Intelligent food heat detection dinner plate
By using a smart food calorie detection plate, which employs an AI camera and weight sensor to identify food types and calculate calories, the convenience and accuracy of calorie management in family meals are solved, achieving efficient diet management.
Patent Information
- Application Number
- CN202511451534.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-11
- Publication Date
- 2026-01-30
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the current technology, it is difficult for people to conveniently and accurately assess the calories they consume at each meal in their daily family diet, and there is a lack of effective tools for calorie management.
A smart food calorie detection plate was designed, which uses an AI recognition camera and supplementary LED lights to work together, combined with a weight sensor and SD card module. It identifies food types and calculates calories through a pre-trained classification model, and uses an OLED display and button design to achieve intuitive data display and storage.
It improves the accuracy and stability of food identification, enhances data storage and traceability capabilities, and increases the convenience and intuitiveness of users' diet management, helping users better control their calorie intake.
Smart Images
Figure CN121421327A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of healthy eating technology, specifically to a smart food calorie detection plate. Background Technology
[0002] With the increasing health awareness of the public, accurate monitoring of daily dietary calorie intake has become an important part of health management. In order to significantly improve the public's awareness and skills in weight management and further reduce the individual obesity rate, dietary control has gradually become an important means of chronic disease prevention, weight loss planning and health management. More and more office workers and the elderly are paying attention to dietary control.
[0003] However, most people currently find it difficult to accurately assess their calorie intake at each meal, especially in daily family meals, where convenient and accurate tools for calorie management are lacking. Therefore, to solve this technical problem, it is necessary to design a device that can automatically identify food types and calculate calorie intake, which would have significant practical value for daily calorie intake and weight management. Summary of the Invention
[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a smart food calorie detection plate, which has the advantages of convenient and efficient dietary calorie monitoring, and solves the problem of difficulty in self-monitoring of calories and self-control of food intake in daily life as described in the background technology.
[0005] (II) Technical Solution To achieve the above-mentioned goal of convenient and efficient monitoring of dietary calories, the present invention provides the following technical solution: a smart food calorie detection plate, including a plate base, wherein a food calorie detection module is provided on the inner side of the plate base; The food calorie detection module includes a circular slot on the top left side of the plate holder, a weight sensor and a sensing plate fixedly installed inside the circular slot, an OLED display and a button digital sensor embedded in the inner top wall of the plate holder, and a main control star core board fixedly installed in the inner bottom wall of the plate holder.
[0006] Preferably, there are two sensing elements, which are located on the inner bottom wall and the opening of the circular groove, respectively. The weight sensor is connected to the opposite side of the two sensing elements, and the detection accuracy range of the weight sensor is set to 5KG.
[0007] Preferably, an AD conversion module is installed on one side of the plate holder, and the weight sensor is connected to the analog input pin of the AD conversion module via a signal output interface.
[0008] Preferably, an L-shaped vertical plate is fixedly connected to one side of the plate holder, and an AI recognition camera and a supplementary LED light are fixedly installed on the inner top wall of the vertical plate. The bottom of the AI recognition camera is provided with a CMOS sensing lens, which is located directly above one of the sensing sheets.
[0009] Preferably, the built-in processor of the AI recognition camera is equipped with a classification model, whose classification information includes carbohydrates, vegetables and meats, for training and analysis of food images from the CMOS sensor lens. The classification model is equipped with a food feature library corresponding to the classification information, which is used to match and compare the extracted food image features with the food feature library to identify the food type.
[0010] Preferably, the OLED display screen is an I2C monochrome screen structure, and the button digital sensor is a normally open digital key structure, respectively configured as button one and button two. The OLED display screen and the button digital sensor are respectively connected to the main control star core board via input pins and output pins.
[0011] Preferably, the plate holder and the SD card module are connected to the main control star core board via an SPI interface to realize data storage writing and interactive reading, and the plate holder is connected to the power module.
[0012] Preferably, the plate holder and the SD card module are connected to the main control star core board via an SPI interface to realize data storage writing and interactive reading, and the plate holder is connected to the power module.
[0013] (III) Beneficial Effects Compared with existing technologies, the present invention provides a smart food calorie detection plate, which has the following beneficial effects: 1. This intelligent food calorie detection plate improves recognition accuracy and stability by employing an AI recognition camera and supplementary LED lights working together. It extracts food features through a pre-trained classification model, combined with intelligent supplementary lighting control to improve recognition accuracy. Then, it works with an AD conversion module and a zero-point error compensation algorithm with preset calibration values to effectively improve the accuracy of food weight detection.
[0014] 2. This intelligent food calorie detection plate uses an SD card module for persistent storage of detection data, and employs SPI interface communication and file system management to enhance data storage and traceability capabilities. Then, through a dual-button design and display screen, it displays the current food information and total calorie data in real time, improving the intuitiveness and convenience of users' diet management. Attached Figure Description
[0015] Figure 1 This is a top-view schematic diagram of the intelligent detection plate of the present invention; Figure 2This is a schematic diagram of the intelligent detection plate of the present invention; Figure 3 for Figure 1 A cross-sectional view from a specific perspective; Figure 4 This is a front view of the vertical plate structure of the present invention; Figure 5 This is a schematic diagram of the control flow of the present invention; Figure 6 This is a schematic diagram of the wiring structure of the module of the present invention; Figure 7 This is a schematic diagram of part of the control program code of the present invention.
[0016] In the diagram: 1. Plate holder; 2. Food calorie detection module; 201. Circular slot; 202. Weight sensor; 203. Sensor sheet; 204. OLED display; 205. Button digital sensor; 206. Main control star core board; 3. AD conversion module; 4. Vertical plate; 5. AI recognition camera; 6. Fill light LED; 7. SD card module; 8. Programming port; 9. Power module. Detailed Implementation
[0017] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] Example 1 In this embodiment, the wiring process for the hardware portion of the intelligent food calorie detection plate includes: 1) The 5V output port of the power module 9 is connected to the 5V pin of the main control star core board 206 to provide main power to the main control star core board 206; it is connected to the VCC pin of the HX711 weight sensor 202 to power the weighing conversion module; and it is connected to the VCC pin of the AI recognition camera 5 to power the AI recognition module. The power supply module 9 is connected to the VCC pin of the OLED display 204 to power the display module; it is connected to the VCC pin of the button digital sensor 205 to power the interaction module; it is connected to the VCC pin of the fill light LED to power the light compensation module; it is connected to the VCC pin of the SD card module 7 to power the data storage module; the GND output of the power supply module 9 and the GND pin of all components are used for unified grounding to form a current loop. 2) The signal output terminal of the weight sensor 202 is connected to the AIN1 (analog signal input) terminal of the AD conversion module 3 to convert the weight into a weak analog signal; the VCC terminal of the AD conversion module 3 is connected to the 5V output terminal of the power supply module 9, and its GND terminal is connected to the GND output terminal for grounding. The SCK clock line of the AD conversion module 3 is connected to the main control star core board 206 to read weight data synchronously through the clock signal. Its DOUT data line is connected to the main control star core board 206 to realize the transmission of the converted digital weight data. 3) The CMOS sensor lens at the bottom of the AI recognition camera 5 is directly facing the sensor 203 at the groove 201 to ensure that food can be completely placed into the lens recognition range; the VCC pin of the AI recognition camera 5 is connected to the 5V output terminal of the power module 9 through a red DuPont wire, and its GND pin is connected to the GND output terminal of the power module 9 through a black DuPont wire to achieve power supply and grounding. The AI recognition camera 5 communicates with the main control star core board 206 via an I2C interface. Its SCL (clock line) is connected to the I2C_SCL pin of the main control star core board 206 through a yellow DuPont wire, and its SDA (data line) is connected to the I2C_SDA pin of the main control star core board 206 through a white DuPont wire. This is used to transmit the recognized food category ID (set as carbohydrates=1, vegetables=2, meat=3) to the main control star core board 206. Meanwhile, the reset pin (RESET) of the AI recognition camera 5 is connected to the GPIO12 pin of the main control star core board 206 through a gray DuPont wire for module reset in abnormal situations, ensuring recognition stability; 4) The supplementary LED light 6 and the AI recognition camera 5 are fixed side by side on the inner top wall of the vertical plate 4, located on the side of the CMOS sensor lens, to ensure that the supplementary lighting range covers the food placement area of the circular slot 201; the VCC pin of the supplementary LED light 6 is connected to the 5V output terminal of the power module 9 through a red DuPont wire, and the GND pin is connected to the GND output terminal of the power module 9 through a black DuPont wire to obtain the working power. The control terminal of the supplementary LED 6 is connected to the GPIO13 pin of the main control star core board 206 via a purple DuPont wire. The main control star core board 206 controls its on / off state via a level signal. When the main control star core board 206 detects that the brightness of the image transmitted by the AI recognition camera 5 is <50cd / m², GPIO13 outputs a high level, triggering the auxiliary light LED 6 to light up; Once the food category is successfully identified, GPIO13 outputs a low level to turn off the auxiliary LED 6 and reduce power consumption. 5) The SD card module 7 is fixedly installed on the inner rear wall of the plate holder 1 and uses the SPI interface to communicate with the main control star core board 206. The VCC pin of the SD card module 7 is connected to the 5V output terminal of the power module 9 through a red DuPont wire, and the GND pin is connected to the GND output terminal of the power module 9 through a black DuPont wire to ensure stable power supply to the module. The SCK (SPI clock) interface of the SD card module 7 is connected to the main control star core board 206 to synchronize the data read and write clock signal of the SPI interface; its MOSI (data transmit) end is connected to the main control star core board 206 to write detection data (including time, weight, heat, etc.) to the SD card; its MISO (data receive) end is connected to the main control star core board 206, and the SD card module 7 feeds back the storage status or reads historical data to the main control star core board 206.
[0019] 6) The OLED display 204 is an I2C monochrome screen structure, which is embedded in the inner top wall of the plate holder 1; its VCC pin is connected to the 5V output terminal of the power module 9 through a red DuPont wire, and its GND pin is connected to the GND output terminal of the power module 9 through a black DuPont wire to obtain the working power. The I2C pins of the OLED display 204 are shared with the I2C interface of the main control star core board 206 (to reduce pin usage). Its SCL (clock line) is connected to the I2C_SCL pin of the main control star core board 206 through a blue DuPont wire (shared with the AI recognition camera 5), and the SDA (data line) is connected to the I2C_SDA pin of the main control star core board 206 through a green DuPont wire (shared with the AI recognition camera 5) to receive display data (current food information, total weight, total calories, prompt information, etc.) sent by the main control star core board 206.
[0020] The button digital sensor 205 is a normally open digital key structure, with two in total (the first is the "Switch Food" button and the second is the "End Detection" button), which is embedded in the inner top wall of the plate holder 1 (on the side near the OLED display 204 for easy user operation). The VCC pins of the button digital sensor 205 are all connected in parallel to the 5V output terminal of the power module 9 through red DuPont wires, and the GND pins are all connected in parallel to the GND output terminal of the power module 9 through black DuPont wires to achieve unified power supply. The signal output of button one (switch food) is connected to the GPIO4 pin of the main control star core board 206 through a blue DuPont wire; the signal output of button two (end detection) is connected to the GPIO2 pin of the main control star core board 206 through a green DuPont wire. When not pressed, the main control star core board detection pin is at a high level; when pressed, the pin is grounded and becomes low, triggering the corresponding operation (switching food / ending detection).
[0021] Example 2 In this embodiment, the power module 9 adopts a portable solution of 18650 lithium battery pack + TP4056 charging module to convert 220V AC power or lithium battery 3.7V voltage into stable 5V DC voltage, providing consistent power supply to all components and avoiding data acquisition errors caused by voltage fluctuations. The weight sensor 202 is a resistance strain gauge sensor, consisting of four strain gauges forming a Wheatstone bridge. When food is placed on the sensor, the strain gauges deform under pressure, causing a change in resistance. The bridge outputs a weak analog voltage signal proportional to the weight. This weak analog signal is then amplified 1000 times by the built-in instrumentation amplifier and converted from analog to digital by the AD conversion module 3 before being transmitted to the main control star core board 206 for identification as digital weight data. The star core board sends a clock signal through P11 (SCK), synchronously reads digital data from P10 (DOUT), and then uses the calibration value 1992 to offset the sensor zero-point error. The AI recognition camera 5 incorporates a built-in CMOS image sensor and deploys pre-trained models for three food categories: carbohydrates, vegetables, and meat. It captures food images through the lens, and the CMOS sensor converts light signals into a digital pixel matrix. The AI processor then calls the pre-trained models to extract key image features (the "particle texture" of carbohydrates, the "green spectrum percentage" of vegetables, and the "fiber structure" of meat). Category matching compares the extracted features with the "feature library" of the three food categories within the model, outputting the category ID with the highest matching degree (e.g., carbohydrates = 1, vegetables = 2, meat = 3). Data transmission transmits the category ID to the star core board in real time via an I2C interface. If the ambient light is insufficient (the star core board determines this based on the brightness of the camera image), a supplementary lighting module is triggered.
[0022] The supplementary lighting module consists of four 0.5W white LEDs forming a supplementary lighting array, controlled by the GPIO13 pin of the star core board. It reads the brightness value of the image transmitted by the camera in real time. If the brightness is insufficient, the star core board outputs a high level to GPIO13 to trigger the LEDs to light up (the supplementary lighting intensity is fixed to avoid strong light reflection). When the camera successfully identifies the food category, the star core board outputs a low level to turn off the supplementary lighting and reduce power consumption.
[0023] Data storage unit: The core board sends an initialization command to the SD card module to check whether the SD card is properly mounted (such as recognizing the SD card capacity and file system); after each test, the core board writes "test time, food category list, single weight / calorie, total weight / calorie" to the SD card. If the user needs to view the history, the core board can be triggered by pressing a button to read the data of the specified date from the SD card and transmit it to the OLED display.
[0024] The core board serves as the central hub for data processing and command issuance, executing the entire process of "receiving-calculating-judging-commanding". It receives the camera category ID via I2C, the weight data of HX711 via P10 / P11, and the button signals via GPIO. Call the preset calorie parameters per 100 grams (carbohydrate ≈ 4 kcal / g, vegetables = 0.25 kcal / g, meat = 1.7 kcal / g), and calculate according to the formula: Calorie of a single food = (food weight ÷ 100) × total calories per 100 grams of the corresponding category = cumulative calorie value of all single foods; Threshold judgment: Compare the total heat with the preset threshold (700kcal) and generate a prompt signal of "heat exceeds standard" or "within control range"; send "weight / heat / prompt information" display command to OLED, send "light up / light down" command to fill light module, and send "data write" command to SD card module.
[0025] When the button module is not pressed, the detection pin of the star core board is at a high level (achieved through a pull-up resistor); when pressed, the pin is grounded and the level changes to a low level. Button 1 (Switch): Triggers the star core board to "clear the current food data cache" and enter the next food detection state; Button 2 (End): Triggers the star core board to "stop data acquisition, calculate total calories, and generate prompt information".
[0026] After receiving instructions from the star core board, the OLED display 204 converts the digital signal into pixel light emission signals. During detection, it displays "current food category + weight + calories"; after detection, it displays "total weight + total calories + prompt information".
[0027] In summary, the specific workflow of this intelligent food calorie detection plate includes: 1) Device initialization (automatically executed after startup): Power module 9 is powered on, providing 5V voltage to all components. The main control star core board 206 starts up and automatically initializes each module: AI-powered camera recognition: Load the "carbohydrate / vegetable / meat" pre-trained model, switch to the "object classification" algorithm, and calibrate the lens; AD conversion module: Loads the document preset calibration value 1992 to offset the sensor zero point error and enters the "weight detection ready" state; LED fill light: Off by default. The star core board reads the initial brightness value of the camera and sets the fill light trigger threshold. SD Card Module: The core board detects the SD card mounting status. If it is normal, it creates a test file for the day. OLED display screen: Displays the initialization interface (e.g., "Food calorie detection plate, ready! Please place food"). Button module: Enters "waiting for operation" state, and the star core board detects changes in button level in real time.
[0028] 2) Single Food Detection (Core Execution Step): The user places a single type of food on the weight sensor; the star core board reads the image brightness through the camera. If the brightness is insufficient, GPIO13 outputs a high level, illuminating the device through the supplementary lighting module until the brightness meets the standard; if the brightness is sufficient, it directly enters the recognition process. The AI-powered camera captures food images, extracts features, and matches categories. The category ID (e.g., carbohydrates = 1, vegetables = 2, meat = 3) is transmitted to the core board via I2C. The HX711 module detects the food weight and transmits the converted digital weight data to the core board via P10 or P11. The core board calculates the calories of a single food meal and temporarily stores "category + weight + calories" in RAM. The OLED display updates in real time, showing the current food information ("Current: carbohydrates, weight: 200g, calories: 260kcal").
[0029] 3) Multiple food switching or detection end (user interaction): The star core board detects the button signal. If button 1 (switching) is detected, the star core board clears the "current food data" in the RAM, the fill light module is turned off (if it is already lit), and the OLED displays "Please place the next food". If button 2 (End) is detected pressed, stop the detection; if no button is pressed, maintain the current display and wait for user operation.
[0030] 4) Output and storage of test results (final stage): The core board calculates the total weight (cumulative weight of all individual tests) and the total heat (cumulative heat of all individual tests). The core board determines whether the total heat exceeds the preset threshold. If the total heat is >700kcal, a "Heat Exceeds Standard" prompt is generated; if the total heat is ≤700kcal, a "Within Control Range" prompt is generated. The final result is displayed on the screen ("Test complete! Total weight: 350g, total calories: 515kcal, you have controlled your calories very well!"). The star core board writes the test data (test time, category list, individual data, total data) to the SD card via SPI (stored to the file of the day). The supplementary lighting module (if already lit) is turned off, and the star core board waits for the user's next operation, or automatically enters standby mode after 30 seconds to reduce power consumption.
[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0032] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A food calorie intelligent detection plate, comprising a plate base (1), characterized in that: The inner side of the dinner plate seat (1) is provided with a food calorie detection module (2); The food calorie detection module (2) comprises a circular slot (201) formed on the top left side of the dinner plate seat (1), a weight sensor (202) fixedly installed on the inner side of the circular slot (201), an inductive sheet (203) fixedly installed on the inner side of the circular slot (201), an OLED display screen (204) and a button digital sensor (205) embedded on the inner top wall of the dinner plate seat (1), and a main control star core board (206) fixedly installed on the inner bottom wall of the dinner plate seat (1).
2. The food calorie smart detection dinner plate of claim 1, wherein: The number of the inductive sheets (203) is two, and each is located on the inner bottom wall and the slot of the circular slot (201), the weight sensor (202) is connected to the opposite side of the two inductive sheets (203), and the detection accuracy range of the weight sensor (202) is 5KG.
3. The smart calorie detection plate of claim 1, wherein: An AD conversion module (3) is installed on one side of the dinner plate seat (1) close to the weight sensor (202), and the weight sensor (202) is connected to the analog input pin signal of the AD conversion module (3) through a signal output interface.
4. The smart calorie detection plate of claim 1, wherein: A vertical plate (4) of L-shaped structure is fixedly connected to one side of the back of the dinner plate seat (1) close to the circular slot (201), an AI recognition camera (5) and a supplementary light LED lamp (6) are fixedly installed on the inner top wall of the vertical plate (4), and a CMOS sensing lens is arranged at the bottom of the AI recognition camera (5) and located directly above one of the inductive sheets (203).
5. The smart calorie detection plate of claim 4, wherein: A classification model is deployed in the built-in processor of the AI recognition camera (5), the classification information of the classification model includes carbohydrates, vegetables and meat, and the classification model is used for training and analyzing food images from the CMOS sensing lens, a food feature library corresponding to the classification information is arranged in the classification model, and the food feature library is used for matching and comparing according to the extracted food image features and the food feature library to identify the food type.
6. The smart calorie detection plate of claim 1, wherein: The OLED display screen (204) is an I2C monochrome screen structure, the button digital sensor (205) is a normally open digital key structure, and the button digital sensor (205) is provided as key one and key two, and the OLED display screen (204) and the button digital sensor (205) are respectively connected to the main control star core board (206) through input pins and output pins.
7. The smart calorie detection plate of claim 1, wherein: An SD card module (7) is arranged on the inner rear wall of the dinner plate seat (1), the SD card module (7) is connected to the main control star core board (206) through an SPI interface, and is used for realizing data storage writing and interactive reading, a burning port (8) and a power module (9) are arranged on the inner right wall of the dinner plate seat (1).
8. The smart calorie detection plate of claim 7, wherein: An SD card module (7) is arranged on the inner rear wall of the dinner plate seat (1), the SD card module (7) is connected to the main control star core board (206) through an SPI interface, and is used for realizing data storage writing and interactive reading, a burning port (8) and a power module (9) are arranged on the inner right wall of the dinner plate seat (1).