An adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition and a control method thereof

Abstract

This invention discloses an adaptive sterilization and deodorization shoe cabinet system and control method based on AI visual recognition. The system includes: an AI visual recognition module for acquiring shoe images and outputting shoe feature vectors such as shoe type, material, and dry / wet status; an environmental perception module for real-time acquisition of environmental data such as temperature, humidity, VOCs, bacteria, and ozone inside the cabinet; an AI intelligent control module for fusing multimodal data such as shoe feature vectors and environmental data, and outputting a processing solution through AI model inference; a sterilization and deodorization function module including blue light lamps, titanium dioxide dry powder, and an ozone generator to perform zoned adaptive sterilization and deodorization; and a safety protection module. This invention can provide a tailored adaptive processing solution for different shoe characteristics, effectively protecting shoe materials and reducing energy consumption. Furthermore, this invention features AI closed-loop safety management and user habit prediction functions, greatly enhancing the intelligent experience and security.

Description

Technical Field

[0001] This invention relates to the field of sterilization and deodorization shoe cabinet systems, and in particular to an adaptive sterilization and deodorization shoe cabinet system and control method based on AI visual recognition. Background Technology

[0002] As people's living standards improve and the demand for smart homes increases, the function of shoe cabinets has evolved from simple storage to include sterilization, deodorization, dehumidification, and intelligent control. While existing smart shoe cabinets have made progress in some aspects, they also have many shortcomings. For example, these smart shoe cabinets typically rely solely on data from environmental sensors such as temperature, humidity, odor, and bacterial concentration to trigger their operation, lacking AI visual recognition capabilities for the shoes themselves. This means they cannot clearly distinguish between different shoe types, materials, wetness / dryness levels, dirt levels, and shoe occupancy, making the target of processing unclear.

[0003] In addition, existing smart shoe cabinets generally suffer from the following problems: First, they typically use fixed modes, times, and intensities for sterilization, deodorization, and dehumidification, failing to adapt to the characteristics of the shoes. This can easily lead to problems such as leather aging, suede deterioration, or mesh damage, while also wasting energy. Second, the control logic of existing smart shoe cabinets is relatively simple, lacking the use of artificial intelligence (AI) for autonomous decision-making, dynamic parameter adjustment, closed-loop feedback, and iterative optimization. Manual or timed activation is still required, resulting in a low level of intelligence. Third, safety protection and ozone residue degradation rely on simple hardware safeguards, lacking AI-based calculations, predictions, and intelligent control, posing safety hazards such as ozone leakage, over-sterilization, and excessive residue levels. Summary of the Invention

[0004] To overcome the aforementioned problems in existing technologies, this invention provides an adaptive sterilization and deodorization shoe cabinet system and control method based on AI visual recognition. This system can intelligently identify the type, material, dryness, dirt, and occupancy status of shoes. Simultaneously, it can generate corresponding processing solutions based on shoe feature data and environmental data to protect shoe materials, reduce energy consumption, and ensure user safety.

[0005] According to a first aspect of the present invention, an adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition is provided, comprising: The AI ​​visual recognition module is located at the top of the cabinet and is used to acquire images of shoes and output characteristic data of shoes, including shoe type, material, dry and wet condition, dirt level, and shoe space occupancy status. The environmental sensing module is used to collect environmental data inside the shoe cabinet; The AI ​​intelligent control module is connected to the AI ​​visual recognition module and the environmental perception module. It is used to integrate the footwear feature data with the environmental data, module operation data and user historical behavior data into the AI ​​model, and infer and output a processing solution. The processing solution includes at least whether to process, the processing area, the sterilization intensity, the ozone concentration, the fan speed and the working time. The sterilization and deodorization module includes a black titanium dioxide photocatalytic dry powder box and an ozone generator, which are used to receive the processing scheme and perform zoned sterilization and deodorization; The safety protection module is used to monitor the cabinet door status and ozone concentration in real time, and to start the degradation process when the ozone concentration exceeds the standard until the ozone concentration is lower than the safety threshold and then unlock the cabinet door.

[0006] According to a second aspect of the present invention, a control method for an adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition is provided, the method comprising: S1: System power-on pre-start, door magnetic signal activates AI intelligent control module; S2: The AI ​​visual recognition module collects images of shoes, identifies shoe type, material, dryness / wetness status, dirt level, and shoe space occupancy status, and outputs shoe feature data; the environmental perception module simultaneously collects environmental data inside the cabinet. S3: The AI ​​intelligent control module integrates the footwear feature data with the environmental data, module operation data and user historical behavior data, inputs them into the AI ​​model, and infers and outputs a processing solution. S4: According to the processing scheme, the sterilization and deodorization module will activate the ozone generator and blue light in the corresponding shoe compartment, and adjust the ozone concentration, blue light duration, fan speed and PTC heating power. S5: Real-time monitoring of residual ozone concentration via an ozone concentration sensor. If the concentration exceeds the safety threshold, the cabinet door will be unlocked and a voice and app notification will be issued.

[0007] Compared with existing technologies, the advantages of this invention are as follows: By introducing an AI visual recognition module, the pain point of existing shoe cabinets being unable to identify the characteristics of shoes can be solved, achieving adaptive and precise sterilization and deodorization for different shoe types and materials, effectively avoiding damage to delicate materials such as leather shoes and suede shoes caused by traditional fixed modes, while significantly reducing ineffective energy consumption; by combining the safety protection module and the sterilization and deodorization module, the safety hazard of ozone leakage is eliminated, ensuring the respiratory health of users. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of an adaptive sterilization and deodorization shoe cabinet system architecture based on AI visual recognition, provided in an embodiment of the present invention.

[0009] Figure 2This is a schematic diagram of the structure of the adaptive sterilization and deodorization shoe cabinet provided in an embodiment of the present invention;

[0010] Figure 3 A cross-sectional structural diagram of the adaptive sterilization and deodorization shoe cabinet provided in an embodiment of the present invention;

[0011] Figure 4 This is a schematic diagram of the control method for an adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition, provided in an embodiment of the present invention.

[0012] Numbered in the diagram: 1-Touchscreen LCD; 2-Cabinet door; 3-Layered, hollow, pull-out shelf; 4-High-definition camera; 5-450nm blue light lamp; 6-Ozone generator; 7-Inner air outlet; 8-Transparent boxed black titanium dioxide powder; 9-PTC constant temperature heating device; 10-First fan; 11-Second fan; 12-VOC detection sensor; 13-Ozone detection sensor; 14-Temperature and humidity sensor; 15-Bacterial concentration sensor; 16-AI main control chip; 17-Filter and activated carbon adsorption; 18-Door magnetic induction switch. Detailed Implementation

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

[0014] Existing technologies rely solely on environmental sensors such as temperature, humidity, and odor sensors to trigger operation, lacking the capability for AI visual recognition of the shoes themselves. This results in unclear processing targets and an inability to distinguish shoe type, material, wetness / dryness, dirt level, and shoe space occupancy status. The AI ​​visual recognition module of this invention acquires shoe images and outputs shoe feature data including shoe type, material, wetness / dryness, dirt level, and shoe space occupancy status, providing a reliable data foundation for subsequent adaptive control.

[0015] Existing technologies suffer from limitations due to their fixed modes, durations, and intensities in sterilization, deodorization, and dehumidification, which can easily lead to problems such as leather aging and suede deterioration. The AI ​​intelligent control module of this invention overcomes these limitations by fusing shoe feature data with environmental data, module operation data, and user historical behavior data, and then using an AI model to output a unified processing solution. This achieves adaptive processing tailored to each shoe, effectively protecting shoe materials and reducing energy consumption.

[0016] Existing technologies have limitations in terms of safety protection and ozone residue degradation, relying on simple hardware protection, which poses a safety hazard of ozone leakage. According to embodiments of the present invention, the safety protection module and the AI ​​intelligent control module work together to monitor ozone concentration in real time and only unlock the cabinet door after the concentration reaches the standard, overcoming this limitation and achieving AI closed-loop safety management, thus thoroughly protecting the personal safety of users.

[0017] The overall system architecture and apparatus of the present invention will be described in detail below.

[0018] like Figures 1 to 3 As shown, this invention provides an adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition. In terms of overall appearance and physical structure, the shoe cabinet system includes a cabinet body, a touch-screen LCD 1, cabinet doors 2, and layered, perforated, pull-out shelves 3. The cabinet dimensions can be designed to be 80cm high, 40cm wide, and 35cm deep, and it is made of environmentally friendly board material, ensuring both structural strength and compliance with modern home environmental requirements. The design of the layered, perforated, pull-out shelves 3 not only allows users to freely adjust the shelf height according to the height of their shoes, but its perforated structure also ensures smooth air circulation within the cabinet, allowing ozone, hot air, and photocatalyst-generated active substances to penetrate into the interior of each pair of shoes without any blind spots.

[0019] In terms of system functional modules, it mainly consists of an AI intelligent control module, an AI visual recognition module, a sterilization and deodorization module, an environmental sensing module, a safety protection module, an auxiliary function module, and a power supply module. All modules are tightly coupled through an internal communication bus and power supply network.

[0020] The AI ​​visual recognition module, located at the top of the cabinet, acquires images of shoes and outputs characteristic data such as shoe type, material, wet / dry condition, dirt level, and occupancy status. The module primarily consists of a high-definition camera (4) and a corresponding lighting unit. The high-definition camera (4) captures panoramic images of the shoes on each shelf inside the cabinet after the door is closed. The lighting unit provides uniform illumination in low-light conditions inside the cabinet, ensuring image clarity.

[0021] The AI ​​intelligent control module, connected to the AI ​​visual recognition module and the environmental perception module, is used to fuse the shoe feature data with environmental data, module operation data, and user historical behavior data, inputting them into the AI ​​model, and inferring and outputting a processing solution. The processing solution includes at least whether to process, the processing area, sterilization intensity, ozone concentration, fan speed, and operating time. The core component of the AI ​​intelligent control module is the AI ​​main control chip 16. This chip can be a SoC with a powerful edge computing NPU (Neural Processing Unit) to meet the computing power requirements for local real-time image processing and multimodal data fusion inference. The AI ​​intelligent control module is connected to the touch LCD screen 1 and the built-in wireless communication unit to realize local visual status display, manual control, and remote interaction with a smartphone APP.

[0022] The environmental sensing module is used to collect environmental data inside the shoe cabinet. It is distributed inside the cabinet and includes a temperature and humidity sensor 14, a VOC detection sensor 12, a bacteria concentration sensor 15, and an ozone detection sensor 13. These sensors form a comprehensive environmental monitoring network, which collects micro-environmental data inside the cabinet in real time and converts it into digital signals for transmission to the AI ​​main control chip 16.

[0023] The sterilization and deodorization module includes a black titanium dioxide photocatalytic dry powder box and an ozone generator, used to receive the processing plan and perform zoned sterilization and deodorization. Specifically, the sterilization and deodorization module includes a 450nm blue light lamp 5, a transparent box of black titanium dioxide (TiO2) dry powder 8, an ozone generator 6, a PTC constant temperature heating device 9, a first fan 10, and a second fan 11. These components perform zoned and graded adaptive sterilization and deodorization operations according to the processing decisions issued by the AI ​​intelligent control module.

[0024] The safety protection module monitors the cabinet door status and ozone concentration in real time, and initiates a degradation process when the ozone concentration exceeds the standard until the ozone concentration falls below the safety threshold, at which point the cabinet door is unlocked. The safety protection module includes a door magnetic induction switch 18 and the aforementioned ozone detection sensor 13. The door magnetic induction switch 18 monitors the opening and closing status of the cabinet door 2 in real time. If the cabinet door is detected to be accidentally opened during equipment operation, or if the ozone detection sensor 13 detects an excessive concentration, the safety protection module will immediately cut off the power to the sterilization and deodorization function module via a hardware interrupt mechanism and trigger an audible and visual alarm.

[0025] The auxiliary function module includes a constant temperature dehumidification unit, an infrared sensor, a voice prompt unit, and a status visualization unit. The status visualization unit is the aforementioned touchscreen LCD 1, used to enhance the human-computer interaction experience. The power supply module is used to supply power, responsible for converting external AC power into the various DC voltages required by each module, and has overvoltage and overcurrent protection functions.

[0026] The first embodiment of the present invention will be described in detail below, focusing on the detailed configuration, working principle and technical effects of each module component.

[0027] In this embodiment, the AI ​​visual recognition module is the first step in realizing system intelligence. The high-definition camera 4 ensures coverage of all the layered, retractable shelves 3 inside the cabinet. When the door magnetic sensor switch 18 detects that the cabinet door 2 is closed, the AI ​​main control chip 16 wakes up the high-definition camera 4. To overcome the problem of insufficient light inside the cabinet, the system automatically turns on the LED supplementary light. After the high-definition camera 4 captures the image of the shoes, it transmits it to the AI ​​main control chip 16.

[0028] The AI ​​main control chip 16 internally runs a deeply optimized lightweight convolutional neural network model. This model has been pre-trained on a large number of footwear datasets and is capable of performing object detection and image classification tasks. Specifically, the image processing flow includes: first, image preprocessing; then, using an object detection algorithm to select the position of each pair of shoes, thereby determining the occupancy status of the shoe positions; next, feature extraction is performed on each selected shoe image region, and the features are fed into multiple parallel classification heads.

[0029] The first classification head identifies the shoe type, outputting results such as leather shoes, sneakers, canvas shoes, boots, high heels, and slippers. The second classification head identifies the material, outputting results such as leather, mesh, suede, rubber, and canvas. The third classification head combines features like color depth and reflectivity of the image to identify the shoe's wet or dry state. The fourth classification head assesses the level of dirtiness by analyzing texture anomalies and stains on the shoe surface. Finally, the AI ​​visual recognition module outputs a multi-dimensional feature vector of the shoe.

[0030] Placing high-definition cameras and running AI models can greatly improve the system's perception capabilities, solving the problem of traditional shoe cabinets working blindly. This allows the system to know exactly what shoes are inside the cabinet and their condition.

[0031] Meanwhile, the environmental sensing module is also working synchronously. Temperature and humidity sensor 14 monitors the temperature and relative humidity inside the cabinet in real time, which is crucial for determining the dryness of the shoes and controlling the PTC constant temperature heating device 9. VOC detection sensor 12 uses a semiconductor gas-sensitive element, enabling highly sensitive detection of volatile organic compounds such as isovaleric acid and butyric acid emitted from inside the shoes that cause foot odor. Bacterial concentration sensor 15 is used to assess the degree of microbial contamination inside the cabinet. Ozone detection sensor 13 uses electrochemical principles to accurately measure the PPM concentration of ozone inside the cabinet, ultimately outputting an environmental data vector.

[0032] The core of the AI ​​intelligent control module lies in its multimodal data fusion decision algorithm. The AI ​​main control chip 16 fuses the aforementioned shoe feature vectors, environmental data vectors, current module operation data, and user historical behavior data stored locally or in the cloud. After unified inference by the AI ​​model, the system outputs a specific processing plan. This processing plan is an instruction set containing multiple control parameters, including at least: whether to process, processing area, working mode, sterilization intensity, ozone concentration target value, fan speed (adjusting the speed of the first and second fans), and working duration.

[0033] After receiving the processing decision, the sterilization and deodorization module begins to execute specific physical and chemical treatment processes. This invention employs an innovative composite solution for sterilization and deodorization technology: a 450nm blue light lamp 5 + transparent boxed black titanium dioxide dry powder 8 + an ozone generator 6.

[0034] While traditional ultraviolet (UV) sterilization is effective, it easily causes aging, yellowing, and brittleness in polymer materials such as leather and rubber. Therefore, this invention innovatively employs a 450nm wavelength blue light lamp. 450nm blue light falls within the visible light range, causing minimal damage to shoe materials, but its photon energy is sufficient to excite a specific photocatalyst. The transparent-boxed black titanium dioxide powder is precisely such a special catalyst. Traditional white titanium dioxide can only absorb UV light, while the doped and modified black titanium dioxide has its band gap effectively reduced, enabling it to absorb 450nm visible blue light. Under blue light irradiation, electron-hole pairs are generated on the surface of the black titanium dioxide powder, reacting with water molecules and oxygen in the air to generate hydroxyl radicals and superoxide anions with extremely strong oxidizing power. These free radicals can non-selectively destroy the cell membranes of bacteria and the protein shells of viruses, while simultaneously completely oxidizing and decomposing VOC odor molecules into harmless carbon dioxide and water.

[0035] Ozone generator 6 ionizes oxygen in the air to generate ozone. Ozone is a strong oxidizing gas with excellent diffusion properties, allowing it to penetrate deep into the hard-to-reach corners of the shoe cavity for sterilization and deodorization. However, high concentrations of ozone can also accelerate material aging. Therefore, the AI ​​intelligent control module strictly controls the output concentration of ozone generator 6 based on the shoe's material.

[0036] The first fan 10 and the second fan 11 work together with the inner air outlet 7 to form a scientific aerodynamic circulation field inside the cabinet. The PTC constant temperature heating device 9 uses a positive temperature coefficient ceramic heating element, which has the characteristics of automatic constant temperature, no open flame, and safety. It is used to heat the circulating air and accelerate the evaporation of moisture inside the shoes.

[0037] During the processing, the AI ​​intelligent control module does not rigidly execute the initial decision but possesses dynamic parameter self-optimization capabilities. The system monitors the processing effect in real time through VOC detection sensor 12 and bacterial concentration sensor 15. If the odor reduction rate is slower than expected, the AI ​​main control chip 16 will dynamically increase the duration of the blue light 5 or slightly increase the fan speed; if the temperature and humidity sensor 14 shows that the humidity is decreasing too quickly, the system will immediately reduce the power of the PTC constant temperature heating device 9. This closed-loop control mechanism, which optimizes as it runs, greatly improves processing efficiency and protects the shoe materials.

[0038] The second embodiment of the present invention will be described in detail below, focusing on the specific steps of the adaptive processing scheme and control method for different footwear characteristics.

[0039] According to embodiments of this application, the AI ​​intelligent control module presets and dynamically adjusts the following exclusive strategies for footwear of different materials and conditions: For leather shoes: Genuine leather contains a large amount of collagen and oils, making it extremely sensitive to high temperatures, strong oxidants, and ultraviolet light. Therefore, the AI ​​intelligent control module outputs a treatment solution of low ozone + blue light catalysis + constant temperature + short-time sterilization. Specifically, the ozone generator 6 is limited to extremely low power operation, maintaining the ozone concentration inside the cabinet within a safe low concentration range of 0.1ppm to 0.3ppm, only used for mild surface antibacterial treatment; gentle deodorization is mainly achieved through the photocatalytic effect generated by the 450nm blue light lamp 5 irradiating the black titanium dioxide dry powder 8; the PTC constant temperature heating device 9 strictly controls the circulating air temperature between 35℃ and 40℃ to prevent the leather from drying out and cracking due to high temperatures; the overall treatment time is controlled within a short range of 10 to 15 minutes.

[0040] For wet shoes: Wet shoes are prone to mold growth and have a strong odor. The AI ​​intelligent control module outputs a treatment solution of dehumidification followed by tiered temperature control for deep sterilization. First, the system starts the PTC constant temperature heating device 9 and fans 1 and 2 at full power, raising the air temperature to 45℃ to 50℃ for rapid hot air circulation dehumidification; when the temperature and humidity sensor 14 detects that the humidity has dropped to the set threshold, the system switches to the tiered temperature control stage, lowering the air temperature to 40℃, and at the same time, the ozone generator 6 and the blue light catalytic system start at full power for deep penetration sterilization and enhanced deodorization; the treatment time can reach 30 to 45 minutes.

[0041] For suede shoes: The suede surface has fine fibers that easily attract dust and odors, and it is very sensitive to the oxidizing effects of ozone, easily leading to fading and damage to the hair follicles. Therefore, the AI ​​intelligent control module outputs an ozone-disabling solution and adopts a low-temperature photocatalytic treatment. Under this solution, the ozone generator 6 is forcibly disabled; the system relies solely on blue light lamps 5 and black titanium dioxide dry powder 8 for photocatalytic deodorization; simultaneously, the PTC constant temperature heating device 9 remains off or operates at only extremely low power, working in conjunction with a fan for low-temperature dehumidification to maximize the protection of the suede texture.

[0042] For mesh shoes: Mesh shoes offer good breathability but easily absorb a lot of sweat, resulting in a strong VOC odor. The AI ​​intelligent control module outputs a high-volume airflow circulation deep deodorization solution. The system adjusts the speed of the first fan 10 and the second fan 11 to the highest level, forming a high-volume, powerful circulation, allowing ozone and photocatalytically generated active free radicals to quickly penetrate the mesh fabric and deeply penetrate the insole and sole for full-cavity sterilization; simultaneously, a medium temperature is used for drying.

[0043] For empty shoe spaces: When the AI ​​visual recognition module determines that a certain floor or area is an empty shoe space, the AI ​​intelligent control module executes a sleep mode, closes the air deflector in the corresponding area, and does not deliver ozone and hot air to that area, thereby significantly reducing the overall energy consumption of the machine.

[0044] See Figure 4 The present invention provides an adaptive sterilization and deodorization shoe cabinet control method based on AI visual recognition, which specifically includes the following detailed steps: Step S1: System power-on pre-start. The power module supplies power to each sub-module, the AI ​​main control chip 16 completes the operating system boot and underlying hardware driver loading, and each sensor unit performs baseline calibration. At this time, the system is in a low-power standby state. When the user opens the cabinet door 2 to put in shoes and closes it again, the level state of the door magnetic sensor switch 18 flips, generating a hardware interrupt signal. This door magnetic signal immediately activates the AI ​​intelligent control module to enter the working state.

[0045] Step S2: After the AI ​​intelligent control module is activated, the high-definition camera 4 of the AI ​​visual recognition module, with the assistance of a supplementary light, acquires panoramic images of the cabinet, identifies shoe type, material, dryness / wetness status, dirt level, and shoe space occupancy status, and outputs shoe feature data. Simultaneously, the sensors of the environmental perception module acquire environmental data inside the cabinet and transmit it to the AI ​​main control chip 16 via the I2C bus.

[0046] Step S3: The AI ​​intelligent control module inputs shoe feature data, environmental data, module operation data, and user historical behavior data into the AI ​​model. The AI ​​model infers and outputs a processing solution. As mentioned earlier, a low-ozone, blue light catalysis, constant temperature, and short-time sterilization treatment solution is used for leather shoes; a dehumidification followed by graded temperature control for deep sterilization is used for wet shoes; ozone is prohibited and low-temperature photocatalysis is used for suede shoes; a high-volume circulation for deep deodorization is used for mesh shoes; and a dormancy mode is implemented for empty shoe slots.

[0047] Step S4: According to the generated processing plan, the sterilization and deodorization module activates the ozone generator 6 and blue light 5 in the corresponding shoe compartment. During execution, the AI ​​main control chip 16 dynamically adjusts the PWM duty cycle of the ozone generator to control the ozone concentration, adjusts the irradiation duration of the blue light, adjusts the speed of the first fan 10 and the second fan 11 to change the airflow, and adjusts the power of the PTC constant temperature heating device 9 to precisely control the temperature, based on real-time data fed back from the sensors.

[0048] Step S5: After the sterilization and deodorization process is completed, a certain concentration of ozone will inevitably remain inside the cabinet. At this time, the system enters the safe degradation stage. The residual ozone concentration is monitored in real time by the ozone detection sensor 13. The system starts the adsorption degradation fan to force the air inside the cabinet to circulate through the filter and activated carbon adsorption unit 17. The abundant high specific surface area micropores of activated carbon can quickly physically adsorb ozone molecules and catalytically decompose them into oxygen on its surface. Only when the ozone detection sensor 13 confirms that the ozone concentration inside the cabinet is lower than the international safety threshold will the AI ​​main control chip 16 send a command to unlock the electronic lock of the cabinet door 2 and issue a reminder that cleaning is complete and the cabinet can be opened safely via the touch screen 1, voice module, and the bound smartphone APP.

[0049] If, during S5 or any operating phase, the door magnetic sensor switch 18 detects that the cabinet door has been forcibly opened, or that the ozone concentration has exceeded the standard abnormally, the system will trigger the highest priority safety interruption, immediately cutting off the power to the ozone generator 6 and the blue light 5, and issuing a rapid buzzer alarm to completely eliminate safety hazards.

[0050] The third embodiment of the present invention will be described in detail below, focusing on the extended functions of the AI ​​user habit learning and prediction module and the auxiliary degradation component.

[0051] According to embodiments of this application, the system also includes an AI user habit learning and prediction module. This module mainly runs in the background of the AI ​​main control chip 16, or collaborates with a cloud server via a wireless network. Its core is based on time series algorithms, such as Long Short-Term Memory (LSTM) networks or Autoregressive Integral Moving Average (ARIMA) models.

[0052] In daily use, the system records data from every user interaction, including: the timestamp of opening the cabinet door each day, the types of shoes put in / taken out, and the frequency of use of specific shoe compartments. After a period of data accumulation (such as one to two weeks), the time series algorithm can uncover patterns in the user's shoe-wearing habits. For example, the algorithm might learn that user A takes out their leather shoes from the second shelf at 8:00 AM every weekday and puts them back in at 6:30 PM; while on weekends, they take out their sneakers at 10:00 AM.

[0053] Based on these learned patterns, the AI ​​user habit learning and prediction module can predict the exact time a user will next use the shoe cabinet. To provide an extremely seamless experience, the system automatically wakes from sleep mode 10 to 20 minutes before the predicted usage time and executes sterilization and deodorization steps S2 to S5. For example, at 7:40 AM, the system automatically performs a rapid blue light catalytic deodorization and constant temperature heating (shoe warming function) on the leather shoes. When the user opens the shoe cabinet at 8:00 AM, not only have the shoes been sterilized and deodorized, but the internal temperature is also comfortable, greatly improving the user's wearing comfort and intelligent experience.

[0054] Furthermore, according to the embodiments of this application, the sterilization and deodorization functional module also includes a detachable filter and an activated carbon adsorption unit 17. This unit is typically installed at the inner air outlet 7 or a dedicated return air outlet. The detachable filter is mainly used to intercept large particulate impurities such as dust and hair that are stirred up during air circulation, preventing them from adhering to the surfaces of the blue light lamp 5, ozone generator 6, or sensors and affecting their performance. The activated carbon adsorption unit uses modified honeycomb activated carbon, which has an extremely high specific surface area and a well-developed microporous structure. After the ozone generator 6 stops working, the system enters the degradation stage, and the fan drives the air containing residual ozone through the activated carbon adsorption unit 17. Activated carbon can not only physically adsorb ozone, but its surface functional groups can also act as a catalyst, significantly reducing the activation energy of ozone decomposition and accelerating the conversion of ozone to oxygen. This combination of physical adsorption and chemical catalysis significantly shortens the ozone residual degradation time from several hours of natural decay to within a few minutes, greatly improving the equipment's turnover rate and safety. Meanwhile, the unit is designed with a detachable structure. The AI ​​main control chip 16 will remind users to replace the filter and activated carbon regularly through the APP based on the cumulative running time and resistance changes of the fan, ensuring the long-term stable operation of the system.

[0055] All embodiments described above are not limited to the embodiments shown, and can of course be combined and applied in combination.

[0056] While the present invention has been described with reference to the embodiments shown in the accompanying drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent embodiments can be made therefrom. Therefore, the true scope of protection of the present invention should be determined by the technical spirit of the appended claims.

[0057] The specific implementations described in the embodiments are examples and do not limit the scope of the embodiments in any way. Furthermore, unless specifically mentioned as "essential," "important," etc., they may not be essential components for applying the present invention.

Claims

1. An adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition, characterized in that, include: The AI ​​visual recognition module is located at the top of the cabinet and is used to acquire images of shoes and output characteristic data of shoes, including shoe type, material, dry and wet condition, dirt level, and shoe space occupancy status. The environmental sensing module is used to collect environmental data inside the shoe cabinet; The AI ​​intelligent control module is connected to the AI ​​visual recognition module and the environmental perception module. It is used to integrate the footwear feature data with the environmental data, module operation data and user historical behavior data into the AI ​​model, and infer and output a processing solution. The processing solution includes at least whether to process, the processing area, the sterilization intensity, the ozone concentration, the fan speed and the working time. The sterilization and deodorization module includes a black titanium dioxide photocatalytic dry powder box and an ozone generator, which are used to receive the processing scheme and perform zoned sterilization and deodorization; The safety protection module is used to monitor the cabinet door status and ozone concentration in real time, and to start the degradation process when the ozone concentration exceeds the standard until the ozone concentration is lower than the safety threshold and then unlock the cabinet door.

2. The adaptive sterilization and deodorization shoe cabinet system according to claim 1, characterized in that, It also includes auxiliary function modules, which include constant temperature dehumidification, infrared sensing, voice prompts and status visualization units.

3. The adaptive sterilization and deodorization shoe cabinet system according to claim 1, characterized in that, The AI ​​intelligent control module provides a treatment scheme for leather shoes, which uses low ozone, blue light catalysis, constant temperature and short-term sterilization; for wet shoes, it provides a treatment scheme for first dehumidifying and then using graded temperature control for deep sterilization; for suede shoes, it uses a treatment scheme that prohibits ozone and uses low-temperature photocatalysis; for mesh shoes, it provides a treatment scheme for high-volume circulation for deep deodorization; and for empty shoe slots, it provides a sleep mode.

4. The adaptive sterilization and deodorization shoe cabinet system according to claim 1, characterized in that, It also includes an AI user habit learning and prediction module, which is used to learn the time and frequency of users wearing shoes based on time series algorithms, predict the next time of use, and complete sterilization and deodorization in advance.

5. The adaptive sterilization and deodorization shoe cabinet system according to claim 1, characterized in that, The sterilization and deodorization module also includes a detachable filter and an activated carbon adsorption unit, which are used to accelerate the degradation of residual ozone after the ozone generator stops working.

6. The adaptive sterilization and deodorization shoe cabinet system according to claim 1, characterized in that, It also includes a power supply module for powering the circuit.

7. A control method for an adaptive sterilization and deodorization shoe cabinet system based on AI visual recognition, characterized in that, The method includes; S1: System power-on pre-start, door magnetic signal activates AI intelligent control module; S2: The AI ​​visual recognition module collects images of shoes, identifies shoe type, material, dryness / wetness status, dirt level, and shoe space occupancy status, and outputs shoe feature data; the environmental perception module simultaneously collects environmental data inside the cabinet. S3: The AI ​​intelligent control module integrates the footwear feature data with the environmental data, module operation data and user historical behavior data, inputs them into the AI ​​model, and infers and outputs a processing solution. S4: According to the processing scheme, the sterilization and deodorization module will activate the ozone generator and blue light in the corresponding shoe compartment, and adjust the ozone concentration, blue light duration, fan speed and PTC heating power. S5: Real-time monitoring of residual ozone concentration via an ozone concentration sensor. If the concentration exceeds the safety threshold, the cabinet door will be unlocked and a voice and app notification will be issued.

8. The control method according to claim 7, characterized in that, Step S3 provides a treatment scheme for leather shoes, which uses low ozone, blue light catalysis, constant temperature and short-term sterilization; for wet shoes, it provides a treatment scheme for dehumidification followed by graded temperature control for deep sterilization; for suede shoes, it uses a treatment scheme that prohibits ozone and uses low-temperature photocatalysis; for mesh shoes, it provides a treatment scheme for high-volume circulation for deep deodorization; and for empty shoe slots, it provides a sleep mode.

9. The control method according to claim 7, characterized in that, The method further includes: learning the user's shoe-wearing time periods and frequencies based on time series algorithms, predicting the next usage time, and executing steps S2-S5 10-20 minutes in advance.

10. The control method according to claim 7, characterized in that, In step S5, when the cabinet door is opened or the ozone concentration exceeds the standard, the power supply to the ozone generator and blue light lamp is immediately cut off and an alarm is triggered.

Images