A virtual reality-based perfume blending method and system

Abstract

This invention discloses a perfume blending method and system based on virtual reality. The method includes: constructing a virtual lobby; presenting the virtual lobby to a user through a virtual reality device; presenting a corresponding themed scene to the user; releasing the scent of the corresponding scent component in response to the user's interaction with the scent component model; collecting the user's EEG data in real time using an EEG signal acquisition device, and simultaneously obtaining the user's subjective rating data for the scent component corresponding to the currently released scent; generating an emotional valence index for the corresponding scent component based on the collected EEG data, and calculating the comprehensive preference score corresponding to the scent component; generating a personalized perfume formula based on the comprehensive preference scores of each scent component; and generating a visual perfume card corresponding to the personalized perfume formula. This invention can generate personalized perfume formulas based on the user's interaction with the model in the virtual scene.

Description

Technical Field

[0001] This invention belongs to the field of perfume blending technology, and in particular relates to a perfume blending method and system based on virtual reality. Background Technology

[0002] With the upgrading of residents' consumption and the refined development of the beauty and personal care industry, perfume products have gradually extended from traditional standardized luxury goods to personalized and customized mass consumer products. Consumers' demand for perfume is no longer limited to branded products, but increasingly seeks exclusive fragrances that highly match their own emotions, preferences, and temperament. However, the core characteristics of perfume olfactory perception are strong subjectivity, transience, and easy fatigue, which have brought many long-standing industry pain points to perfume selection, recommendation, and personalization. At the same time, the development of digital technology has also provided new possibilities for technological innovation in this field.

[0003] In the traditional perfume selection and customization field, existing models have significant technical and experiential shortcomings. In the traditional offline fragrance testing model, consumers mainly rely on fragrance strips and the perfume bottle opening to experience the scent. First, a single visit to the store allows for the testing of only a limited number of scents, and repeated testing in a short period can easily lead to olfactory fatigue, causing a rapid decline in olfactory sensitivity and an inability to accurately distinguish subtle differences in fragrance details. Second, the complex ambient odors in offline stores cannot provide consumers with an undisturbed, pure olfactory experience, making it easy for consumers' olfactory impressions to be skewed by the environment, making it difficult to form accurate fragrance memories. Third, personalized customization has a very high barrier to entry. Ordinary consumers lack systematic perfumery expertise and cannot translate their vague olfactory preferences into the selection and proportioning of raw materials for top, middle, and base notes. Furthermore, one-on-one customization services by professional perfumers are costly, time-consuming, and prone to communication errors, making it difficult to achieve large-scale popularization. In the field of online intelligent recommendation, most existing AI-based perfume recommendation systems rely on purely subjective information such as user-filled preference questionnaires, historical purchase data, and social media tags to make recommendations. Their core flaw lies in the lack of real-time physical perception of the scent by the user. Users cannot truly experience the scent characteristics of the recommended fragrance before placing an order, resulting in a significant deviation between the recommendation results and the user's actual expectations. The return rate for online perfumes is much higher than that for other beauty products, making it impossible to achieve truly accurate and personalized recommendations.

[0004] In recent years, virtual reality (VR) technology, with its core advantage of immersive interaction, has been widely used in digital content, new retail, and other fields. Meanwhile, the development of olfactory digital twin technology has also enabled the digital control and precise release of scents. Related technologies have already introduced olfactory feedback into VR scenes for perfume testing and blending assistance. Among them, Chinese utility model patent CN215769666U discloses an immersive VR device for perfume blending. This device uses a VR image as its core, combined with a movable perfume box and a test tube rack containing perfume samples. It outputs virtual scene images through a lens-type eye mask, and synchronously releases perfume scents through scent diffusion channels on the perfume box. Simultaneously, it incorporates a sound unit to achieve basic fusion of sight, hearing, and smell, simulating real-world scenarios for virtual perfume testing. This helps consumers avoid sales pressure from offline guides, prevent impulsive purchases, and enhance the immersive experience of perfume testing to a certain extent.

[0005] However, the aforementioned existing technologies are merely hardware devices for immersive fragrance testing, which can only achieve the passive release of scents from pre-stored perfume samples. They have not built a systematic perfume blending interaction method, cannot achieve real-time interaction between users and fragrance components in virtual scenes, and cannot automatically generate personalized perfume formulas based on user experience feedback. They can only achieve the fragrance testing experience of finished perfumes and do not have the capability for customized perfume production. Summary of the Invention

[0006] The purpose of this invention is to provide a perfume blending method and system based on virtual reality, which partially solves or alleviates the above-mentioned shortcomings in the prior art and can generate personalized perfume formulas based on the interaction between the user and the model in the virtual scene.

[0007] To solve the aforementioned technical problems, the present invention specifically adopts the following technical solution: A first aspect of the present invention is to provide a perfume blending method based on virtual reality, comprising the following steps: A virtual lobby and multiple themed scene entrances are set up within the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type. A virtual lobby is presented to users through virtual reality devices, and corresponding fragrances are released in response to user interaction with the theme scene entrance for users to rate. Based on the highest-rated fragrance, the corresponding themed scenarios are presented to the user; in response to the user's interaction with the fragrance component model, the scent of the corresponding fragrance component is released. When the aroma components are released, the user's brainwave data is collected in real time through an EEG signal acquisition device, and the user's subjective rating data of the aroma components corresponding to the currently released aroma is obtained at the same time. Based on the collected EEG data, an emotional valence index for the corresponding aroma component is generated. The emotional valence index is then weighted and fused with the subjective rating data to calculate the comprehensive preference score corresponding to the aroma component. Based on the overall preference score of each aroma component, the preferred aroma components are selected from the preset aroma component library by category into top notes, middle notes and base notes. The blending ratio of each preferred aroma component is calculated based on the overall preference score to generate a personalized perfume formula. Based on the personalized perfume formula, a visual perfume card corresponding to the formula is generated.

[0008] Furthermore, the emotional valence index is calculated using the following formula: ; Calculate; where VEI(s) i ) for experiencing aroma components i Emotional valence index at the time; P α,F4 (s i For user experience, aroma components i At that time, the average power of the α band of the F4 channel in the frontal lobe of the EEG signal; P α,F3 (s i For user experience, aroma components i At that time, the average power of the α band of the F3 channel in the frontal lobe of the EEG signal.

[0009] Furthermore, the overall preference score for aroma components is calculated using the following formula: ; Calculate; where P(s) i Ln(s) represents the overall preference score for aroma component si. i ( ) represents the normalized subjective rating data, VEI norm (s i ω is the normalized emotional valence index. I ω is the weighting coefficient for subjective ratings. O Let be the weighting coefficients of the normalized sentiment valence index, and satisfy ω I +ω O= 1.

[0010] Furthermore, the weighting coefficient is set to ω by default. I =0.6、ω O =0.4, and can be dynamically adjusted according to application scenarios or user profiles.

[0011] Furthermore, the arousal index and attention index are introduced to expand and fuse multiple indicators. The expanded comprehensive preference score is calculated as follows: P ext (s i)=ω I *L n (s i )+ω O *VEI norm (s i )+ω A *A norm (s i )+ω T *T norm (s i ); Among them, P ext (s i ) represents the expanded comprehensive preference score, Ln(s) i ( ) represents the normalized subjective rating data, VEI norm (s i A is the normalized emotional valence index. norm (s i T represents the normalized arousal index. norm (s i ω represents the normalized attention index; I ω is the weighting coefficient for subjective ratings. O ω is the weighting coefficient of the normalized emotional valence index. A ω is the weighting coefficient of the wakefulness index. T Let be the weighting coefficient of the attention index, and satisfy ω I +ω O +ω A +ω T =1.

[0012] Furthermore, the steps for generating personalized fragrance formulations include: All aroma components in the preset aroma component library are divided into three categories according to their aroma attributes: top notes, middle notes, and base notes. For all aroma components experienced by users in each category, they are sorted from high to low according to their overall preference scores. Based on the preset formula complexity parameters, select the top-ranked aroma components with the highest overall preference scores from the ranking results of each type of aroma component as preferred aroma components. For each fragrance family, the volume ratio of the preferred fragrance component within the corresponding fragrance family is calculated based on the proportion of the overall preference score of each preferred fragrance component to the total overall preference score of all preferred fragrance components in that family. This process ultimately generates a complete personalized perfume formula.

[0013] Furthermore, the volume ratio of various aroma components within a certain chromatic layer is determined using the following formula: ; Calculate; where, This represents the volume ratio of the flavoring agent with serial number j in the corresponding flavor layer. This represents the overall preference score for spice number j. For spice with index k, N represents the overall preference score. T This corresponds to the amount of fragrance components in the fragrance layer.

[0014] Furthermore, when generating visual perfume cards, the top, middle, and base notes of the personalized perfume formula are respectively mapped to the foreground, midground, and background areas of the image; the visual proportion and volume ratio of each fragrance component are directly proportional.

[0015] Furthermore, the visualized perfume card is equipped with a unique identifier; users can trigger the playback of augmented reality (AR) content through the identifier to trace the perfume formula generation process.

[0016] The present invention also provides a perfume blending system based on virtual reality, comprising: The virtual scene building module is used to build a virtual lobby and multiple themed scene entrances displayed in the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type. The virtual lobby interactive module is used to present a virtual lobby to users through virtual reality devices. In response to users' interaction with the theme scene entrance, it releases the corresponding fragrance for users to rate. The themed scene interaction module is used to present the corresponding themed scene to users based on the highest-rated fragrance type; in response to the user's interaction with the fragrance component model, it releases the scent of the corresponding fragrance component. The subjective rating data calculation module is used to collect the user's brainwave data in real time through the brainwave signal acquisition device when the aroma component is released, and at the same time obtain the user's subjective rating data for the aroma component corresponding to the currently released aroma. The comprehensive preference score calculation module is used to generate an emotional valence index for the corresponding aroma component based on the collected EEG data, and to calculate the comprehensive preference score corresponding to the aroma component by weighted fusion of the emotional valence index and the subjective rating data. The personalized perfume formula generation module is used to select preferred fragrance components from a preset fragrance component library according to the comprehensive preference score of each fragrance component, categorized into top notes, middle notes, and base notes, and calculate the blending ratio of each preferred fragrance component based on the comprehensive preference score to generate a personalized perfume formula. The visual perfume card generation module generates a visual perfume card corresponding to the personalized perfume formula.

[0017] Beneficial effects: This invention establishes a phased interactive path for rapid initial screening in a virtual lobby and in-depth exploration of themed scenes. It first identifies users' core preferences by classifying fragrance types, and then conducts in-depth experiences with individual fragrance components, significantly reducing ineffective fragrance testing and fundamentally lowering the probability of olfactory fatigue. It achieves millisecond-level synchronization between VR interaction and scent release, allowing users' actions of grasping the fragrance component model to precisely trigger the start and stop of the corresponding scent. Combined with audiovisual scenes that match the fragrance style, it constructs an immersive experience that integrates multiple senses, completely eliminating environmental interference and a sense of disconnect in the experience.

[0018] This invention constructs an emotional valence index calculation model adapted to olfactory scenarios, and is compatible with both multi-channel and single-channel EEG acquisition devices. It can accurately capture the subconscious emotional responses of users when experiencing fragrances, transforming olfactory preferences that cannot be directly quantified into standardized, calculable indicators. A dynamically adjustable weighted fusion algorithm is designed to balance users' explicit subjective intentions and implicit physiological preferences. The default weight takes into account both subjective expression and subconscious real reactions, and can be dynamically adjusted according to user profiles and application scenarios. It supports the extended fusion of multiple indicators such as arousal and attention, and can adopt differentiated screening logic for different fragrance attributes of top, middle, and base notes, further improving the refinement of preference modeling.

[0019] This invention transforms abstract formula data and user emotional tags into personalized, visual perfume cards, achieving a synesthetic transformation from olfactory experience to visual presentation. The cards combine artistry, functionality, and commemorative significance, naturally aligning with social media platforms for viral brand dissemination. By uniquely identifying the card and binding it to the user's entire experience data, users can scan the card to review the customization process and emotional changes, significantly extending the customization experience cycle and strengthening the emotional connection between users and the brand. The end-to-end technical solution boasts strong scenario adaptability, allowing for lightweight deployment in offline beauty collection stores and perfume retail outlets, as well as customization services in high-end perfumery studios. It can also be used for brand metaverse marketing and perfume product user research. Furthermore, it is compatible with hardware devices of varying costs, from low-cost single-channel EEG devices to professional multi-channel equipment, demonstrating feasibility for large-scale commercial implementation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0021] Figure 1 This is a flowchart of the present invention; Figure 2 This is a schematic diagram of the structural principle of the present invention.

[0022] Figure 3 This is a flowchart of the present invention.

[0023] Figure 4 This is a schematic diagram of the system hardware architecture and how it is worn.

[0024] Figure 5 This is a schematic diagram of a rapid fragrance preference test scenario and a fragrance exploration scenario in the lobby.

[0025] Figure 6 This is a schematic diagram illustrating the interaction and rating interface in the scenario for detailed exploration.

[0026] Figure 7 A schematic diagram illustrating the creation of a fragrance customization result that interacts with AR. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0028] In this document, suffixes such as "module," "part," or "unit" used to denote elements are used only for the purpose of illustrative purposes and have no specific meaning in themselves. Therefore, "module," "part," or "unit" may be used interchangeably.

[0029] In this document, the terms "upper," "lower," "inner," "outer," "front," "rear," "one end," and "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the present invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0030] In this document, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0031] In this document, "and / or" includes any and all combinations of one or more of the listed related items.

[0032] In this article, "multiple" means two or more, that is, it includes two, three, four, five, etc.

[0033] Example 1: like Figure 1 , Figure 3 As shown, this embodiment provides a perfume blending method based on virtual reality, which specifically includes the following steps: S1, build a virtual lobby and multiple themed scene entrances displayed in the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type.

[0034] like Figure 4 As shown, the purpose of this step is to build a hierarchical virtual interaction system to solve the problems of mixed fragrances, high user decision-making threshold, and lack of systematic guidance in the olfactory experience in traditional perfume testing. It provides a standardized immersive interactive platform for subsequent user-staged fragrance testing, data collection, and preference locking. This step is built using the Unity / Unreal Engine and integrates SteamVR / OpenXR plugins for compatibility with mainstream VR headsets. All scene assets, interaction logic, and data binding are pre-configured during the initialization phase to ensure smooth, lag-free interaction for users after entering the system.

[0035] The virtual lobby is the first interactive interface and central node of the entire process after users enter the VR system. It is responsible for user guidance, initial screening of fragrance categories, and scene transitions.

[0036] like Figure 5As shown, for example, the virtual lobby can adopt a standardized circular / circular symmetrical layout, with the overall space being an open and bright environment to avoid the VR dizziness caused by enclosed spaces. The initial user position and system guidance interaction area are set up in the center of the lobby, with a built-in virtual guide NPC / text and image guidance panel to explain the system operation rules, fragrance testing process, and interaction methods to users. Multiple themed scene entrances are evenly distributed along the circular edge, with the number of entrances corresponding one-to-one with the number of preset fragrance categories. The default configuration is 5 entrances, each matching one of the 5 core fragrance categories commonly used in the perfume industry: floral, fruity, herbal, spicy, and woody.

[0037] Each themed scene entrance uses a visual design that strongly matches the corresponding fragrance. For example, the fruity fragrance entrance uses a combination of glowing fresh fruit, the floral fragrance entrance uses a combination of bouquets / gardens, and the woody fragrance entrance uses a combination of logs / pine branches. At the same time, each entrance is equipped with a glowing, highlighted border and the fragrance name text label. When the user hovers the controller over the entrance, the entrance model will trigger a dynamic feedback of enlargement and highlighting, and a brief description panel of the fragrance will pop up, explaining the style characteristics, applicable scenarios, and core aroma features of the fragrance to the user.

[0038] Each themed scene entry point is assigned a unique fragrance ID. This ID is uniquely bound to the fragrance category ID in the preset fragrance component library, the corresponding channel ID of the olfactory release device, and the category label of the user rating data. During the setup phase, end-to-end data mapping has been completed. All user interactions and rating data are categorized, labeled, and stored using the fragrance ID, providing a data basis for subsequently pushing corresponding themed scenes based on the highest-rated fragrance.

[0039] like Figure 6 As shown, each themed scene is divided into corresponding fragrance raw materials according to the general fragrance rules of the perfume industry. Each area displays several 3D models of fragrance components under that fragrance category and corresponding fragrance attributes. A natural, user-friendly interactive scene is constructed. By finding glowing interactive points in the scene, users explore fragrances. When a user approaches an interactive point, the corresponding fragrance information panel is triggered, establishing a complete understanding of perfume fragrance notes. For example, in the spicy fragrance themed scene, a medicinal kitchen scene is used, with models of spicy raw materials such as ginger and pepper displayed on the cooking station, and models of Chinese medicinal materials such as cinnamon, nutmeg, cloves, and fennel displayed in the medicine cabinet.

[0040] Aroma component models can be created using realistic or cartoon-style 3D modeling. The models are matched with the actual fragrance raw materials. For example, lemon aroma components correspond to a fresh lemon model, rose aroma components correspond to a rose flower model, and sandalwood aroma components correspond to a sandalwood model, ensuring that users can quickly establish a cognitive association with the aroma through visual means.

[0041] S2 controls the virtual reality module to display a virtual lobby to the user. In response to the user's interaction with the theme scene entrance in the virtual lobby, it controls the scent release module to release the corresponding fragrance for the user to rate.

[0042] The purpose of this step is to guide users to quickly experience and subjectively evaluate the core fragrance type through a low-barrier, immersive, and rapid fragrance trial interaction. This accurately identifies the fragrance category with high user preference and its aroma development stage (such as the middle note). The final fragrance formula will then incorporate these choices based on a predefined top, middle, and base note mixing ratio (R). T R M R B ) Calculations are combined (e.g., if the preferred fragrance type is in the middle notes, then the mixing ratio is mainly middle notes, corresponding to a mixing ratio of R). T :R M :R B =3:4:3), while avoiding the problems of olfactory fatigue and difficulty in selection caused by traditional blind testing of all categories.

[0043] The system uses the spatial positioning module of the VR device and the interaction component of the controller to capture the user's position changes and operation behavior in the virtual hall in real time. It sets two-level trigger rules for the interaction between the user and the theme scene entrance to ensure that the interaction logic conforms to the user's operating habits. At the same time, it achieves precise odor release control to avoid accidental triggering and cross-odor interference.

[0044] Level 1 Trigger. Proximity Interaction Pre-Response. The system monitors the spatial distance between the user's virtual position and the theme scene entrance in real time. When the user moves to within 1 meter of the entrance, it determines that the user has the intention to experience the fragrance and immediately executes two actions simultaneously. First, the entrance model triggers a highlighted and enlarged dynamic effect, and a brief panel pops up simultaneously displaying the fragrance's name, core scent characteristics, and applicable scenarios. Second, a pre-start command is sent to the olfactory control module to wake up the corresponding fragrance's scent release channel, preparing for subsequent scent release and eliminating the audiovisual-olfactory asynchrony problem caused by hardware startup delay.

[0045] Level 2 Trigger. Continuous Interaction Confirms Execution. When a user stays within 1 meter of the entrance for more than 2 seconds, the system determines that the user has confirmed the experience of the fragrance, immediately locks the scent channel for the current fragrance, officially sends the scent release command, and simultaneously blocks trigger signals from other theme scene entrances to avoid cross-contamination and olfactory confusion caused by the simultaneous release of multiple fragrances; only after the scent release process for the current fragrance is completely completed and the residual scent is cleared will the interactive trigger permissions for other entrances be restored.

[0046] The entire scent release process is highly synchronized with the visual feedback of the VR scene. During the release, the entrance model continuously plays dynamic effects of the fragrance flowing. After the release ends, the effects stop synchronously, and a rating prompt pops up in front of the user's field of vision, enhancing the synergistic experience of sight, hearing, and smell. After the release is complete, the system automatically triggers the residual odor removal process. The air pump of the olfactory device uses reverse suction to remove residual odor from the pipes and outlet, avoiding interference between different fragrances and ensuring the purity of the subsequent fragrance experience.

[0047] S3 controls the data acquisition module to obtain the user's rating of the fragrance type corresponding to the theme scene entrance, and controls the virtual reality module to present the corresponding theme scene to the user based on the fragrance type with the highest rating.

[0048] The purpose of this step is to provide users with an experience of a single aroma component within the user's most preferred aroma type. This is achieved through interactive triggering and synchronous release with the scent, enabling olfactory perception of a single aroma component.

[0049] Step S2 aims to allow users to complete the experience and rating of all themed scene stone tablet entrances. After that, the system backend immediately sorts all the normalized subjective ratings of the fragrances in descending order and automatically locks the fragrance with the highest rating as the user's core preferred fragrance. If two or more fragrances have the same rating, the system automatically compares the user's dwell time and interaction frequency during the corresponding fragrance experience. The fragrance with a longer dwell time and higher interaction frequency is selected as the highest rated fragrance. At the same time, after all experience ratings are completed, the stone tablet gate lights up, allowing users to choose to enter.

[0050] Upon entering the corresponding themed scene, the user's initial position is set in the entrance guidance area of ​​the themed scene. Within their field of vision, they can clearly see the overall layout of the scene. For example, the floral fragrance scene plays garden ambient sounds in sync, and the woody fragrance scene plays forest wind sounds in sync, achieving basic synergy of multiple senses: sight, hearing, and smell.

[0051] The system uses the spatial positioning module of the VR headset and the motion capture component of the controller to capture the user's spatial position, controller movements, and viewing direction, and sets three-level trigger rules for the interaction between the user and the aroma component model: Level 1 Trigger. Hover Preview Interaction: When the user's controller ray is aimed at the aroma component model in the scene, and the contact time between the ray and the model's collider exceeds 0.5 seconds, the system determines that the user has the intention to view the ingredient. At this time, the model triggers a dynamic effect of highlighting and slowly rotating, and a details information panel pops up next to the model. The panel displays the name, fragrance attributes, scent description, and lasting fragrance duration of the aroma component.

[0052] Level 2 Trigger. Grab Confirmation Interaction: When the user presses the grab button on the controller and completes the grabbing action on the aroma component model, the system determines that the user has confirmed the experience of that aroma component and immediately triggers the following interaction: First, it locks the odor release channel corresponding to that aroma component and simultaneously blocks the interaction trigger permissions of all other aroma component models in the scene to prevent accidental user operation that would trigger multiple channels to release simultaneously; second, it sends a synchronous release command to the olfactory control module and a synchronous start command to the data acquisition module, achieving synchronization of odor release, VR visual feedback, and data acquisition; third, the model follows the user's controller movement while simultaneously playing particle effects depicting the flow of aroma, enhancing the immersive interactive experience.

[0053] Three-level trigger. The release and completion interaction automatically triggers the end process after the system completes the preset duration of scent release: First, the controller vibrates once to inform the user that the experience is over, and simultaneously releases the model's grip lock, allowing the model to automatically return to its original position; second, the corresponding scent release channel is closed, and the residual scent cleanup process is initiated; third, a rating panel pops up directly in front of the user's field of vision, guiding the user to complete a subjective rating, while simultaneously unlocking the interaction locks of other scent component models, allowing the user to experience the next scent component.

[0054] S4, in response to the user's interaction with the aroma component model within the theme scene, controls the aroma release module to release the aroma of the corresponding aroma component for the user to rate, and controls the data acquisition module to collect the user's EEG data.

[0055] The system collects data from the frontal lobe region of the user's brain (F3 and F4 channels). The EEG signals in this region are directly related to human emotional responses and attention levels, and serve as the data source for subsequent preference index calculations.

[0056] After single-segment EEG data acquisition, the data is preprocessed, including filtering and noise reduction, and artifact removal. The Welch method (Hamming window, window length 2 seconds, overlap 50%) is used to calculate the average power of each core frequency band for the preprocessed clean signal. The power values ​​of the alpha, beta, and theta frequency bands of the F3 and F4 channels in the frontal lobe are extracted.

[0057] Meanwhile, the system adopts an integer rating system from 1 to 10 points. The rating panel uses a dual input mode of linear slider and number keys, which is fully adapted to the operation habits of VR controllers. The left side of the panel is marked "Dislike" corresponding to 1 point, and the right side is marked "Like" corresponding to 10 points. The top of the panel displays the name of the aroma component, 3D thumbnail model and fragrance attributes corresponding to the current rating, which visually strengthens the correspondence between the rating and the aroma component and avoids users from making rating mismatches.

[0058] After a user submits a rating, the system immediately standardizes and normalizes the original rating data from 1 to 10, linearly mapping the original rating to the [0,1] interval, eliminating the dimensional differences between subjective ratings and EEG physiological indicators, and providing a unified data standard for the subsequent weighted fusion calculation of the two types of data; the original rating and the normalized rating data are stored synchronously and bound to the raw material ID of the current aroma component.

[0059] In some embodiments, the scent release module can employ an existing smart fragrance device. This smart fragrance device integrates a communication module (such as a wireless communication module like Bluetooth) and has multiple atomizers (each atomizer can release a different fragrance). When it receives a corresponding trigger signal, it can release the corresponding fragrance based on that signal. Of course, the trigger signal carries identification information such as the name or number of the fragrance to be released, allowing the smart fragrance device to control the corresponding atomizer to release the corresponding fragrance based on this identification. This smart fragrance device is existing technology and will not be described in detail here.

[0060] S5, the control data processing module generates the emotional valence index of the corresponding aroma component based on the collected user EEG data, and calculates the comprehensive preference score of the aroma component based on the emotional valence index and the user's subjective rating data of the aroma component.

[0061] The emotional valence index is a core indicator for quantifying users' subconscious emotional preferences for aroma components. It reflects the degree of pleasure (positive / negative) and intensity of users' experience with aromas. Its calculation is based on the EEG data collected and preprocessed in step S4.

[0062] Specifically, the aroma components of user experience are extracted from the preprocessed EEG data. i At that time, the average power values ​​of the α band (8-13Hz) of the F3 (left frontal lobe) and F4 (right frontal lobe) channels in the brain were denoted as P. α,F3 (s i ) and P α,F4 (s i The power difference in this frequency band is a core indicator recognized in the industry for judging emotional valence. A decrease in the power of the left frontal alpha wave and an increase in the power of the right frontal wave correspond to positive pleasant emotions, while the opposite corresponds to negative aversion emotions.

[0063] The emotional valence index described in this embodiment uses the following formula: ; Calculate; where VEI(s) i For user experience, aroma components i Emotional valence index at the time; P α,F4 (s i For user experience, aroma components iAt that time, the average power of the α band of the F4 channel in the frontal lobe of the EEG signal; P α,F3 (s i For user experience, aroma components i At that time, the average power of the α band of the F3 channel in the frontal lobe of the EEG signal.

[0064] To maintain consistency with the normalized interval [0,1] of the subjective ratings, the system adjusts the original VEI (s) i The data was linearly normalized and converted into a normalized emotional valence index (VEI). norm (s i The formula is: ; Normalized VEI norm (s i The value range of ) is [0,1], where 0 corresponds to extreme aversion, 1 corresponds to extreme pleasure, and 0.5 corresponds to no obvious emotional tendency, ensuring consistency with the normalized data L of subjective ratings. n (s i The unification of dimensions provides a foundation for subsequent weighted fusion.

[0065] The normalized emotional valence index is weighted and fused with the normalized subjective rating data to balance the weights of physiological subconscious responses and users' explicit expressed preferences. It also supports multi-indicator expansion and fusion, as detailed below: The basic weighted fusion method is suitable for lightweight scenarios involving quick fragrance selection. The fusion formula is: ; Wherein, P(s) i ) is the aroma component s i The overall preference score, Ln(s) i ( ) represents the normalized subjective rating data, VEI norm (s i ω is the normalized emotional valence index. I ω is the weighting coefficient for subjective ratings. O Let be the weighting coefficients of the normalized sentiment valence index, and satisfy ω I +ω O= 1. In this embodiment, the weighting coefficient is set to ω by default. I =0.6、ω O =0.4, and can be dynamically adjusted according to application scenarios or user profiles.

[0066] For scenarios requiring extreme precision, such as high-end customization and user research, the system introduces arousal and attention indices into the basic model to construct a multi-dimensional extended fusion model. This more comprehensively captures users' multi-dimensional psychophysiological responses to fragrances. The formula is as follows: Pext (s i )=ω I *L n (s i )+ω O *VEI norm (s i )+ω A *A norm (s i )+ω T *T norm (s i ); Among them, P ext (s i ) represents the expanded comprehensive preference score, Ln(s) i ( ) represents the normalized subjective rating data, VEI norm (s i A is the normalized emotional valence index. norm (s i T represents the normalized arousal index. norm (s i ω represents the normalized attention index; I ω is the weighting coefficient for subjective ratings. O ω is the weighting coefficient of the normalized emotional valence index. A ω is the weighting coefficient of the wakefulness index. T Let be the weighting coefficient of the attention index, and satisfy ω I +ω O +ω A +ω T =1.

[0067] This formula uses an attention index to screen top notes, which form the first impression of the perfume, and the level of attention directly reflects the user's initial impression preference. It uses an arousal index to screen middle and base notes, which determine the core tone of the perfume, and arousal reflects the user's deep emotional response to the fragrance, thus achieving refined screening of fragrance raw materials.

[0068] S6: After user experience, the fragrance components are sorted according to the overall preference score. The formula generation module selects the top-ranked fragrance components from the preset fragrance component library according to the top, middle and base notes as preferred fragrance components. The volume ratio of each preferred fragrance component is calculated to generate a personalized perfume formula.

[0069] The purpose of this step is to quantify abstract user preferences and transform them into standardized, personalized perfume formulas that conform to perfume industry standards and can be directly used for physical blending. Specifically, the steps to generate personalized perfume formulas include: S61 divides all aroma components in the preset aroma component library into three categories: top notes, middle notes, and base notes according to their aroma attributes. For all aroma components experienced by users in each category, sort them from high to low according to their overall preference scores.

[0070] Based on the perfume industry's common standards for fragrance volatility and lasting power, each fragrance component is labeled with a unique fragrance attribute. Fragrances with high volatility and lasting power <2 hours are labeled as top notes, those with medium volatility and lasting power 2-6 hours are labeled as middle notes, and those with low volatility and lasting power >6 hours are labeled as base notes.

[0071] Each fragrance component is labeled with its corresponding fragrance category, including floral, fruity, herbal, spicy, and woody, which corresponds one-to-one with the theme scene classification.

[0072] The system calls the preset aroma component library and divides all aroma components in the library into three independent sets according to the pre-labeled aroma attributes: top notes, middle notes, and base notes, denoted as top note set T, middle note set M, and base note set B, respectively, to ensure that each aroma component belongs to only one aroma category.

[0073] From the three fragrance families, the system filters out fragrance components for which the user has completed a full immersive experience and whose overall preference score is valid. These components are then sorted in descending order of overall preference score, with higher-ranked components indicating higher user preference and thus higher priority in subsequent filtering stages. If two or more components have the same overall preference score, the system automatically compares their corresponding Normalized Emotional Valence (VEI). norm , with VEI norm Components with higher numerical values ​​are prioritized, and users' subconscious physiological preferences are used as the core basis for typological sorting to ensure that the sorting results are more in line with users' real olfactory preferences.

[0074] S62, based on the preset formula complexity parameters, select the several aroma components with the highest comprehensive preference scores from the ranking results of each type of aroma component as preferred aroma components.

[0075] The formula complexity parameter defines the number of preferred ingredients selected for each of the top, middle, and base notes, denoted as the number of top note selections, N. T Number of screenings in the middle stage N M Post-screening quantity N B The system offers multiple configurable options to suit the needs of different users.

[0076] For example, the default beginner configuration: the system defaults to the industry-standard beginner-friendly recipe structure, setting N... T =2、N M =3、N B=2. This configuration has been tested and verified by a large number of users. It can ensure the aroma layers of the formula without making the blending too difficult due to too many ingredients, and it is suitable for the needs of ordinary users in offline retail stores.

[0077] Minimalist recipe configuration: such as setting N T =1、N M =2、N B =1, suitable for beginners who prefer a single core fragrance and have very low blending difficulty; Enriching recipe configurations: such as setting N T =3、N M =4、N B =3, suitable for advanced users with some perfumery experience who pursue a richer and more complex aroma; Professional custom configuration: It allows users to manually adjust the number of each fragrance note to be selected, and set the maximum number of single fragrance notes to be selected to 5, so as to avoid too many ingredients causing fragrance conflicts and chaos, which is in line with the general standards of the perfumery industry.

[0078] Based on the formula mode selected by the user, the corresponding formula complexity parameters are loaded to determine the number of preferred ingredients to be screened for each of the three fragrance notes.

[0079] Based on the loaded parameters, select the top-ranked N from the three fragrance notes, sorted in descending order. T Pre-regulatory components, N M Medium-regulating components, N B These base notes, as preferred aroma components for the corresponding fragrance notes, are included in the ingredient list of the final formulation.

[0080] After screening, the system automatically calls the preset fragrance compatibility database to perform compatibility checks on all the selected preferred ingredients, checking for aroma conflicts and incompatibilities between the ingredients. If conflicting ingredients are found, the system automatically replaces the conflicting ingredients with the next-ranked ingredient with the same fragrance notes in the sorting, and simultaneously informs the user of the reason for the replacement, ensuring that the final selected ingredients are harmonious and compatible in aroma, comply with professional perfumery standards, and avoid generating formulas with chaotic aromas that cannot be used in practice.

[0081] S63. For each type of fragrance attribute, the volume ratio of the preferred fragrance component in the corresponding fragrance attribute is calculated according to the proportion of the comprehensive preference score of each preferred fragrance component to the total comprehensive preference score of all preferred fragrance components in that category, and finally a complete personalized perfume formula is generated.

[0082] Specifically, in this embodiment, the volume ratio of various aroma components within a certain layer is determined using the following formula: ; Calculate; where, This represents the volume ratio of the flavoring agent with serial number j in the corresponding flavor layer. This represents the overall preference score for spice number j. For spice with index k, N represents the overall preference score. T This corresponds to the amount of fragrance components in the fragrance layer.

[0083] For the top, middle, and base notes, the preferred ingredients are substituted into the core formula to calculate the volume ratio of each ingredient within its corresponding note. For example, if two preferred ingredients are selected for the top note with overall preference scores of 0.8 and 0.6 respectively, and the total score is 1.4, then the volume ratios of the two ingredients within the top note are approximately 57.14% (0.8 / 1.4) and 42.86% (0.6 / 1.4), respectively. The sum of these two ratios is 100%, which meets the normalization requirements within the fragrance note.

[0084] The system combines the preferred ingredients and corresponding volume ratios within the three fragrance notes with a predefined top, middle, and base note mixing ratio derived from initial preference assessments, conforming to industry-standard ranges. This process calculates the final volume percentage of each ingredient in the entire perfume bottle, generating a complete, standardized perfume formula. The final output formula includes: a unique formula ID, a fragrance style tag, a list of preferred ingredients for each fragrance note, the volume ratio of each individual ingredient within the fragrance note, and the final volume percentage of each individual ingredient in the entire bottle. Users can directly follow the formula to manually blend their perfumes.

[0085] S7 controls the card generation module to generate visual perfume cards based on personalized perfume formulas.

[0086] like Figure 7 As shown, the list of preferred fragrance components for top, middle, and base notes extracted from personalized perfume formulas, the volume ratio of each component within the corresponding fragrance note, the final proportion of the whole perfume, and fragrance style labels such as fresh floral, woody spicy, and citrus fruity are used as the basis for constructing the visual elements of the card.

[0087] Simultaneously extract the comprehensive preference scores, normalized emotional valence index, arousal and attention indicators of each component calculated in step S5, and generate user-specific emotional tags such as pleasant and soothing, vibrant and refreshing, and calm and reserved based on numerical features. These tags serve as the basis for defining the style, color, and lighting of the card image, realizing the synesthetic transformation from olfactory experience to visual presentation.

[0088] Extract the globally unique ID of the recipe, the unique identifier of the user experience, the timestamp of the recipe generation, and the key data index of the experience process to provide a data foundation for the unique identification binding of cards and the association of AR content.

[0089] Following the fragrance logic of top, middle, and base notes, the basic hierarchical structure of the image is defined. The template is: generate a vertical perfume card art image, with the image strictly divided into three levels: foreground, middle ground, and background. The foreground corresponds to the top notes of the perfume, the middle ground corresponds to the middle notes of the perfume, and the background corresponds to the base notes of the perfume.

[0090] The aroma components and volume ratios in the formula are quantified into control parameters for visual elements in the image. For example, the foreground contains components 1 and 2 representing the top notes, and the element's proportion, clarity, color saturation, and visual prominence are directly proportional to the volume ratio of the corresponding components. The midground contains components 1, 2, and 3 representing the middle notes, serving as the core subject of the image. The background contains components 1 and 2 representing the base notes, presented in a blurred form to create the atmosphere. This layer achieves a one-to-one correspondence between formula ratios and visual elements; the higher the proportion of a component, the higher its visual weight in the image, ensuring that the visual content of each card perfectly matches the exclusive formula.

[0091] In addition, the visual perfume card is equipped with a unique identifier, such as a QR code; users can trigger the playback of augmented reality (AR) content through the identifier to trace the perfume formula generation process.

[0092] The card's QR code is linked to the formula data. Users can scan the QR code to view the complete digital formula, ingredient details, and mixing instructions, allowing them to access formula information anytime without carrying a physical card. The unique card identifier is linked to the entire user experience data process, and key scenes, emotional change curves, rating records, and aroma component experience videos of the user's experience in the VR environment are associated with the card ID and stored on the cloud server, providing a complete data source for subsequent AR content playback and experience review; The card identifier is linked to the user's unique account. Users can view all historically generated formulas and corresponding cards in their account at any time. The system supports secondary adjustments to formulas and regeneration of cards, enabling long-term preservation of the user's exclusive fragrance profile.

[0093] Example 2: like Figure 2 As shown, the present invention also provides a perfume blending system based on virtual reality, comprising: The virtual scene building module is used to build a virtual lobby and multiple themed scene entrances displayed in the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type. The virtual lobby interaction module is used to present a virtual lobby to users through virtual reality devices. In response to the user's interaction with the theme scene entrance, the scent release module releases the corresponding fragrance for the user to rate. The themed scene interaction module is used to present the corresponding themed scene to users based on the highest-rated fragrance type; in response to the user's interaction with the fragrance component model, it releases the scent of the corresponding fragrance component. The subjective rating data calculation module is used to collect the user's brainwave data in real time through the brainwave signal acquisition device when the aroma component is released, and at the same time obtain the user's subjective rating data for the aroma component corresponding to the currently released aroma. The comprehensive preference score calculation module is used to generate an emotional valence index for the corresponding aroma component based on the collected EEG data, and to calculate the comprehensive preference score corresponding to the aroma component by weighted fusion of the emotional valence index and the subjective rating data. The personalized perfume formula generation module is used to select preferred fragrance components from a preset fragrance component library according to the comprehensive preference score of each fragrance component, categorized into top notes, middle notes, and base notes, and calculate the blending ratio of each preferred fragrance component based on the comprehensive preference score to generate a personalized perfume formula. The visual perfume card generation module generates a visual perfume card corresponding to the personalized perfume formula.

[0094] In some embodiments, the odor release module connected to the above system includes an atomizer array, an air pump for driving the atomizer array (all atomizers are connected to one air pump), and a main control board; the atomizer array includes multiple odor channels, each odor channel being provided with a liquid sample of a certain fragrance; preferably, each odor channel is provided with a micro-atomizer and a one-way valve; the micro-atomizer, the one-way valve, and the air pump are electrically connected to the main control board; the main control board is configured to, upon triggering by the virtual lobby interaction module, control the opening of the one-way valve of the corresponding odor channel and control the air pump to drive the micro-atomizer of the corresponding odor channel to release the corresponding fragrance.

[0095] Specifically, the scent release module can utilize existing smart fragrance devices. These devices integrate a communication module (such as Bluetooth or other wireless communication modules) and multiple atomizers (each atomizer can release a different fragrance). When it receives a corresponding trigger signal, it can release the corresponding fragrance based on that signal. Of course, the trigger signal carries identification information such as the name or number of the fragrance to be released, allowing the smart fragrance device to control the corresponding atomizer to release the appropriate fragrance based on this identification. This smart fragrance device is existing technology and will not be described in detail here.

[0096] For example, 27 atomizers are set up, and each atomizer is electrically connected to a driver board.

[0097] For example, the communication control section of this scent release module uses two Arduino Uno R3 development boards as the main control modules. Each development board connects to 13-14 atomizer driver boards via an I / O expansion board (the first board connects to 14, the second board connects to 13, for a total of 27). Each development board is connected to the host computer via a USB cable and occupies a different COM port (e.g., COM3 and COM4). Each driver board is assigned a unique digital pin or I2C address (i.e., a unique port identifier; correspondingly, the system pre-stores the unique port identifiers and corresponding fragrance types for each of the 27 atomizers, so the trigger signal includes the unique port identifier corresponding to the fragrance to be released), and implements the function of receiving serial port commands to control the start and stop of the corresponding atomizer.

[0098] Specifically, C# scripts are written in the Unity engine to communicate with two Arduino Uno R3 development boards via the SerialPort class. The controller (such as an Xbox controller) receives input through Unity's Input System. When the user presses any button or joystick on the controller, Unity sends a command (i.e., a trigger signal, such as "A1,ON,1000" indicating that atomizer #1 is turned on for 1000ms) to the corresponding COM port. After parsing the command, the Arduino Uno R3 development board drives the corresponding pin to output a high level, thereby driving the corresponding atomizer to operate.

[0099] Furthermore, the device supports a maintenance mode: after the host computer sends a self-test command, the two Arduino Uno R3 development boards activate 27 atomizers in sequence, with each atomizer spraying for 1 second. Users can judge whether it is normal by observing the spray sequence and mist volume.

[0100] When a driver board needs to be replaced, simply remove the corresponding M3 screw, unplug the connecting cable, replace the new board, and power on again. The system will automatically recognize it through the COM port and pin markings, without needing to modify the overall code.

[0101] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a computer terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0103] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A perfume blending method based on virtual reality, characterized in that, Includes the following steps: A virtual lobby and multiple themed scene entrances are set up within the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type; A virtual lobby is presented to users through virtual reality devices, and corresponding fragrances are released in response to user interaction with the theme scene entrance for users to rate. Based on the highest-rated fragrance, present users with corresponding themed scenarios; In response to user interaction with the aroma component model, the corresponding aroma component is released. When the aroma components are released, the user's brainwave data is collected in real time through the EEG signal acquisition device, and the user's subjective rating data of the aroma components corresponding to the currently released aroma is obtained at the same time. Based on the collected EEG data, an emotional valence index for the corresponding aroma component is generated. The emotional valence index is then weighted and fused with the subjective rating data to calculate the comprehensive preference score corresponding to the aroma component. Based on the overall preference score of each aroma component, the preferred aroma components are selected from the preset aroma component library by category into top notes, middle notes and base notes. The blending ratio of each preferred aroma component is calculated based on the overall preference score to generate a personalized perfume formula. Based on the personalized perfume formula, a visual perfume card corresponding to the formula is generated.

2. The perfume blending method based on virtual reality according to claim 1, characterized in that, The emotional valence index uses the following formula: ； Calculate; where VEI(s) i For user experience, aroma components i Emotional valence index at the time; P α,F4 (s i For user experience, aroma components i At that time, the average power of the α band of the F4 channel in the frontal lobe of the EEG signal; P α,F3 (s i For the user experience, the fragrance components are... i At that time, the average power of the α band of the F3 channel in the frontal lobe of the EEG signal.

3. The perfume blending method based on virtual reality according to claim 2, characterized in that, The formula for calculating the overall preference score for aroma components is as follows: ； Calculate; where P(s) i ) is the aroma component s i The overall preference score, Ln(s) i ( ) represents the normalized subjective rating data, VEI norm (s i ω is the normalized emotional valence index. I ω is the weighting coefficient for subjective ratings. O Let be the weighting coefficients of the normalized sentiment valence index, and satisfy ω I +ω O= 1.

4. The perfume blending method based on virtual reality according to claim 3, characterized in that, The weighting coefficient is set to ω by default. I =0.6、ω O =0.4, and can be dynamically adjusted according to application scenarios or user profiles.

5. The perfume blending method based on virtual reality according to claim 1, characterized in that, By introducing the arousal index and attention index to expand and fuse multiple indicators, the extended comprehensive preference score is calculated as follows: P ext (s i )=ω I *L n (s i )+ω O *VEI norm (s i )+ω A *A norm (s i )+ω T *T norm (s i ); Among them, P ext (s i ) represents the expanded comprehensive preference score, Ln(s) i ( ) represents the normalized subjective rating data, VEI norm (s i A is the normalized emotional valence index. norm (s i T represents the normalized arousal index. norm (s i ω represents the normalized attention index; I ω is the weighting coefficient for subjective ratings. O ω is the weighting coefficient of the normalized emotional valence index. A ω is the weighting coefficient of the wakefulness index. T Let be the weighting coefficient of the attention index, and satisfy ω I +ω O +ω A +ω T =1.

6. The perfume blending method based on virtual reality according to claim 1, characterized in that, The steps involved in creating a personalized fragrance formula include: All aroma components in the preset aroma component library are divided into three categories according to their aroma attributes: top notes, middle notes, and base notes. For all aroma components experienced by users in each category, they are sorted from high to low according to their overall preference scores. Based on the preset formula complexity parameters, select the top-ranked aroma components with the highest overall preference scores from the ranking results of each type of aroma component as preferred aroma components. For each fragrance family, the volume ratio of the preferred fragrance component within the corresponding fragrance family is calculated based on the proportion of the overall preference score of each preferred fragrance component to the total overall preference score of all preferred fragrance components in that family, thereby generating a personalized perfume formula.

7. A perfume blending method based on virtual reality according to claim 1, characterized in that, The volume ratio of various aroma components within a certain tonality layer is determined using the following formula: ； Calculate; where, This represents the volume ratio of the flavoring agent with serial number j in the corresponding flavor layer. This represents the overall preference score for spice number j. For spice with index k, N represents the overall preference score. T This corresponds to the amount of fragrance components in the fragrance layer.

8. The perfume blending method based on virtual reality according to claim 1, characterized in that, When generating a visual perfume card, the top, middle, and base notes of the personalized perfume formula are respectively mapped to the foreground, midground, and background areas of the image; the visual proportion and ratio of each fragrance component are proportional; and / or, the visual perfume card is assigned a unique identifier; users can trigger the playback of augmented reality (AR) content through the identifier to trace back the perfume formula generation process.

9. A perfume blending system based on virtual reality, characterized in that, include: The virtual scene building module is used to build a virtual lobby and multiple themed scene entrances displayed in the virtual lobby; each themed scene entrance links to a themed scene, each themed scene corresponds to a fragrance type, and each themed scene contains several fragrance component models under that fragrance type. The virtual lobby interaction module is used to present a virtual lobby to users through virtual reality devices. In response to the user's interaction with the theme scene entrance, the scent release module releases the corresponding fragrance for the user to rate. The theme scene interaction module is used to present corresponding theme scenes to users based on the highest-rated fragrance type; In response to user interaction with the aroma component model, the corresponding aroma component is released. The subjective rating data calculation module is used to collect the user's brainwave data in real time through the brainwave signal acquisition device when the aroma component is released, and at the same time obtain the user's subjective rating data for the aroma component corresponding to the currently released aroma. The comprehensive preference score calculation module is used to generate an emotional valence index for the corresponding aroma component based on the collected EEG data, and to calculate the comprehensive preference score corresponding to the aroma component by weighted fusion of the emotional valence index and the subjective rating data. The personalized perfume formula generation module is used to select preferred fragrance components from a preset fragrance component library according to the comprehensive preference score of each fragrance component, categorized into top notes, middle notes, and base notes, and calculate the blending ratio of each preferred fragrance component based on the comprehensive preference score to generate a personalized perfume formula. The visual perfume card generation module generates a visual perfume card corresponding to the personalized perfume formula.

10. A perfume blending system based on virtual reality according to claim 9, characterized in that, The odor release module includes: An atomizer array, an air pump for driving the atomizer array, and a main control board; the atomizer array includes multiple scent channels, each scent channel is provided with a liquid sample of a certain fragrance, and each scent channel is provided with a one-way valve; the one-way valve and the air pump are electrically connected to the main control board; the main control board is configured to, upon triggering by the virtual lobby interactive module, control the opening of the one-way valve of the corresponding scent channel and control the air pump to drive the corresponding scent channel to release the corresponding fragrance.

Images