Mattress and pillow recommendation method and device based on 3D structured light, equipment and medium

Abstract

The invention provides a mattress and pillow recommendation method and device based on 3D structured light, equipment and a medium, and the method comprises the steps: obtaining image data and preference data of a target client, extracting key parameters of the image data, and inputting the key parameters and the preference data into a preset mattress recommendation model and a pillow recommendation model; calculating the hardness of the mattress by the mattress recommendation model, and calculating and matching based on the hardness of the mattress, a preset mapping rule and a preset mattress hardness information base to obtain a mattress hardness grade and a target mattress; the mattress hardness is input into a pillow recommendation model, the pillow recommendation model calculates the pillow supporting height according to the key parameters, the preference data and the mattress hardness, and a target pillow is obtained through matching in a preset pillow database based on the key parameters, the pillow supporting height and preset pillow material depression information; and sending the mattress hardness grade, the target mattress, the pillow supporting height and the target pillow to a client so as to solve the problem that mattress and pillow recommendation is inaccurate and unscientific.

Description

Recommended methods, devices, equipment, and media for mattresses and pillows based on 3D structured light. Technical Field

[0001] This invention relates to the field of computer vision technology, and in particular to a method, apparatus, device, and medium for recommending mattresses and pillows based on 3D structured light. Background Technology

[0002] As living standards improve, people are paying increasing attention to sleep quality. Mattresses and pillows, as key products affecting sleep quality, are directly related to spinal health and muscle fatigue relief; therefore, choosing the right mattress and pillow has become an important need. However, currently, consumers still mainly rely on traditional methods such as user trial experiences and salesperson recommendations when purchasing mattresses and pillows.

[0003] This traditional approach has significant drawbacks: users' brief experience of lying down in a store is highly subjective and cannot accurately predict the long-term support effect of the mattress on the spine; moreover, the fit of the same mattress varies significantly among users of different body types and sleeping positions, and sales staff make recommendations based solely on personal experience without the support of scientific data, making it difficult to guarantee the accuracy of the recommendations.

[0004] Therefore, the inventors of this application have found that existing mattress and pillow recommendation methods generally suffer from inaccurate, unscientific, and inefficient recommendations, and cannot achieve a coordinated match between mattress firmness and pillow height, making it difficult to meet consumers' personalized and scientific purchasing needs. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for recommending mattresses and pillows based on 3D structured light, aiming to solve the problems of inaccurate, unscientific, and inefficient mattress and pillow recommendations.

[0006] In a first aspect, embodiments of the present invention provide a mattress and pillow recommendation method based on 3D structured light, comprising: acquiring image data and preference data of a target customer; extracting key parameters from the image data; and inputting the key parameters and preference data into a preset mattress recommendation model and a pillow recommendation model; the mattress recommendation model calculating mattress firmness based on the key parameters and preference data, and calculating and matching a mattress firmness level and a target mattress based on the mattress firmness, preset mapping rules, and a preset mattress firmness information database; inputting the mattress firmness into the pillow recommendation model, which calculates pillow support height based on the key parameters, preference data, and mattress firmness, and matching a target pillow in a preset pillow database based on the key parameters, pillow support height, and preset pillow material indentation information; and sending the mattress firmness level, the target mattress, the pillow support height, and the target pillow to a client.

[0007] Secondly, embodiments of the present invention also provide a mattress and pillow recommendation device based on 3D structured light, which includes a unit for performing the above-described method.

[0008] Thirdly, embodiments of the present invention also provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0009] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, can implement the above-described method.

[0010] This application provides a method, apparatus, device, and medium for recommending mattresses and pillows based on 3D structured light. It uses 3D structured light technology to scan and acquire three-dimensional images and preference data of a target customer, extracting key body parameters. These parameters and preference data are then input into a mattress recommendation model and a pillow recommendation model for collaborative calculation. This ensures that the calculated mattress firmness matches the target customer's actual body shape and preferences, and that the pillow support height is adapted to the recommended mattress firmness and the target customer's actual body shape and preferences. This achieves precise and coordinated matching between the mattress and pillow, enabling automated, high-precision, and high-efficiency acquisition of the user's human morphological characteristics. Based on ergonomic principles, it scientifically recommends the most suitable mattress and pillow for the target customer's health needs, solving the problems of inaccurate, unscientific, and inefficient mattress and pillow recommendations currently in place. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 is a schematic flowchart of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 2 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 3 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 4 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 5 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 6 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 7 is a schematic diagram of a sub-process of the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 8 is a schematic diagram of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention. Figure 9 is a schematic diagram of a sub-process of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 10 is a schematic diagram of a sub-process of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 11 is a schematic diagram of a sub-process of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 12 is a schematic diagram of a sub-process of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 13 is a schematic diagram of a sub-process of a mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention; Figure 14 is a schematic block diagram of a mattress and pillow recommendation device based on 3D structured light provided in an embodiment of the present invention; Figure 15 is a schematic block diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0015] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0016] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0017] Please refer to Figure 1, which is a schematic flowchart of the mattress and pillow recommendation method based on 3D structured light provided in this embodiment of the invention. In this application, the mattress and pillow recommendation method based on 3D structured light is applied to the fields of smart home and healthy sleep management. In high-end home furnishing stores or sleep experience centers, customers stand in the 3D structured light scanning area, and the device quickly acquires their three-dimensional body image. Simultaneously, customers input their preference data through a terminal. The system backend can generate a personalized report within minutes, including the specific mattress and pillow product models, mattress zone firmness settings, overall mattress firmness settings, pillow zone support height, and pillow support height suggestions. Sales personnel can use this report for precise product guidance and adjustments, while customers can intuitively understand the scientific basis of the recommendations. This application scenario transforms traditional experience-based purchasing into data-driven precise matching, significantly improving the shopping experience, product satisfaction, and sleep health levels.

[0018] This application provides a method, apparatus, computer device, and storage medium for recommending mattresses and pillows based on 3D structured light. The method includes: acquiring image data and preference data of a target customer; extracting key parameters from the image data; and inputting the key parameters and preference data into preset mattress recommendation models and pillow recommendation models; the mattress recommendation model calculates mattress firmness based on the key parameters and preference data, and calculates and matches a mattress firmness level and a target mattress based on the mattress firmness, preset mapping rules, and a preset mattress firmness information database; the mattress firmness is input into the pillow recommendation model, which calculates pillow support height based on the key parameters, preference data, and mattress firmness, and matches a target pillow in a preset pillow database based on the key parameters, pillow support height, and preset pillow material indentation information; and the mattress firmness level, the target mattress, the pillow support height, and the target pillow are sent to a client.

[0019] This application utilizes 3D structured light technology to scan and acquire three-dimensional images and preference data of target customers, extracts their key body parameters, and inputs these parameters and preference data into a mattress recommendation model and a pillow recommendation model for collaborative calculation. This ensures that the calculated mattress firmness matches the target customer's actual body shape and preference needs, and that the pillow support height is adapted to the recommended mattress firmness and the target customer's actual body shape and preference needs. This achieves precise and coordinated matching between the mattress and pillow, enabling automated, high-precision, and high-efficiency acquisition of the user's human morphological characteristics. Based on ergonomic principles, it scientifically recommends the most suitable mattress and pillow for the target customer's health needs, solving the current problems of inaccurate, unscientific, and inefficient mattress and pillow recommendations.

[0020] Figure 1 is a flowchart illustrating the mattress and pillow recommendation method based on 3D structured light provided in an embodiment of the present invention. As shown in Figure 1, the method includes the following steps S10-S40.

[0021] S10. Acquire image data and preference data of the target customer, extract key parameters from the image data, and input the key parameters and preference data into preset mattress recommendation models and pillow recommendation models. Specifically, this embodiment is a data acquisition and data processing flow, mainly including two parts: data acquisition hardware and data processing module. The data acquisition hardware is a mirror composed of two 3D structured light cameras with a display function. The cameras emit invisible infrared light onto the human body surface and capture the deformed pattern. The three-dimensional point cloud data of the human body surface is calculated using the triangulation principle. This device can complete a full-body scan within one second with millimeter-level accuracy. Data processing module: Deployed on a local server or in the cloud, this module receives the raw point cloud data and performs preprocessing and parameter extraction tasks to obtain the key parameters of the target customer.

[0022] Specifically, 3D structured light refers to the technology of acquiring three-dimensional information of an object by projecting structured light patterns; the target customer refers to consumers who have purchasing needs for mattresses and pillows; the image data refers to a dataset containing three-dimensional geometric information of the target customer's body surface, acquired through non-contact sensing devices such as depth cameras and 3D scanners. The preference data refers to customer subjective selection information obtained through questionnaires, interactive interfaces, or historical records, including datasets of preferences for product hardness, material, size, quality, price, replacement frequency, sleeping posture, sleep quality, sleep duration, and sleep state; the key parameters are the customer's three-dimensional morphology transformed into a set of quantifiable feature values, which are quantitative indicators of key parts of the human body, including data such as shoulder width, scoliosis angle, lumbar physiological curvature depth, shoulder-to-hip ratio, cervical curvature height, volume, and weight.

[0023] The mattress recommendation model is a machine learning model (such as a gradient boosting tree or neural network) based on the relationship between human body pressure distribution and support mechanics and trained on ergonomics, used to output the appropriate mattress type, firmness level and zoned support scheme.

[0024] The pillow recommendation model is a model trained based on the geometric features of the head, neck and shoulders and the association rules of sleeping posture. It is used to output the appropriate pillow height, zone customization scheme and material type.

[0025] In practice, a 3D structured light acquisition device scans the target customer to obtain image data containing information such as body contours, spinal and cervical spine physiological curves. Simultaneously, interactive acquisition methods are used to collect the target customer's preference data. The acquired image data is preprocessed to remove interfering information and extract key parameters such as shoulder width, waist-to-hip ratio, body pressure distribution, and cervical curvature. These standardized key parameters, along with the preference data, are synchronously input into preset mattress and pillow recommendation models, providing core data support for generating accurate recommendations. This embodiment utilizes 3D structured light technology for data acquisition and processing, allowing the target customer to replace a lengthy trial lying process with a scan that takes only a few seconds. It can also detect spinal posture problems that are difficult to observe with the naked eye, thereby achieving precise matching and improving the efficiency and accuracy of mattress and pillow selection.

[0026] In one embodiment, as shown in FIG2, step S10 includes steps S11-S12.

[0027] S11. Preprocess the image data to generate a three-dimensional human body mesh model; S12. Extract parameters from the three-dimensional human body mesh model to obtain the key parameters.

[0028] Specifically, the target customer is first scanned using a 3D structured light acquisition device to obtain image data. This image data refers to the original point cloud or depth image sequence containing the three-dimensional spatial information of the target customer's body surface, acquired by a depth sensing device. Simultaneously, the target customer's preference data is collected through interactive acquisition. The acquired image data is then preprocessed, including background noise removal, data calibration, stitching and fusion of multi-view data, and coordinate normalization, to eliminate interference and generate a complete, smooth, and accurately representative three-dimensional human body mesh model. This model is a digital model composed of numerous interconnected triangular facets that precisely reflects the geometric shape of the customer's body surface.

[0029] Next, key parameters are extracted based on the three-dimensional human body mesh model. Through the feature recognition algorithm built into the model, key parameters closely related to mattress support and pillow fit, such as shoulder width, scoliosis angle, lumbar physiological curvature depth, shoulder-to-hip ratio, cervical curvature height, volume, and weight, are accurately extracted. The standardized key parameters and preference data are then synchronously input into the preset mattress recommendation model and pillow recommendation model to provide core data support for generating accurate recommendation results in the future.

[0030] This embodiment completes the key parameter extraction steps from the original image to quantify body features.

[0031] S20. The mattress recommendation model calculates the mattress firmness based on the key parameters and the preference data, and calculates and matches the mattress firmness level and target mattress based on the mattress firmness, preset mapping rules and preset mattress firmness information database.

[0032] Specifically, the mattress firmness is a continuous numerical index calculated by a model to quantitatively characterize the mattress's support properties. The preset mapping rule is a set of predefined thresholds or functional relationships used to classify continuous firmness values ​​into discrete firmness levels. In this embodiment, the mapping is based on the firmness calculation method in the national standard GB / T43007-2023 "Test and Evaluation Method for Mattress Firmness Level Distribution". The preset mattress firmness information database is a data set storing detailed attribute records for various mattress models, with each record containing at least the measured firmness value or nominal level of the mattress.

[0033] In practice, standardized key parameters and preference data are simultaneously input into a pre-defined mattress recommendation model. The model combines these features and uses a built-in algorithm to accurately calculate a quantitative value for mattress firmness suitable for the customer, i.e., mattress firmness. The model then retrieves pre-defined mapping rules to convert the calculated mattress firmness into a corresponding firmness level. Simultaneously, it accesses a pre-defined mattress firmness database, performing a matching query to select one or more candidate mattresses with a firmness consistent with or closest to the calculated firmness. Finally, it outputs one or more highly recommended target mattress models and their detailed parameters, thus completing the personalized matching from user characteristics to specific products.

[0034] This embodiment not only considers the key parameters of the target customer but also incorporates their preferences to provide a "tailor-made" solution for each customer, providing a data foundation for customized mattress production. Therefore, the mattress recommendation method in this embodiment can automatically and accurately acquire the user's human morphological characteristics and, based on ergonomic principles, scientifically recommend the most suitable mattress for the user's spinal health needs.

[0035] In one embodiment, as shown in FIG3, the key parameters include volume data and weight data of each part of the preset human body partition, and step S20 includes steps S21-S23.

[0036] S21. Calculate the basic firmness of each zone based on the volume data and weight data of each zone; S22. Calculate the overall mattress reference firmness based on the basic firmness of each zone and a preset calculation algorithm; S23. Correct the mattress reference firmness according to the preference data to obtain the mattress firmness.

[0037] Specifically, the preset human body partitions are based on ergonomically defined body parts corresponding to the mattress support areas. In this embodiment, the human body is divided into five partitions: head, shoulders and back, waist, hips, and legs. The volume and weight data are quantified indicators obtained through voxelization analysis and density estimation of the three-dimensional human body mesh model. The volume data is the three-dimensional volume quantification value of each partition, and the weight data is the weight quantification value of each partition. The partition base firmness refers to the support firmness value adapted to that part, calculated based on the volume and weight data of a single partition. The preset calculation algorithm refers to the pre-set calculation logic that integrates the base firmness of each partition to obtain the overall firmness of the mattress. The mattress reference firmness refers to the overall initial firmness quantification value of the mattress obtained by combining the base firmness of all partitions. The mattress firmness refers to the final adapted firmness quantification value determined after correction by preference data.

[0038] During implementation, the mattress recommendation model first calculates the basic firmness of each zone based on the volume and weight data of each zone using built-in logic (e.g., combining the allowable pressure per unit area corresponding to the physiological characteristics of the zone, and using mechanical support formulas, etc.) to meet the requirements for balanced support.

[0039] Subsequently, the mattress recommendation model uses a preset calculation algorithm (such as a weighted average algorithm) to comprehensively calculate the basic firmness of all zones and determine the benchmark firmness of the mattress that represents the overall support requirements.

[0040] Finally, the mattress recommendation model combines preference data to analyze customers' subjective needs for firmness (such as sleeping posture, firmness preference, etc.), and makes targeted adjustments to the benchmark firmness of the mattress to obtain a mattress firmness that accurately matches the customer's body shape and subjective needs.

[0041] This embodiment first calculates the mattress firmness of different zones, then calculates the overall firmness of the mattress using a weighted average, and finally adjusts it based on customer preferences to obtain the final mattress firmness requirement. This allows for accurate mattress recommendations and improves the user experience.

[0042] In one embodiment, as shown in FIG4, step S21 is followed by steps S211-S213.

[0043] S211. Based on the key parameters and the preset correction algorithm, the basic softness and hardness of each partition are corrected to obtain the recommended softness and hardness of each partition. S212. Based on the recommended softness and hardness of each partition and the preset mapping rules, the partition softness and hardness level corresponding to each partition is generated. S213. The partition softness and hardness level of each partition is output to the client.

[0044] Specifically, the preset correction algorithm is a set of compensatory calculation rules built into the model, based on human biomechanics and physiological characteristics. Its function is to correct the basic mattress firmness—a mechanical indicator derived from static geometric data such as volume and weight—by combining it with key parameters (such as lumbar curvature and pelvic tilt angle) into quantitative parameters that better meet the dynamic support and comfort needs of the target customer. The zoned recommended firmness is the final suggested firmness value obtained after correction by this algorithm for each preset human body zone (head, shoulders and back, waist, hips, and legs).

[0045] The "district hardness level" refers to the grade label corresponding to the recommended hardness of each zone; the "client" refers to the terminal device used by target customers or sales personnel for information interaction.

[0046] In practice, the mattress recommendation model calls upon more refined physiological characteristic data from the key parameters, such as lumbar curvature, pelvic tilt angle, estimated elasticity parameters of soft tissues, or pressure sensitivity coefficients of bony prominences within specific zones, and inputs them into a preset correction algorithm. This algorithm fine-tunes and calibrates the basic firmness of each zone based on predefined mathematical relationships or empirical formulas, thereby outputting a set of more accurate and personalized recommended firmness values ​​for each zone.

[0047] Subsequently, the mattress recommendation model forms a mapping based on the preset mapping rules. In this embodiment, the mapping is based on the softness and hardness calculation method in the national standard "GB / T43007-2023 Test and Evaluation Method for Mattress Hardness Grade Distribution". That is, a pre-established lookup table or threshold function that maps the continuous softness and hardness numerical range to a finite discrete level (such as "level 1", "level 2", "level 3", "level 4", "level 5", etc.) and classifies the recommended softness and hardness value of each zone to its corresponding zone softness and hardness level.

[0048] Finally, the system encapsulates the generated results—a series of paired data for "zone area - firmness level"—in a structured manner and transmits them to the client via the application programming interface (API). Upon receiving this data, the client software or interface can display it to the user in the form of a graphical zoning diagram, table, or configuration parameter list, or send it directly to the adjustable mattress's control system, thus completing the entire process from data processing to end-user application.

[0049] In this embodiment, the basic hardness of each zone can be accurately calculated by correcting the basic hardness of the zones, and the recommended hardness level of each zone can be matched based on the recommended hardness of the zones. The hardness level of each zone can be displayed to the client, which can realize the customization needs of different zones. Customers can customize mattresses according to the hardness level corresponding to each zone, so as to provide each customer with a "tailor-made" customization solution, improve the user experience and brand professional image.

[0050] In one embodiment, as shown in FIG5, the key parameter includes lumbar curvature, the partition includes the lumbar region, and step S211 includes steps S2111-S2113.

[0051] S2111. Determine the absolute value of the difference between the lumbar curvature and the standard lumbar curvature and the preset first lumbar threshold and the preset second lumbar threshold; S2112. If the absolute value of the difference is greater than the preset first lumbar threshold, increase the support stiffness by a preset first percentage on the basic stiffness of the lumbar region to obtain the recommended stiffness of the lumbar region; S2113. If the absolute value of the difference is less than the preset second lumbar threshold, decrease the support stiffness by a preset second percentage on the basic stiffness of the lumbar region to obtain the recommended stiffness of the lumbar region.

[0052] Specifically, the key parameters include lumbar curvature, which refers to a quantitative value characterizing the degree of physiological curvature of the lumbar spine, measured through a three-dimensional human body model, such as the lumbar lordosis angle or radius of curvature. The standard lumbar curvature refers to the ideal lumbar curvature reference value set according to ergonomic and clinical medical consensus. The preset first lumbar threshold and the preset second lumbar threshold are two pre-set boundary values ​​for judging the degree of curvature abnormality. The preset first lumbar threshold is usually the critical point for excessive curvature, and the preset second lumbar threshold is the critical point for insufficient curvature. The preset first percentage and the preset second percentage are pre-determined support stiffness adjustment ratios corresponding to different degrees of abnormality.

[0053] In practice, the mattress recommendation model first calculates the difference between the user's actual lumbar curvature obtained from key parameters and the standard lumbar curvature. Then, the absolute value of this difference is compared with preset first and second lumbar vertebral thresholds.

[0054] If the absolute value of the difference is greater than the preset first lumbar vertebra threshold, the user's lumbar curvature is determined to be significantly greater than the standard value (i.e., excessive lumbar lordosis). In this case, the system will add a preset first percentage of support hardness to the already calculated basic hardness values ​​for the lumbar region, for example, increasing the hardness by 15%, to calculate the recommended hardness for the lumbar region. Conversely, if the absolute value of the difference is less than the preset second lumbar vertebra threshold, the user's lumbar curvature is determined to be significantly less than the standard value (i.e., straightening of the lumbar curvature). In this case, the system will reduce the support hardness by a preset second percentage of the basic hardness for the lumbar region, for example, reducing the hardness by 10%, to generate the recommended hardness for the lumbar region.

[0055] This embodiment achieves personalized calibration of general mechanical calculation results based on precise physiological structural deviations, ensuring that the corrected lumbar region zone-recommended stiffness can accurately match the customer's lumbar health needs.

[0056] In one embodiment, as shown in FIG6, the key parameters include the pelvic tilt angle, the partition includes the hip, and step S211 includes steps S2114-S2115.

[0057] S2114. Determine the magnitude of the pelvic tilt angle compared to the preset angle threshold; S2115. If the pelvic tilt angle is greater than the preset angle threshold, reduce the support hardness by a preset third percentage on the basic softness and hardness of the hip area to obtain the recommended softness and hardness of the hip area.

[0058] Specifically, the pelvic tilt angle refers to the angle value calculated by analyzing the posture of a three-dimensional human body mesh model in the sagittal plane, used to quantify the degree of forward or backward rotation of the pelvis, usually with forward tilt as positive. The preset angle threshold is a pre-set critical angle value used to determine whether the pelvic tilt has reached a level requiring targeted intervention. The preset third percentage is a pre-determined reduction ratio of support stiffness corresponding to situations where the pelvic tilt angle exceeds the threshold.

[0059] In practice, the mattress recommendation model first obtains the specific value of the user's pelvic tilt angle from the key parameters. Then, it executes a judgment logic, directly comparing the pelvic tilt angle value with a preset angle threshold. If the comparison result shows that the pelvic tilt angle is greater than the preset angle threshold, it is determined that the user has significant anterior pelvic tilt. In this case, for the preset human body zone of the buttocks, the system will initiate a correction calculation: using the already calculated basic firmness of this zone as a baseline value, a firmness value equivalent to a preset third percentage is reduced. For example, if the preset third percentage is 10%, then the basic firmness of the buttocks zone is multiplied by (1-10%). After this multiplication operation or equivalent numerical reduction operation, the result is determined as the final recommended firmness for the buttocks zone.

[0060] This correction logic is based on the biomechanical principle that anterior pelvic tilt alters the stress characteristics of the buttocks in a lying position. It aims to optimize pressure distribution and spinal alignment by appropriately reducing the local support stiffness, thereby ensuring that the corrected buttock stiffness accurately matches the client's pelvic position and guarantees the comfort and scientific nature of the support.

[0061] In one embodiment, as shown in FIG7, step S20 includes steps S24-S25.

[0062] S24. Calculate the mattress firmness level corresponding to the mattress firmness according to the preset mapping rule; S25. Match the mattress with the closest firmness in the preset mattress firmness information database as the target mattress according to the mattress firmness.

[0063] Specifically, the mattress recommendation model determines which preset firmness range the mattress falls into by searching or calculating, and outputs the label corresponding to that range as the final mattress firmness level.

[0064] Subsequently, the mattress recommendation model uses the mattress firmness as a query condition and matches it against a pre-set mattress firmness information database. The core of the matching process lies in calculating the difference (e.g., absolute difference) between the firmness attribute value of each mattress in the database and the target "mattress firmness" value, and then sorting them according to the magnitude of the difference. The mattress recommendation model selects one or more mattresses with the smallest difference, i.e., the closest firmness, and identifies them as the target mattress.

[0065] In this embodiment, the entire process relies on precise quantitative comparison and standardized rules to ensure accurate classification of mattress firmness and scientific matching of target mattresses, providing customers with accurate and intuitive mattress recommendation results.

[0066] Furthermore, to optimize the results, this process can often be combined with other mattress attributes (such as material and zoning technology) in the database and user preference data for secondary filtering and sorting, so as to ensure that the final recommended target mattress not only matches the firmness but also meets the user's expectations in terms of overall characteristics.

[0067] S30. Input the mattress firmness into the pillow recommendation model. The pillow recommendation model calculates the pillow support height based on the key parameters, the preference data and the mattress firmness, and matches the target pillow in the preset pillow database based on the key parameters, the pillow support height and the preset pillow material indentation information.

[0068] In this embodiment, the pillow support height is a key quantitative recommendation value output by the pillow recommendation model. Specifically, it refers to the expected pillow height under natural pressure, recommended to achieve ideal head and neck support and spinal alignment after considering mattress firmness, user physiological characteristics, and preferences. The pillow material indentation information is a dataset pre-stored in the system describing the height compression ratio of different pillow filling materials (such as memory foam, latex, down, etc.) under typical head pressure. The preset pillow database is a structured information library containing detailed parameters of various pillow products. Each record includes at least the pillow's identification information, material composition, nominal height, firmness level, and corresponding indentation characteristic parameters.

[0069] In practice, the system calculates the mattress firmness, along with the key parameters and preference data, and inputs them into a preset pillow recommendation model. This model first calculates and generates a personalized pillow support height recommendation based on the key parameters (neck curvature height, shoulder width) and preference data (sleeping posture), combined with the mattress firmness.

[0070] Then, based on key parameters, pillow support height, and preset pillow material indentation information, the system queries and matches the preset pillow database. By comparing the attribute parameters of each pillow product in the database, it selects one or more products that best match the calculated suggestions in terms of support height, material characteristics, and softness / firmness, and determines them as the recommended target pillow.

[0071] This embodiment shifts from passive, subjective feelings to active, scientific assessments based on human morphology and biomechanics, matching the firmness of pillows and mattresses to recommend the most suitable pillow for the user's health needs, resulting in more convincing and accurate recommendations.

[0072] In one embodiment, as shown in FIG8, step S30 may include steps S31-S32.

[0073] S31. Calculate the basic support height of the pillow based on the key parameters and the preference data; S32. Compensate the basic support height of the pillow based on the firmness of the mattress to obtain the pillow support height.

[0074] Specifically, the basic pillow support height refers to the suggested height initially calculated by the pillow recommendation model based on user physiological characteristics and subjective preferences, without considering mattress characteristics, to achieve ideal head and neck support. The compensation here specifically refers to the numerical adjustment of the basic pillow support height according to the mattress firmness, based on the interaction between the mattress and pillow as a collaborative support system.

[0075] In practice, the pillow recommendation model first integrates physiological characteristic data related to the weight of the neck, shoulders, and head, as well as customer preference data from key parameters. Through built-in algorithms, it analyzes physiological adaptation needs and subjective preference tendencies to accurately calculate the basic support height of the pillow.

[0076] Subsequently, the model incorporates mattress firmness parameters for systematic compensation calculations. The model has a pre-defined compensation function describing the "mattress-pillow" support coupling relationship. Its core logic is: when the mattress firmness value indicates an overall soft mattress, the body sinks deeper into the mattress during sleep, resulting in a lower relative height between the shoulders and the bed surface. To maintain the natural curvature of the cervical spine, the pillow height needs to be increased accordingly to compensate. Conversely, when the mattress is firm, the pillow height needs to be appropriately reduced. The model inputs the mattress firmness value into this compensation function, calculates a height compensation amount (which can be positive or negative), and algebraically adds it to the basic pillow support height to obtain a more scientifically accurate recommended pillow support height that balances the mattress's influence.

[0077] This embodiment calculates the pillow support height that is suitable for the target customer's physiological characteristics, subjective preferences, and mattress support status by using key parameters, preference data, and mattress firmness, providing accurate parameter support for subsequent target pillow matching.

[0078] For example, the formula for calculating the basic support height of the pillow in this embodiment includes: if the sleeping posture is supine, then the basic support height of the pillow = neck curvature height + head weight compensation; if the sleeping posture is side-lying, then the basic support height of the pillow = (left shoulder width + right shoulder width) / 2 * adjustment coefficient + neck curvature height; if the sleeping posture is prone, then the basic support height of the pillow = neck curvature height * reference coefficient, and the minimum height is 2cm.

[0079] Furthermore, the logic for systematically calculating pillow support height based on mattress firmness is as follows: first, calculate the difference between the recommended mattress firmness and the standard mattress firmness, and compensate 1cm for every 10 units of firmness difference to obtain the compensation amount; the final recommended pillow support height = basic pillow support height + compensation amount.

[0080] In one embodiment, as shown in FIG9, the key parameter includes neck curvature height, and step S30 may include steps S33-S35.

[0081] S33. Calculate the actual height of each pillow material based on the pillow support height and the indentation information; S34. Calculate the absolute value of the difference between the actual height of each pillow and the neck curve height; S35. Select the pillow with the smallest absolute value of the difference, match the corresponding pillow material in the preset pillow database, and select the pillow with the closest actual height under that material as the target pillow.

[0082] Specifically, the cervical curvature height is a key anatomical parameter extracted from the client's three-dimensional human body model. It specifically refers to the vertical measurement value of the filling space required to maintain the natural physiological forward curvature of the cervical spine and is the core basis for determining the pillow's fit. The actual pillow height is derived from the recommended pillow support height (i.e., the desired effective support height) as the target, based on the pre-stored indentation information (i.e., specific data describing the height compression ratio of the material when subjected to standard head pressure), and through a reverse calculation formula. This is the initial nominal height that the pillow of that material should have in the uncompressed state.

[0083] In practice, the pillow recommendation model first uses the pillow support height and the indentation information to perform parallel calculations to calculate the corresponding actual pillow height for each material (such as memory foam, latex, down, polyester fiber, etc.) recorded in the database.

[0084] Next, the system proceeds to the matching and quantification evaluation phase: the calculated actual height of each pillow is compared with the user's neck curvature height, and the absolute value of the difference between the two is calculated. This absolute value of the difference directly represents the degree of conformity between the theoretical support surface of the pillow and the natural curve space of the user's neck, serving as an objective indicator of the quality of the match. The system iterates through the comparisons, selecting the calculation result with the smallest absolute value of the difference. This result simultaneously indicates the optimal matching material and the theoretically optimal actual pillow height for that material.

[0085] Ultimately, the system uses this optimal material as the key screening criterion, searches all pillow products of this material in the preset pillow database, selects the pillow of this material with different support heights that is closest to the actual height of the aforementioned optimal pillow, and determines it as the final target pillow. This completes the accurate and personalized recommendation from physiological parameters to specific products, ensuring that the pillow support is accurately matched with the neck curvature height, and guaranteeing the scientific and comfortable nature of neck support.

[0086] In one embodiment, as shown in FIG10, the preference data includes sleep posture, which includes lying on one's side or lying on one's back, and steps S311-S314 may be included after step S31.

[0087] S311. Determine whether the target customer's sleeping posture is side-lying or supine; S312. If side-lying, calculate the first zone pillow support height of each area in the preset pillow partition according to the preset first calculation rule and the basic pillow support height; S313. If supine, calculate the second zone pillow support height of each area in the preset pillow partition according to the preset second calculation rule and the basic pillow support height; S314. Output the first zone pillow support height or the second zone pillow support height corresponding to each area to the client.

[0088] Specifically, the sleep posture is a classification information representing the user's habitual sleeping posture, obtained from user-provided preference data. In this embodiment, it specifically refers to the two main categories of "side-lying" and "supine". The basic pillow support height is an initial value of overall support height calculated by the model based on key physiological parameters and preference data such as the user's neck curvature height, shoulder width, and head weight. The preset pillow zones refer to several functional areas pre-divided on the pillow geometry model according to ergonomics and head, neck, and shoulder contact mechanics, such as the left wing zone, central zone, right wing zone, and neck support zone. The first calculation rule and the second calculation rule are two sets of height allocation algorithm logics preset for side-lying and supine postures, respectively. The first zone pillow support height refers to the appropriate support height of each pillow zone when side-lying; the second zone pillow support height refers to the appropriate support height of each pillow zone when supine.

[0089] In practice, after calculating the basic pillow support height as described above, the subsequent process is advanced based on the pillow recommendation model. The model first extracts sleep posture information from the preference data to determine whether the target customer's sleep posture is side-lying or supine. If side-lying, it calls a preset first calculation rule, combining the obtained basic pillow support height with the functional positioning of preset pillow zones, to calculate the first-zone pillow support height for each zone. If supine, it calls a preset second calculation rule, calculating the second-zone pillow support height for each zone based on the basic pillow support height. After completing the corresponding zone support height calculation, the model synchronously outputs the first-zone or second-zone pillow support height for each area to the client, providing the customer with a clear view of the pillow zone support requirements adapted to their sleeping posture, and also providing precise zone parameter support for subsequent customized pillow needs.

[0090] This embodiment refines the initially calculated basic pillow support height into smaller zones based on key user sleeping posture preference data, and outputs personalized support parameters for each zone. This allows for customized services to be provided to target customers, enhances the brand's professional image, and provides a data foundation for customized pillow production.

[0091] In one embodiment, as shown in FIG11, the preset pillow partitions include a left wing area, a central area, a right wing area, and a neck support area, and step S312 may include steps S3121-S3123.

[0092] S3121. Calculate the first zone pillow support height of the left wing area and the right wing area according to the preset first parameter and the basic pillow support height; S3122. Calculate the first zone pillow support height of the central area according to the preset second parameter and the basic pillow support height; S3123. Calculate the first zone pillow support height of the neck support area according to the preset third parameter and the basic pillow support height; wherein, the preset first parameter is greater than the preset third parameter, and the preset third parameter is greater than the preset second parameter.

[0093] Specifically, the preset pillow zones refer to the functional area divisions predefined on the pillow structure to meet the differentiated support needs of the head, neck, and shoulders under different sleeping positions. In this embodiment, they specifically include the left and right wing zones for compensating for shoulder height when lying on one's side, the central zone for supporting the head, and the neck support zone specifically for supporting the physiological curvature of the cervical spine. The preset first parameter, preset second parameter, and preset third parameter are proportional coefficients or weighting factors pre-set in the first calculation rule to convert the baseline value of the pillow's basic support height into the specific height of each zone.

[0094] In practice, after determining that the user is lying on their side, the pillow recommendation model invokes a preset first calculation rule. The core logic of this rule is based on the fact that in a side-lying position, there is a height difference between the human shoulder and the bed surface, requiring the edge area of ​​the pillow to provide additional support to maintain the horizontal alignment of the spine; at the same time, the head needs to be nestled in the center of the pillow to avoid lateral flexion, while the neck needs independent and moderate support to maintain its natural forward curvature. The rule reflects these differentiated needs through three different preset parameters.

[0095] First, the pillow recommendation model multiplies the basic pillow support height by a larger preset first parameter to calculate the required first-zone pillow support height for the left and right wing areas. Setting a larger first parameter ensures that the wing areas receive significantly higher support than the basic height, effectively filling the shoulder space. Second, the system multiplies the basic pillow support height by the smallest preset second parameter to calculate the first-zone pillow support height for the central area, making this area relatively low to provide more space for the head. Finally, the system multiplies the basic pillow support height by a preset third parameter with a value between the first and second parameters to calculate the first-zone pillow support height for the neck support area, providing moderate elevation to conform to the neck curve. The relative values ​​of the three parameters (preset first parameter > preset third parameter > preset second parameter) ensure that the final generated zone height distribution conforms to the biomechanical requirements of side-lying.

[0096] Therefore, this embodiment ensures that the left and right wing areas have the highest support height, the neck support area is next, and the central area is the lowest, based on the relationship between parameters. This adapts to the physiological force posture of the head and neck when lying on one's side, providing precise parameter support for the subsequent customized pillow zoning.

[0097] In one embodiment, as shown in FIG12, the preset pillow partitions include a left wing area, a central area, a right wing area, and a neck support area, and step S313 may include steps S3131-S3132.

[0098] S3131. Calculate the second zone pillow support height of the neck support area according to the preset fourth parameter and the basic pillow support height; S3132. The second zone pillow support height of the left wing area, the central area and the right wing area is the basic pillow support height.

[0099] Specifically, the preset fourth parameter is a proportional coefficient or adjustment factor pre-set in the second calculation rule, specifically used to calculate the height of the neck support zone in the supine position. Its value is determined based on biomechanical studies of the amount of cervical spine elevation required when supine. The second zone pillow support height is the specific height value calculated for each zone after applying the second calculation rule.

[0100] In practice, after the pillow recommendation model determines that the user's sleeping posture is supine based on preference data, it immediately invokes a preset second calculation rule. The design logic of this rule stems from the fact that in the supine position, the contact area between the human head and the bed surface is relatively large, and the main requirement is to prevent the head from tilting back excessively and to maintain the natural forward curvature of the cervical spine. Therefore, the rule provides special treatment for the neck support area, while providing balanced support for other main head support areas.

[0101] Specifically, the pillow recommendation model first calculates the dedicated second-zone pillow support height for the neck support area by multiplying the basic pillow support height by a preset fourth parameter. This parameter is typically set to a value greater than 1, aiming to elevate the neck support area slightly above the basic height to effectively fill the space below the cervical lordosis and provide support. Simultaneously, for the left wing, central, and right wing areas, a second calculation rule stipulates that their second-zone pillow support height is directly equal to the basic pillow support height, without further differentiation. This means that in a supine position, the model assumes these three main contact areas of the head should receive a uniform support height to maintain a neutral head position and smoothly transition with the specifically elevated neck support area, forming an overall support surface that conforms to supine biomechanical requirements. Finally, the system outputs these four zones and their calculated second-zone support heights as the results.

[0102] Therefore, this embodiment is adapted to the physiological force posture of the head and neck when lying supine, providing precise parameter support for the subsequent customized pillow partitioning.

[0103] S40. Send the mattress firmness rating, the target mattress, the pillow support height, and the target pillow to the client.

[0104] Specifically, after completing all calculations and matching logic, the recommendation engine in the system's backend encapsulates the four core result elements—mattress firmness rating, target mattress (including model, image, key parameters, and other detailed information), pillow support height, and target pillow (including model, image, material, and other detailed information)—into a structured data packet. The system then transmits this encapsulated data packet from the server to the client via a predefined application programming interface (API) or network communication protocol. Upon receiving this data packet, the client application calls the corresponding data parsing module to parse and extract the content, and finally renders and displays the recommendation results to the user in a clear and easy-to-read format combining text and graphics through a graphical user interface (GUI). For example, a dedicated recommendation results page might present suggested mattress and pillow models, key parameters, and visual illustrations, thus completing the closed loop from system calculation to user perception.

[0105] In one embodiment, as shown in FIG13, steps S41-S44 may be included before step S40.

[0106] S41. Input the key parameters, mattress firmness, and pillow support height into a preset matching model; S42. The matching model verifies the support effect of the mattress firmness and pillow support height based on the key parameters, and determines whether the support effect of the mattress firmness and pillow support height is qualified; S43. If the support effect is qualified, send the mattress firmness level, the target mattress, the pillow support height, and the target pillow to the client; S44. If the support effect is unqualified, return to the mattress recommendation model and the pillow recommendation model to recalculate the mattress firmness level, the target mattress, the pillow support height, and the target pillow.

[0107] Specifically, the matching model is a pre-defined computational model built upon principles of human biomechanics, spinal health medicine, and pressure distribution. Its core function is to simulate and evaluate the effects of initially recommended mattress and pillow parameter combinations. The support effect verification refers to the process by which the matching model, using the input original user key parameters (such as volume and weight, spinal curvature, etc.), combined with the mattress firmness and pillow support height initially recommended by the aforementioned mattress and pillow recommendation models, constructs a human-bedding interaction scenario in a virtual environment and performs quantitative analysis. A satisfactory support effect means that the comprehensive score obtained from the verification analysis reaches or exceeds a preset quality threshold, indicating that the recommended solution can provide support that meets healthy sleep standards for the user.

[0108] In practice, after the mattress recommendation model outputs the mattress firmness rating and target mattress, and the pillow recommendation model outputs the pillow support height and target pillow, the system does not immediately send these results. Instead, it first inputs the original key parameters, along with the mattress firmness (specific values) and pillow support height calculated by the two models, into a preset matching model. This matching model reconstructs the user's personalized physiological model based on the key parameters and uses the recommended mattress firmness and pillow height as boundary conditions to perform multi-angle biomechanical simulation and adaptive calculations.

[0109] The calculation process mainly includes four aspects: First, calculating the mattress firmness adaptability score. The model compares the input mattress firmness value with the ideal firmness range calculated based on key parameters such as user weight, body shape, and spinal curvature, while also considering subjective firmness preferences in the data. A quantitative score between 0 and 100 is generated through a preset algorithm. Second, calculating the pillow height adaptability score. The model compares the recommended pillow support height with the theoretical ideal height value derived from cervical curvature height, shoulder width, and sleeping posture, and considers the user's height preference to generate another quantitative score within the range of 0 to 100. Third, calculating and analyzing spinal alignment. The model simulates the deviation of the curves of the cervical, thoracic, and lumbar vertebrae from the ideal physiological curves when the user is in a typical sleeping posture under the recommended configuration, and outputs the alignment score of each segment of the spine. Fourth, simulating pressure distribution. The model calculates the magnitude and uniformity of pressure between the main weight-bearing parts of the body and the contact surface of the bedding.

[0110] The matching model integrates the four scores mentioned above into a comprehensive support effect score according to preset weights. This comprehensive score is compared with a preset pass / fail threshold to determine whether the support effect of the mattress firmness and the pillow support height is satisfactory. If the comprehensive score is greater than or equal to the threshold, it is deemed satisfactory, and the system then allows the complete recommendation results (the mattress firmness level, the target mattress, the pillow support height, and the target pillow) to be packaged and sent to the client.

[0111] If the overall score is below the threshold, it is deemed unqualified. The system will return the specific defects identified in this verification analysis (such as a low mattress firmness adaptability score or an insufficient pillow height adaptability score) as explicit feedback to the mattress recommendation model and the pillow recommendation model. Based on this specific score feedback, the two recommendation models will adjust their internal calculation parameters or matching strategies, re-execute the recommendation calculation and product matching process, generate new recommendation parameters and product lists, and resubmit them to the matching model for verification.

[0112] Furthermore, the system can accumulate user data to inform research and development, helping mattress companies understand the body characteristics and support needs of different groups of people, which can then be used to optimize existing products and develop new ones.

[0113] This iterative cycle will continue until a satisfactory support effect is achieved or the maximum number of iterations set by the system is reached. This ensures that the final recommended solution delivered to the client is an optimized result that has been verified from multiple dimensions and meets the individual health needs of the user, thereby improving the scientific, accurate and intelligent level of the system's recommended mattresses and pillows.

[0114] This application utilizes 3D structured light technology to scan and acquire three-dimensional images and preference data of target customers, extracts their key body parameters, and inputs these parameters and preference data into a mattress recommendation model and a pillow recommendation model for collaborative calculation. This ensures that the calculated mattress firmness matches the target customer's actual body shape and preference needs, and that the pillow support height is adapted to the recommended mattress firmness and the target customer's actual body shape and preference needs. This achieves precise and coordinated matching between the mattress and pillow, enabling automated, high-precision, and high-efficiency acquisition of the user's human morphological characteristics. Based on ergonomic principles, it scientifically recommends the most suitable mattress and pillow for the target customer's health needs, solving the current problems of inaccurate, unscientific, and inefficient mattress and pillow recommendations.

[0115] Meanwhile, this application's recommendation method shifts from passive subjective feelings to proactive, scientific assessment based on human morphology and biomechanics, matching the firmness of pillows and mattresses for more convincing and accurate recommendations, thus enhancing the scientific rigor of the method. Furthermore, using 3D structured light technology, a scan of just a few seconds can replace a lengthy trial lying process, revealing spinal posture issues that are difficult to observe with the naked eye, achieving precise matching and high efficiency. Moreover, this method not only considers the target customer's height and weight but also delves into details such as spinal curvature and hip curves, providing a "tailor-made" solution for each user and offering a data foundation for customized mattress and pillow production. In terms of customer experience, the entire recommendation process is technologically advanced, intuitively interactive, and produces professional reports, significantly enhancing the customer shopping experience and brand image.

[0116] Figure 14 is a schematic block diagram of a mattress and pillow recommendation device 300 based on 3D structured light according to an embodiment of the present invention. As shown in Figure 14, corresponding to the above-described mattress and pillow recommendation method based on 3D structured light, the present invention also provides a mattress and pillow recommendation device 300 based on 3D structured light. This mattress and pillow recommendation device 300 includes units for executing the above-described mattress and pillow recommendation method based on 3D structured light, and the device can be configured in a computer device. Specifically, referring to Figure 14, the mattress and pillow recommendation device 300 based on 3D structured light includes an extraction unit 301, a mattress recommendation unit 302, a pillow recommendation unit 303, and a sending unit 304.

[0117] Extraction unit 301 is used to acquire image data and preference data of the target customer, extract key parameters from the image data, and input the key parameters and preference data into a preset mattress recommendation model and pillow recommendation model; Mattress recommendation unit 302 is used for the mattress recommendation model to calculate mattress firmness based on the key parameters and preference data, and to calculate and match mattress firmness level and target mattress based on mattress firmness, preset mapping rules, and preset mattress firmness information database; Pillow recommendation unit 303 is used to input the mattress firmness into the pillow recommendation model, the pillow recommendation model calculates pillow support height based on the key parameters, preference data, and mattress firmness, and matches target pillow in a preset pillow database based on the key parameters, pillow support height, and preset pillow material indentation information; Sending unit 304 is used to send the mattress firmness level, target mattress, pillow support height, and target pillow to the client.

[0118] In one embodiment, the mattress recommendation unit 302 includes a first calculation unit, a correction unit, a generation unit, a first output unit, and a first matching unit.

[0119] A first calculation unit is used to calculate the basic firmness of each zone based on the volume data and weight data of each zone; calculate the overall mattress baseline firmness based on the basic firmness and a preset calculation algorithm; and calculate the mattress firmness level corresponding to the firmness according to the preset mapping rules. A correction unit is used to correct the mattress baseline firmness based on the preference data to obtain the mattress firmness; and to correct the basic firmness of each zone based on the key parameters and a preset correction algorithm to obtain the recommended firmness for each zone. A generation unit is used to generate the corresponding firmness level for each zone based on the recommended firmness and a preset mapping rule. A first output unit is used to output the firmness level of each zone to the client. A first matching unit is used to match the mattress with the closest firmness in the preset mattress firmness information database as the target mattress.

[0120] In one embodiment, the correction unit includes a first judgment unit and a first correction unit.

[0121] The first judgment unit is used to determine the magnitude of the absolute value of the difference between the lumbar curvature and the standard lumbar curvature and the preset first lumbar threshold and the preset second lumbar threshold; and to determine the magnitude of the pelvic tilt angle and the preset angle threshold. The first correction unit is used to, if the absolute value of the difference is greater than the preset first lumbar threshold, increase the support hardness of the lumbar region by a preset first percentage to obtain the recommended hardness of the lumbar region; if the absolute value of the difference is less than the preset second lumbar threshold, decrease the support hardness of the lumbar region by a preset second percentage to obtain the recommended hardness of the lumbar region; and if the pelvic tilt angle is greater than the preset angle threshold, decrease the support hardness of the buttock region by a preset third percentage to obtain the recommended hardness of the buttock region.

[0122] In one embodiment, the pillow recommendation unit 303 includes a second calculation unit, a compensation unit, a second matching unit, a second judgment unit, a partition calculation unit, and a second output unit.

[0123] The second calculation unit is used to calculate the basic support height of the pillow based on the key parameters and the preference data; the compensation unit is used to compensate the basic support height of the pillow based on the firmness of the mattress to obtain the pillow support height; the second matching unit is used to calculate the actual pillow height corresponding to each pillow material based on the pillow support height and the indentation information; calculate the absolute value of the difference between each actual pillow height and the neck curve height; select the actual pillow height with the smallest absolute value of the difference to match the corresponding pillow material in the preset pillow database and the pillow with the closest actual pillow height under that material as the target pillow; the second judgment unit is used to determine whether the target customer's sleeping posture is side-lying or supine; the partition calculation unit calculates the first partition pillow support height of each area in the preset pillow partition according to the preset first calculation rule and the basic support height of the pillow if the customer is side-lying; if the customer is supine, calculates the second partition pillow support height of each area in the preset pillow partition according to the preset second calculation rule and the basic support height of the pillow; the second output unit outputs the first partition pillow support height or the second partition pillow support height corresponding to each area to the client.

[0124] In one embodiment, the partitioning calculation unit includes a third calculation unit.

[0125] The third calculation unit is used to calculate the first zone pillow support height of the left wing area and the right wing area according to the preset first parameter and the basic pillow support height; calculate the first zone pillow support height of the central area according to the preset second parameter and the basic pillow support height; calculate the first zone pillow support height of the neck support area according to the preset third parameter and the basic pillow support height; wherein the preset first parameter is greater than the preset third parameter, and the preset third parameter is greater than the preset second parameter; and is used to calculate the second zone pillow support height of the neck support area according to the preset fourth parameter and the basic pillow support height; the second zone pillow support height of the left wing area, the central area, and the right wing area is the basic pillow support height.

[0126] In one embodiment, the 3D structured light-based mattress and pillow recommendation device 300 also includes a verification model.

[0127] A verification model is used to input the key parameters, mattress firmness, and pillow support height into a preset matching model. The matching model verifies the support effect of the mattress firmness and pillow support height based on the key parameters, and determines whether the support effect of the mattress firmness and pillow support height is qualified. If the support effect is qualified, the mattress firmness level, the target mattress, the pillow support height, and the target pillow are sent to the client. If the support effect is unqualified, the mattress recommendation model and the pillow recommendation model are returned to recalculate the mattress firmness level, the target mattress, the pillow support height, and the target pillow.

[0128] In one embodiment, the extraction unit 301 includes a preprocessing unit and a first extraction unit.

[0129] The preprocessing unit is used to preprocess the image data to generate a three-dimensional human body mesh model; the first extraction unit is used to extract parameters from the three-dimensional human body mesh model to obtain the key parameters.

[0130] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned mattress and pillow recommendation device based on 3D structured light and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0131] The aforementioned mattress and pillow recommendation device 300 based on 3D structured light can be implemented as a computer program that can run on the computer device shown in Figure 15.

[0132] Please refer to Figure 15, which is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a terminal or a server. The terminal can be an electronic device with communication functions, such as a smartphone, tablet, laptop, desktop computer, personal digital assistant, or wearable device. The server can be a standalone server or a server cluster composed of multiple servers.

[0133] Referring to Figure 15, the computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0134] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a mattress and pillow recommendation method based on 3D structured light.

[0135] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0136] The internal memory 504 provides an environment for the execution of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a mattress and pillow recommendation method based on 3D structured light.

[0137] The network interface 505 is used for network communication with other devices. Those skilled in the art will understand that the structure shown in FIG15 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. A specific computer device 500 may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0138] The processor 502 is used to run a computer program 5032 stored in a memory to implement the steps of the above-described mattress and pillow recommendation method based on 3D structured light.

[0139] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0140] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0141] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program includes program instructions. When executed by a processor, the program instructions cause the processor to perform the steps of the above-described method for recommending mattresses and pillows based on 3D structured light.

[0142] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0143] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software 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 implementations should not be considered beyond the scope of this invention.

[0144] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0145] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the device of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0147] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for recommending mattresses and pillows based on 3D structured light, characterized in that, The method includes: acquiring image data and preference data of the target customer; extracting key parameters from the image data; and inputting the key parameters and preference data into a preset mattress recommendation model and a pillow recommendation model; the mattress recommendation model calculates mattress firmness based on the key parameters and preference data, and calculates and matches a mattress firmness level and a target mattress based on the mattress firmness, preset mapping rules, and a preset mattress firmness information database; the mattress firmness is input into the pillow recommendation model, which calculates pillow support height based on the key parameters, preference data, and mattress firmness, and matches a target pillow in a preset pillow database based on the key parameters, pillow support height, and preset pillow material indentation information; and the mattress firmness level, the target mattress, the pillow support height, and the target pillow are sent to the client.

2. The method according to claim 1, characterized in that, The key parameters include volume and weight data of each part in the preset human body partitions. The mattress recommendation model calculates the mattress firmness based on the key parameters and the preference data by: calculating the basic firmness of each partition according to the volume and weight data of each part; calculating the overall mattress reference firmness based on the basic firmness of the partitions and the preset calculation algorithm; and correcting the mattress reference firmness according to the preference data to obtain the mattress firmness.

3. The method according to claim 2, characterized in that, The step of calculating the basic hardness of each partition based on the volume data and weight data of each partition includes: correcting the basic hardness of each partition based on the key parameters and a preset correction algorithm to obtain the recommended hardness of each partition; generating the partition hardness level corresponding to each partition based on the recommended hardness of each partition and a preset mapping rule; and outputting the partition hardness level of each partition to the client.

4. The method according to claim 3, characterized in that, The key parameters include lumbar curvature, and the zones include the lumbar region. The step of correcting the basic hardness of each zone based on the key parameters and a preset correction algorithm to obtain the recommended hardness of each zone includes: determining the absolute value of the difference between the lumbar curvature and the standard lumbar curvature and the magnitude of a preset first lumbar threshold and a preset second lumbar threshold; if the absolute value of the difference is greater than the preset first lumbar threshold, then a preset first percentage of support hardness is added to the basic hardness of the lumbar region to obtain the recommended hardness of the lumbar region; if the absolute value of the difference is less than the preset second lumbar threshold, then a preset second percentage of support hardness is reduced to the basic hardness of the lumbar region to obtain the recommended hardness of the lumbar region.

5. The method according to claim 3, characterized in that, The key parameters include the pelvic tilt angle, and the partitions include the buttocks. The step of adjusting the basic softness and hardness of each partition according to the key parameters and the preset correction algorithm to obtain the recommended softness and hardness of each partition includes: determining the size of the pelvic tilt angle and the preset angle threshold; if the pelvic tilt angle is greater than the preset angle threshold, then reducing the support hardness by a preset third percentage on the basic softness and hardness of the buttocks partition to obtain the recommended softness and hardness of the buttocks partition.

6. The method according to claim 1, characterized in that, The steps of calculating and matching the mattress firmness level and target mattress based on the mattress firmness, preset mapping rules, and preset mattress firmness information database include: calculating the mattress firmness level corresponding to the mattress firmness according to the preset mapping rules; and matching the mattress with the closest firmness in the preset mattress firmness information database as the target mattress.

7. The method according to claim 1, characterized in that, The pillow recommendation model calculates the pillow support height based on the key parameters, the preference data, and the mattress firmness, including: calculating the basic pillow support height based on the key parameters and the preference data; and compensating the basic pillow support height based on the mattress firmness to obtain the final pillow support height.

8. The method according to claim 7, characterized in that, The key parameters include cervical curvature height. The step of matching the target pillow in a preset pillow database based on the key parameters, the pillow support height, and preset pillow material indentation information includes: calculating the actual pillow height corresponding to each pillow material according to the pillow support height and the indentation information; calculating the absolute value of the difference between each actual pillow height and the cervical curvature height; selecting the pillow with the smallest absolute value of the difference, matching the corresponding pillow material in the preset pillow database, and the pillow with the closest actual pillow height under that material as the target pillow.

9. The method according to claim 7, characterized in that, The preference data includes sleep posture, which includes side-lying or supine sleeping. After the step of calculating the basic pillow support height based on the key parameters and the preference data, the following steps are included: determining whether the target customer's sleep posture is side-lying or supine; if side-lying, calculating the first zone pillow support height of each area in the preset pillow partition according to the preset first calculation rule and the basic pillow support height; if supine, calculating the second zone pillow support height of each area in the preset pillow partition according to the preset second calculation rule and the basic pillow support height; and outputting the first zone pillow support height or the second zone pillow support height corresponding to each area to the client.

10. The method according to claim 9, characterized in that, The preset pillow partitions include a left wing area, a central area, a right wing area, and a neck support area. If the user is lying on their side, the step of calculating the first partition pillow support height of each area within the preset pillow partitions according to a preset first calculation rule and the basic pillow support height includes: calculating the first partition pillow support height of the left and right wing areas according to a preset first parameter and the basic pillow support height; calculating the first partition pillow support height of the central area according to a preset second parameter and the basic pillow support height; and calculating the first partition pillow support height of the neck support area according to a preset third parameter and the basic pillow support height. Wherein, the preset first parameter is greater than the preset third parameter, and the preset third parameter is greater than the preset second parameter.

11. The method according to claim 9, characterized in that, The preset pillow partitions include a left wing area, a central area, a right wing area, and a neck support area. If the patient is lying supine, the step of calculating the second partition pillow support height of each area in the preset pillow partitions according to the preset second calculation rule and the basic pillow support height includes: calculating the second partition pillow support height of the neck support area according to the preset fourth parameter and the basic pillow support height; the second partition pillow support heights of the left wing area, the central area, and the right wing area are the basic pillow support heights.

12. The method according to claim 1, characterized in that, Before the step of sending the mattress firmness rating, the target mattress, the pillow support height, and the target pillow to the client, the process includes: inputting the key parameters, the mattress firmness, and the pillow support height into a preset matching model; the matching model verifies the support effect of the mattress firmness and the pillow support height based on the key parameters, and determines whether the support effect of the mattress firmness and the pillow support height is qualified; if the support effect is qualified, the mattress firmness rating, the target mattress, the pillow support height, and the target pillow are sent to the client; if the support effect is unqualified, the mattress recommendation model and the pillow recommendation model are returned to recalculate the mattress firmness rating, the target mattress, the pillow support height, and the target pillow.

13. The method according to claim 1, characterized in that, The steps of acquiring image data and preference data of target customers and extracting key parameters from the image data include: preprocessing the image data to generate a three-dimensional human body mesh model; and extracting parameters from the three-dimensional human body mesh model to obtain the key parameters.

14. A mattress and pillow recommendation device based on 3D structured light, characterized in that, Includes a unit for performing the method as described in any one of claims 1-13.

15. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in any one of claims 1-13.

16. A storage medium, characterized in that, The storage medium stores a computer program, which includes program instructions that, when executed by a processor, can implement the method as described in any one of claims 1-13.

Images