparatus, before acquiring the second image of the laundry of the washing and treating apparatus, the method further comprises: acquiring preset shooting parameters, wherein the preset shooting parameters comprise one or more of an exposure metering mode, an exposure parameter, a glare suppression, an ISO sensitivity, and a white balance; locking the shooting parameters of the image acquisition device as the preset shooting parameters.

6. The method according to any one of claims 1 to 4, characterized in that, After performing multiple dyeing analysis according to the second image to obtain a second recognition result, the method further comprises: in a case where the first recognition result represents no dyeing risk and the second recognition result represents no dyeing risk, determining that the laundry has no dyeing risk, and controlling the washing and treating apparatus to work according to the washing program.

7. A laundry color identification device of a washing and drying apparatus, characterized by, comprise: a first acquisition unit, configured to acquire a first image of laundry of a washing and treating apparatus; a first analysis unit, configured to, in a case where no washing program is selected, perform single-dyeing analysis according to the first image to obtain a first recognition result, wherein the first recognition result is used to represent whether the laundry has dyeing risk in single-dyeing identification; the first analysis unit comprises an extraction module, a clustering module, a second determination module, and a third determination module, the extraction module is configured to extract colors in the first image and convert the colors to a color space; the clustering module is configured to use a clustering algorithm to cluster the colors in the color space to obtain at least one color cluster; the second determination module is configured to, in a case where the number of the color clusters is greater than or equal to a preset number, the area of each color cluster is greater than or equal to a preset area, and the brightness difference between any two color clusters is greater than or equal to a preset brightness difference, determine that the first recognition result is dyeing risk; the third determination module is configured to, in a case where the number of the color clusters is greater than or equal to the preset number, the area of each color cluster is greater than or equal to the preset area, and the brightness difference between any two color clusters is greater than or equal to the preset brightness difference, determine that the first recognition result is no dyeing risk; a second acquisition unit, configured to, in a case where the first recognition result represents no dyeing risk, acquire a second image of the laundry of the washing and treating apparatus; a second analysis unit, configured to, in a case where the washing program has been selected, perform multiple-dyeing analysis according to the second image to obtain a second recognition result, wherein the second recognition result is used to represent whether the laundry has dyeing risk in multiple-dyeing identification; a first processing unit, configured to, in a case where the first recognition result represents dyeing risk or the second recognition result represents dyeing risk, determine that the laundry has dyeing risk, and generate an alarm information.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the laundry color recognition method of the washing and treating apparatus in any one of claims 1 to 6.

9. A laundry appliance system, characterized by, comprise: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including programs for performing the method for identifying the color of laundry of the washing and caring apparatus according to any one of claims 1 to 6.

Images