Control methods for intelligent food preservation equipment, intelligent food preservation equipment and intelligent control system

Abstract

This application relates to a control method for intelligent food preservation equipment, intelligent food preservation equipment, and intelligent control system. The method includes acquiring food type data, spectral data of the food surface, image data of the food, and current temperature and humidity data of the target area within the intelligent food preservation equipment; performing data fusion based on the spectral data, image data, temperature and humidity data, and food type data to obtain multi-dimensional fusion feature data; obtaining the current freshness data of the target food based on the multi-dimensional fusion feature data and a preset correspondence; and dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target food. The technical solution provided in this application can delay the spoilage process of the food in the target area and improve the accuracy of food preservation control.

Description

Technical Field

[0001] This application relates to the technical field of smart refrigerators, and in particular to control methods, smart preservation equipment and smart control systems for smart preservation equipment. Background Technology

[0002] With the improvement of people's quality of life and the promotion and popularization of technologies such as the Internet, big data, artificial intelligence, and voice interaction, more and more traditional lifestyles are gradually changing. The use of home appliances has gradually moved towards intelligence, bringing more convenience to users while the functions of various home appliances are becoming more diversified.

[0003] However, the preservation methods used in existing refrigerators have limitations, which can easily lead to a decrease in the freshness of food. If users do not take the food out in time, as the storage time increases, the food is very likely to suffer from nutrient loss, flavor decay, texture changes, and even spoilage. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a technical solution for controlling an intelligent preservation device, an intelligent preservation device, and an intelligent control system. Specifically, this application fuses spectral data, image data, temperature and humidity data, and food type data to obtain multi-dimensional fused feature data. Based on the multi-dimensional fused feature data and a preset correspondence, the current freshness data of the target food can be determined. Furthermore, based on the current freshness data of the target food, the temperature and humidity data in the target area can be dynamically adjusted. The technical solution provided by this application can slow down the spoilage process of the food in the target area and improve the accuracy of food preservation control.

[0005] On one hand, embodiments of this application provide a control method for an intelligent food preservation device, the method comprising: Acquire the food type data of the target food in the target area within the intelligent preservation device, the spectral data of the surface of the target food, the image data of the target food, and the current temperature and humidity data of the target area; Data fusion is performed based on the spectral data, the image data, the temperature and humidity data, and the food type data to obtain multidimensional fused feature data; Based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target ingredient is obtained. The preset correspondence is a mapping relationship between the reference multidimensional fusion feature data and the reference freshness data. The freshness data represents the degree of spoilage of the target ingredient. Based on the current freshness data of the target ingredient, the temperature and humidity data in the target area are dynamically adjusted.

[0006] Furthermore, the data fusion based on the spectral data, the image data, the temperature and humidity data, and the food type data to obtain multidimensional fused feature data includes: Data extraction and processing are performed based on the spectral data to obtain multiple spectral characteristic wavelengths associated with the target food ingredient; Based on the spectral reflectance characteristics corresponding to each of the multiple spectral characteristic wavelengths, a spectral feature vector is determined. Based on the image data, multiple grayscale texture features corresponding to the target food ingredient are determined; Based on the aforementioned multiple grayscale texture features, an image feature vector is determined; The multidimensional fusion feature data is obtained by stitching together the spectral feature vector, the image feature vector, the temperature and humidity data, and the food type data.

[0007] Furthermore, the step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: If the current freshness data of the target ingredient is less than the first freshness threshold, the current temperature and humidity data in the target area shall be maintained.

[0008] Furthermore, the step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: When the current freshness data of the target ingredient is greater than or equal to a first freshness threshold and the current freshness data of the target ingredient is less than or equal to a second freshness threshold, a first freshness deviation is determined based on the first freshness threshold and the freshness data. Both the first freshness threshold and the second freshness threshold correspond to the type of the target ingredient, and the second freshness threshold is greater than the first freshness threshold. Based on the first freshness deviation and the first preset relationship, the first temperature and humidity adjustment data is determined. The first preset relationship is the mapping relationship between the food type, the first freshness deviation and the first temperature and humidity adjustment data. Based on the first temperature and humidity adjustment data and the current temperature and humidity data in the target area, the first temperature and humidity control data is obtained; Adjust the temperature and humidity in the target area to the first temperature and humidity control data.

[0009] Furthermore, the step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: If the current freshness data of the target ingredient is greater than the second freshness threshold, a second freshness deviation is determined based on the second freshness threshold and the freshness data. The second freshness threshold corresponds to the type of ingredient of the target ingredient. Based on the second freshness deviation and the second preset relationship, the second temperature and humidity adjustment data are determined. The second preset relationship is the mapping relationship between the food type, the second freshness deviation and the second temperature and humidity adjustment data. Based on the second temperature and humidity adjustment data and the current temperature and humidity data in the target area, the second temperature and humidity control data is obtained; The temperature and humidity in the target area are adjusted to the second temperature and humidity control data, and a freshness warning message is generated. The second temperature and humidity control data is greater than the first temperature and humidity control data.

[0010] Furthermore, obtaining the current freshness data of the target ingredient based on the multi-dimensional fused feature data and the preset correspondence includes: The multidimensional fusion feature data is input into the freshness prediction model to predict the freshness, thereby obtaining the current freshness data of the target ingredient. The freshness prediction model is obtained by training a preset model based on the reference multidimensional fusion feature data and the reference freshness data.

[0011] Furthermore, before acquiring the food type data of the target food in the target area within the intelligent preservation device, the spectral data of the surface of the target food, the image data of the target food, and the current temperature and humidity data in the target area, the method further includes: Acquire reference food type data, reference spectral data of the food surface, reference image data of the food, and reference temperature and humidity data of the food in the simulation chamber; The food ingredients in the simulation chamber were subjected to destructive testing, and reference freshness data of the food ingredients after the destructive testing was obtained; Based on the reference spectral data, the reference image data, the reference temperature and humidity data, and the reference food type data, data fusion is performed to obtain reference multidimensional fusion feature data; The freshness prediction model is obtained by training the model based on the reference multidimensional fusion feature data and the reference freshness data, and the preset correspondence is realized based on the freshness prediction model.

[0012] Further, determining the spectral feature vector based on the spectral reflectance characteristics corresponding to each of the plurality of spectral feature wavelengths includes: The spectral reflectance features corresponding to each of the multiple spectral feature wavelengths are normalized and interpolated to obtain the spectral feature vector.

[0013] On the other hand, embodiments of this application provide an intelligent preservation device, which includes a controller for executing the control method of the intelligent preservation device as described above.

[0014] On the other hand, embodiments of this application provide an intelligent control system, which includes a server and an intelligent preservation device. The server is communicatively connected to the intelligent preservation device and is capable of data interaction with the intelligent preservation device. The intelligent preservation device is used to execute the control method of the intelligent preservation device as described above.

[0015] Implementing this application will have the following beneficial effects: This application fuses spectral data, image data, temperature and humidity data, and food type data to obtain multidimensional fusion feature data. Based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target food can be determined. Then, based on the current freshness data of the target food, the temperature and humidity data in the target area can be dynamically adjusted. The technical solution provided by this application can delay the spoilage process of the food in the target area and improve the accuracy of food preservation control. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic diagram of an implementation environment provided for an embodiment of this application; Figure 2 A flowchart illustrating a control method for an intelligent food preservation device provided in an embodiment of this application; Figure 3 A schematic diagram of the structure of a control device for an intelligent food preservation device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0020] Please see Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application, such as... Figure 1 As shown, the implementation environment may include intelligent preservation device 01 and server 02. In practical applications, intelligent preservation device 01 and server 02 can be connected through wireless or wired communication to realize interaction between intelligent preservation device 01 and server 02.

[0021] In this embodiment, server 02 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Further, server 02 can include physical devices, specifically network communication units, processors, and memory, etc., and software running on the physical device, specifically applications, etc. For example, server 02 can be a mobile client, specifically a mobile phone, tablet, or electronic device with applications. Specifically, server 02 can store data sent by intelligent preservation device 01, and further, can send remote services to intelligent preservation device 01. For example, server 02 can provide services such as preservation mode selection and online upgrades of the preservation program.

[0022] In some embodiments, the intelligent preservation device 01 can be an intelligent refrigerator. The intelligent preservation device 01 is equipped with a spectral sensor, a camera device, and a temperature and humidity sensor. The spectral sensor is used to acquire food type data and spectral data of the food surface in the target area. The camera device is used to acquire image data of the target food. The temperature and humidity sensor is used to detect temperature and humidity data in the target area. It should be noted that the intelligent preservation device 01 includes multiple independently controlled areas, and each area is equipped with a spectral sensor, a camera device, and a temperature and humidity sensor to more accurately acquire data from different zones.

[0023] In addition, it should be noted that, Figure 1 The diagram shown is merely a schematic representation of an implementation environment, which may include more or fewer nodes; this application makes no limitation on this.

[0024] Figure 2 This is a flowchart illustrating a control method for an intelligent food preservation device provided in an embodiment of this application. This specification provides the operational steps of the method described in the embodiments or flowchart, but based on conventional or non-inventive methods, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In practical applications, the methods shown in the embodiments or accompanying drawings can be executed sequentially or in parallel. Specifically, as shown... Figure 2 As shown, the control method for this intelligent preservation device may include the following steps: S101: Acquire the food type data, spectral data of the food surface, image data of the food, and current temperature and humidity data of the target area within the intelligent preservation device.

[0025] In this embodiment, the intelligent preservation device includes multiple independently controlled areas, each used to store different foods. Corresponding temperature and humidity control can be applied to these different areas to slow down the spoilage process and improve the precision of food preservation. In one specific embodiment, the intelligent preservation device can be a smart refrigerator. The device includes a spectral sensor, a camera, a temperature and humidity sensor, a humidity regulator, and a temperature regulator. It should be noted that each of the multiple areas is equipped with a spectral sensor, a camera, a temperature and humidity sensor, a humidity regulator, and a temperature regulator to monitor and control the temperature and humidity of each independently controlled area in real time. The detected data can then be processed to obtain temperature and humidity control data for precise control of the temperature and humidity in different areas.

[0026] In practical applications, spectral data is used as optical data to characterize the surface properties of food ingredients. Specifically, spectral data includes characteristic wavelengths and spectral reflectance, which can be acquired by spectral sensors in intelligent preservation equipment. It should be noted that food type data can also be indirectly obtained through data measured by spectral sensors. Specifically, by using spectral sensors to measure the reflection / absorption characteristics of target food ingredients at specific wavelengths, the resulting spectral curves contain intrinsic information such as the chemical composition and tissue structure of the target food ingredients. Furthermore, different types of food ingredients will exhibit different spectral characteristics in specific wavelength bands due to differences in chemical composition. By analyzing the differences in these spectral characteristics, a mapping relationship between spectral characteristics and type labels can be established, thereby determining the type data of the target food ingredients. Image data includes RGB images, which can be acquired by camera devices in intelligent preservation equipment.

[0027] S102: Data fusion is performed based on spectral data, image data, temperature and humidity data, and food type data to obtain multi-dimensional fused feature data.

[0028] In this embodiment of the application, multidimensional fusion feature data characterizes the feature data of the target food in different dimensions. Specifically, spectral data, image data, temperature and humidity data and food type data can be vector-encoded to form multidimensional fusion feature data with the above data vector features. Then, based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target food can be obtained.

[0029] In an optional implementation, step S102 may include: S1021: Data extraction and processing based on spectral data to obtain multiple spectral characteristic wavelengths associated with the target food ingredient; S1022: Determine the spectral feature vector based on the spectral reflectance characteristics corresponding to each of the multiple spectral feature wavelengths; S1023: Based on image data, determine multiple grayscale texture features corresponding to the target ingredient; S1024: Determine the image feature vector based on multiple grayscale texture features; S1025: Multidimensional fused feature data is obtained by stitching together spectral feature vectors, image feature vectors, temperature and humidity data and food type data.

[0030] In the embodiments of this application, the spectral characteristic wavelengths associated with the target food ingredient refer to the key absorption / reflection bands identified by spectral analysis technology that can characterize the composition, quality, or state of the target food ingredient. The spectral characteristic wavelength is a characteristic band range. Specifically, the Continuous Projection Algorithm (SPA) can be used to select characteristic wavelengths from the preprocessed spectral data to obtain multiple spectral characteristic wavelengths associated with the target food ingredient. The preprocessing includes spectral data denoising, smoothing, or baseline correction.

[0031] In one specific embodiment, the image data is an RGB image of the target ingredient. The RGB image of the target ingredient can be converted into a grayscale image of the target ingredient to extract multiple grayscale texture features corresponding to the target ingredient. Then, an image feature vector can be constructed based on the multiple grayscale texture features. Specifically, the multiple grayscale texture features can be vectorized and normalized. Next, the normalized feature vectors are horizontally concatenated to obtain the concatenated feature vector. For example, the concatenated feature vector can be represented as [feature 1_norm, feature 2_norm, ..., feature n_norm]. Further, the concatenated feature vector is subjected to dimensionality reduction processing to obtain the image feature vector corresponding to the target ingredient.

[0032] Furthermore, this application obtains multidimensional fusion feature data by splicing spectral feature vectors, image feature vectors, temperature and humidity data, and food type data. The multidimensional fusion feature data characterizes the target food from different feature dimensions, thereby compensating for the limitations of single features, capturing multi-level data, reducing the impact of noise, and thus improving the recognition accuracy of the target food, enhancing robustness, and achieving complementary features and comprehensive food characterization.

[0033] In an optional implementation, step S1022 may include: S10221: Normalize the difference of the spectral reflectance features corresponding to multiple spectral feature wavelengths to obtain the spectral feature vector.

[0034] Specifically, the normalized difference processing involves performing a difference operation on the spectral reflectance features corresponding to any two different bands among multiple spectral feature wavelengths to obtain a normalized spectral feature vector. Specifically, this is achieved by dividing the difference between the spectral reflectance features corresponding to any two different bands among multiple spectral feature wavelengths by their sum. By performing normalized difference processing on the spectral reflectance features corresponding to each of the multiple spectral feature wavelengths, the spectral reflectance features corresponding to each of the multiple spectral feature wavelengths can be normalized to the same metric, facilitating subsequent processing of the spectral feature vector.

[0035] S103: Based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target ingredient is obtained. The preset correspondence is the mapping relationship between the reference multidimensional fusion feature data and the reference freshness data. The freshness data represents the degree of spoilage of the target ingredient. In this embodiment, the preset correspondence can be implemented based on a freshness prediction model. The freshness prediction model is obtained by training a preset model based on reference multidimensional fusion feature data and reference freshness data. The preset model can be a neural network model, etc. Freshness data is used to objectively and quantitatively characterize the physical / chemical / biological parameters of the degree of spoilage or quality change of food. It reflects the actual freshness level of food through measurable data (such as specific gas concentration, spectral characteristics, number of microorganisms, etc.).

[0036] In practical applications, freshness data includes one or more of the following: microbial count, volatile basic nitrogen content, and chlorophyll content.

[0037] In an optional implementation, step S103 includes: S1031: Input the multidimensional fusion feature data into the freshness prediction model to predict the freshness and obtain the current freshness data of the target ingredient. The freshness prediction model is obtained by training the preset model based on the reference multidimensional fusion feature data and the reference freshness data.

[0038] In this embodiment, multidimensional fusion feature data is used as input data for the freshness prediction model, and the current freshness data of the target ingredient is used as the label for the freshness prediction model. Here, the freshness data represents the degree of spoilage of the target ingredient. The reference multidimensional fusion feature data refers to the feature data used to train the preset model, that is, the multidimensional fusion features (such as the fusion vector of features such as color, texture, and spectrum) extracted from the ingredient sample with known freshness data. The reference multidimensional fusion feature data is used as the input feature of the preset model, and the reference freshness data refers to the real freshness label or score corresponding to the above multidimensional fusion features. It is usually obtained by expert annotation, physicochemical index detection or objective measurement, and is used as the target value when training the preset model. Then, the preset model can be trained based on the reference multidimensional fusion feature data and the reference freshness data to obtain the freshness prediction model. Here, the preset model is a neural network model.

[0039] Specifically, by monitoring the multi-dimensional fusion feature data of the target ingredients in the target area in real time and inputting it into the freshness prediction model, the current freshness data of the target ingredients can be obtained. Based on the freshness data, temperature and humidity control instructions can be generated, and the current temperature and humidity in the target area can be adjusted in a closed loop and dynamically. This proactively and accurately delays the spoilage of the target ingredients in the target area, realizing precise control from extensive storage to on-demand preservation.

[0040] In an optional implementation, prior to step S101, the method further includes: S1011: Acquire reference food type data, reference spectral data of food surface, reference image data of food, and reference temperature and humidity data of food in the simulation chamber. S1012: Conduct destructive testing on the food in the simulation chamber and obtain reference freshness data of the food after the destructive testing; S1013: Based on reference spectral data, reference image data, reference temperature and humidity data, and reference food type data, data fusion is performed to obtain reference multidimensional fusion feature data; S1014: A freshness prediction model is obtained by training the model based on reference multidimensional fusion feature data and reference freshness data, and the preset correspondence is implemented based on the freshness prediction model.

[0041] In this embodiment, the simulation chamber is a controllable environment chamber, meaning that the environmental factors in the simulation chamber are adjustable. The simulation chamber is also equipped with a spectral sensor, a camera, and a temperature and humidity sensor to acquire reference food type data, reference spectral data of the food surface, reference image data of the food, and reference temperature and humidity data within the simulation chamber. The reference food type data refers to the types of food placed in the simulation chamber for the experiment; the reference spectral data is spectral data processed from the data sensed by the spectral sensor in the simulation chamber; the reference image data is image data processed from the data sensed by the camera in the simulation chamber; and the reference temperature and humidity data is the temperature and humidity data sensed by the temperature and humidity sensor in the simulation chamber.

[0042] In practical applications, reference spectral data, reference image data, reference food type data, and reference freshness data can be obtained for food under different levels of damage. This allows for data fusion based on the reference spectral data, reference image data, reference temperature and humidity data, and reference food type data to obtain reference multidimensional fusion feature data. Furthermore, a freshness prediction model can be trained based on the reference multidimensional fusion feature data and the reference freshness data to predict the freshness of food, thereby improving the accuracy of freshness determination and enabling a precise preservation process from state perception to proactive intervention.

[0043] It should be noted that the methods for determining the reference spectral data, reference image data, and reference multidimensional fusion feature data can refer to the methods for determining the spectral data, image data, and multidimensional fusion feature data, respectively, and will not be elaborated here.

[0044] S104: Based on the current freshness data of the target ingredient, dynamically adjust the temperature and humidity data in the target area.

[0045] In this embodiment of the application, the temperature and humidity data in the target area are dynamically adjusted by obtaining the current freshness data of the target ingredient, so as to delay the spoilage process of the target ingredient in the target area and improve the accuracy of the preservation control of the target ingredient.

[0046] In an optional implementation, step S104 may include: S1041: If the current freshness data of the target ingredient is less than the first freshness threshold, maintain the temperature and humidity data in the current target area.

[0047] In this embodiment, the first freshness threshold is the target freshness threshold. Different target ingredients correspond to different first freshness thresholds. Specifically, when the current freshness data of the target ingredient is less than the first freshness threshold, it indicates that the current temperature and humidity environment in the target area is relatively ideal. This can maintain the current temperature and humidity data in the target area to ensure that the target ingredient can maintain its freshness. At this time, there is no need to regulate the temperature and humidity in the target area, thereby avoiding physical deterioration such as microbial activity, cell membrane rupture, accelerated tissue softening, and browning caused by sudden changes in temperature and humidity. This ensures that the state of the target ingredient is relatively stable.

[0048] In an optional implementation, step S104 may include: S1042: When the current freshness data of the target ingredient is greater than or equal to the first freshness threshold and the current freshness data of the target ingredient is less than or equal to the second freshness threshold, a first freshness deviation is determined based on the first freshness threshold and the freshness data. Both the first freshness threshold and the second freshness threshold correspond to the type of the target ingredient, and the second freshness threshold is greater than the first freshness threshold. S1043: Based on the first freshness deviation and the first preset relationship, determine the first temperature and humidity adjustment data. The first preset relationship is the mapping relationship between the type of food, the first freshness deviation and the first temperature and humidity adjustment data. S1044: Based on the first temperature and humidity adjustment data and the current temperature and humidity data in the target area, obtain the first temperature and humidity control data; S1045: Adjust the temperature and humidity in the target area to the first temperature and humidity control data.

[0049] In this embodiment, the second freshness threshold is a set threshold that is greater than the first freshness threshold. Different target ingredients correspond to different second freshness thresholds. The first preset relationship is implemented based on a first preservation strategy lookup table. The first preservation strategy lookup table is indexed by the type of ingredient and the first freshness deviation, and stores the corresponding first temperature and humidity adjustment data. The first temperature and humidity adjustment data is the adjustment amount of temperature and humidity, and the first temperature and humidity control data is the temperature and humidity data that needs to be adjusted. Specifically, the first temperature and humidity control data can be obtained by subtracting the current temperature and humidity data in the target area from the first temperature and humidity adjustment data.

[0050] In practical applications, when the current freshness data of the target ingredient is greater than or equal to the first freshness threshold and less than or equal to the second freshness threshold, it indicates that the target ingredient is spoiling at an accelerated rate. Therefore, based on the deviation between the first freshness threshold and the current freshness data of the target ingredient, and the correspondence between the deviation and temperature and humidity adjustment, the direction and magnitude of temperature and humidity adjustment can be determined to adaptively adjust the temperature and humidity of the current target area in order to slow down the spoilage process of the target ingredient.

[0051] In an optional implementation, step S104 may include: S1046: If the current freshness data of the target ingredient is greater than the second freshness threshold, determine the second freshness deviation based on the second freshness threshold and the freshness data. The second freshness threshold corresponds to the type of ingredient of the target ingredient. S1047: Based on the second freshness deviation and the second preset relationship, determine the second temperature and humidity adjustment data. The second preset relationship is the mapping relationship between the type of food, the second freshness deviation and the second temperature and humidity adjustment data. S1048: Based on the second temperature and humidity adjustment data and the current temperature and humidity data in the target area, obtain the second temperature and humidity control data; S1049: Adjust the temperature and humidity in the target area to the second temperature and humidity control data, and generate freshness warning information. The second temperature and humidity control data is greater than the first temperature and humidity control data.

[0052] In this embodiment, the second preset relationship is implemented based on a second preservation strategy lookup table. The second preservation strategy lookup table uses the type of food and the second freshness deviation as indexes and stores the corresponding second temperature and humidity adjustment data. The second temperature and humidity adjustment data is the adjustment amount of temperature and humidity, and the second temperature and humidity control data is the temperature and humidity data after adjustment. Specifically, the second temperature and humidity control data can be obtained by subtracting the current temperature and humidity data and the second temperature and humidity adjustment data in the target area. The second temperature and humidity control data is greater than the first temperature and humidity control data. This can be understood as the less fresh the target food is, the greater the adjustment range, thereby delaying the spoilage of the target food.

[0053] In practical applications, when the current freshness data of the target ingredient is greater than the second freshness threshold, it indicates that the spoilage of the target ingredient has entered a rapid stage. Therefore, based on the deviation between the second freshness threshold and the current freshness data of the target ingredient, and the correspondence between the deviation and temperature and humidity adjustments, the direction and magnitude of temperature and humidity adjustments can be determined to slow down the spoilage process. At the same time, an early warning message is generated to remind users to use the target ingredient as soon as possible to avoid food spoilage and waste, ensure safe use, and reduce economic losses.

[0054] As can be seen from the above technical solutions of the embodiments of this application, the following technical effects are achieved: This application fuses spectral data, image data, temperature and humidity data, and food type data to obtain multidimensional fusion feature data. Based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target food can be determined. Then, based on the current freshness data of the target food, the temperature and humidity data in the target area can be dynamically adjusted. The technical solution provided by this application can delay the spoilage process of the food in the target area and improve the accuracy of food preservation control.

[0055] This application also provides a control device for an intelligent food preservation device, such as... Figure 3 The diagram shown is a structural schematic of a control device for an intelligent food preservation device provided in an embodiment of this application. The control of the intelligent food preservation device includes: The acquisition module 10 is used to acquire the food type data, spectral data of the food surface, image data of the food, and current temperature and humidity data of the target area within the intelligent preservation device.

[0056] The multidimensional fusion feature data determination module 20 is used to perform data fusion based on spectral data, image data, temperature and humidity data and food type data to obtain multidimensional fusion feature data.

[0057] The freshness data determination module 30 is used to obtain the current freshness data of the target ingredient based on the multidimensional fusion feature data and the preset correspondence. The preset correspondence is the mapping relationship between the reference multidimensional fusion feature data and the reference freshness data. The freshness data represents the degree of spoilage of the target ingredient.

[0058] The dynamic adjustment module 40 is used to dynamically adjust the temperature and humidity data in the target area based on the current freshness data of the target ingredient.

[0059] Furthermore, the multidimensional fusion feature data determination module 20 includes: The wavelength determination unit 201 is used to perform data extraction processing based on spectral data to obtain multiple spectral characteristic wavelengths associated with the target food ingredient.

[0060] The spectral feature vector determination unit 202 is used to determine the spectral feature vector based on the spectral reflectance characteristics corresponding to each of the multiple spectral feature wavelengths.

[0061] The grayscale texture feature determination unit 203 is used to determine multiple grayscale texture features corresponding to the target food based on image data.

[0062] The image feature vector determination unit 204 is used to determine the image feature vector based on multiple grayscale texture features.

[0063] The multidimensional fusion feature data determination unit 205 is used to perform splicing processing based on spectral feature vectors, image feature vectors, temperature and humidity data and food type data to obtain multidimensional fusion feature data.

[0064] Furthermore, the dynamic adjustment module 40 includes: The first adjustment unit 401 is used to maintain the temperature and humidity data in the current target area when the current freshness data of the target ingredient is less than the first freshness threshold.

[0065] Furthermore, the dynamic adjustment module 40 includes: The first freshness deviation determination unit 402 is used to determine a first freshness deviation based on the first freshness threshold and the freshness data when the current freshness data of the target ingredient is greater than or equal to a first freshness threshold and the current freshness data of the target ingredient is less than or equal to a second freshness threshold. The first freshness threshold and the second freshness threshold are both corresponding to the type of ingredient of the target ingredient, and the second freshness threshold is greater than the first freshness threshold.

[0066] The first temperature and humidity adjustment data determination unit 403 is used to determine the first temperature and humidity adjustment data based on the first freshness deviation and the first preset relationship. The first preset relationship is the mapping relationship between the type of food, the first freshness deviation and the first temperature and humidity adjustment data.

[0067] The first temperature and humidity control data determination unit 404 is used to obtain the first temperature and humidity control data based on the first temperature and humidity adjustment data and the current temperature and humidity data in the target area.

[0068] The second adjustment unit 405 is used to adjust the temperature and humidity in the target area to the first temperature and humidity control data.

[0069] Furthermore, the dynamic adjustment module 40 includes: The second freshness deviation determination unit 406 is used to determine the second freshness deviation based on the second freshness threshold and the freshness data when the current freshness data of the target ingredient is greater than the second freshness threshold. The second freshness threshold corresponds to the type of ingredient of the target ingredient.

[0070] The second temperature and humidity adjustment data determination unit 407 is used to determine the second temperature and humidity adjustment data based on the second freshness deviation and the second preset relationship. The second preset relationship is the mapping relationship between the type of food, the second freshness deviation and the second temperature and humidity adjustment data.

[0071] The second temperature and humidity control data determination unit 408 is used to obtain second temperature and humidity control data based on the second temperature and humidity adjustment data and the current temperature and humidity data in the target area.

[0072] The third adjustment unit 409 is used to adjust the temperature and humidity in the target area to the second temperature and humidity control data and generate freshness warning information. The second temperature and humidity control data is greater than the first temperature and humidity control data.

[0073] Furthermore, the freshness data determination module 30 includes: The model training unit 301 is used to input multidimensional fusion feature data into the freshness prediction model to predict the freshness and obtain the current freshness data of the target ingredient. The freshness prediction model is obtained by training a preset model based on the reference multidimensional fusion feature data and the reference freshness data.

[0074] Furthermore, the control device for the intelligent preservation equipment also includes: The reference data acquisition module 50 is used to acquire reference food type data, reference spectral data of food surface, reference image data of food, and reference temperature and humidity data of food in the simulation chamber.

[0075] The testing module 60 is used to conduct destructive tests on the food in the simulation chamber and obtain reference freshness data of the food after the destructive test.

[0076] The reference multidimensional data determination module 70 is used to perform data fusion based on reference spectral data, reference image data, reference temperature and humidity data and reference food type data to obtain reference multidimensional fusion feature data.

[0077] The freshness prediction model determination module 80 is used to train the model based on the reference multidimensional fusion feature data and the reference freshness data to obtain the freshness prediction model, and the preset correspondence is implemented based on the freshness prediction model.

[0078] Furthermore, the spectral feature vector determination unit 202 includes: The difference processing subunit 2021 is used to perform normalized difference processing on the spectral reflectance features corresponding to multiple spectral feature wavelengths to obtain spectral feature vectors.

[0079] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0080] This application embodiment also provides an intelligent preservation device, which includes a controller and a spectral sensor, a camera device, and a temperature and humidity sensor installed on the intelligent preservation device. The spectral sensor is used to acquire food type data and spectral data of the food surface in the target area. The camera device is used to acquire image data of the target food. The temperature and humidity sensor is used to detect temperature and humidity data in the target area. It should be noted that the intelligent preservation device includes multiple independently controlled areas, and each area is equipped with a spectral sensor, a camera device, and a temperature and humidity sensor to more accurately acquire data from different zones. The controller is used to execute the control method of the intelligent preservation device provided in the above method embodiment.

[0081] For example, the controller includes a main controller and multiple zone controllers. The main controller is used to acquire food type data, spectral data of the food surface, image data of the food, and current temperature and humidity data of the target area within the intelligent preservation equipment. It then performs data fusion based on the spectral data, image data, temperature and humidity data, and food type data to obtain multi-dimensional fusion feature data. Based on the multi-dimensional fusion feature data and a preset correspondence, it obtains the current freshness data of the target food. The preset correspondence is a mapping relationship between the multi-dimensional fusion feature data and reference freshness data. The freshness data represents the degree of spoilage of the target food. Based on the current freshness data of the target food, the controller dynamically adjusts the temperature and humidity data in the target area. Multiple zone controllers are respectively set up in different independently controlled areas within the intelligent preservation equipment. When the main controller determines the temperature and humidity adjustment data for different zones, it distributes the corresponding temperature and humidity adjustment data to the zone controllers, enabling the zone controllers to control the temperature and humidity of their respective areas, thereby improving the accuracy of temperature and humidity control.

[0082] This application embodiment also provides an intelligent control system, which includes a server 400 and an intelligent preservation device. The server 400 is communicatively connected to the intelligent preservation device and can interact with the intelligent preservation device for data exchange. Specifically, the server 400 can provide remote services to the intelligent preservation device. For example, the server 400 can provide services such as preservation mode selection and online upgrade of the preservation program.

[0083] In one specific embodiment, such as Figure 4The diagram illustrates the structure of a server according to an embodiment of this application. The server 400 can vary significantly depending on its configuration or performance, and may include one or more processors 410 (e.g., one or more processors) and storage 430, and one or more storage media 420 (e.g., one or more mass storage devices) for storing applications 423 or data 422. The memory 430 and storage media 420 can be temporary or persistent storage. The programs stored in the storage media 420 may include one or more modules, each module including a series of instruction operations on the server. Furthermore, the processor 410 may be configured to communicate with the storage media 420 and execute the series of instruction operations in the storage media 420 on the server 400. The server 400 may also include one or more power supplies 460, one or more wired or wireless network interfaces 450, one or more input / output interfaces 440, and / or one or more operating systems 421, such as Windows Server™, Mac OSX™, Unix™, Linux™, FreeBSD™, etc.

[0084] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system and server embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0086] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control method for an intelligent food preservation device, characterized in that, The method includes: Acquire the food type data of the target food in the target area within the intelligent preservation device, the spectral data of the surface of the target food, the image data of the target food, and the current temperature and humidity data of the target area; Data fusion is performed based on the spectral data, the image data, the temperature and humidity data, and the food type data to obtain multidimensional fused feature data; Based on the multidimensional fusion feature data and the preset correspondence, the current freshness data of the target ingredient is obtained. The preset correspondence is a mapping relationship between the reference multidimensional fusion feature data and the reference freshness data. The freshness data represents the degree of spoilage of the target ingredient. Based on the current freshness data of the target ingredient, the temperature and humidity data in the target area are dynamically adjusted.

2. The method according to claim 1, characterized in that, The data fusion based on the spectral data, image data, temperature and humidity data, and food type data yields multidimensional fused feature data, including: Data extraction and processing are performed based on the spectral data to obtain multiple spectral characteristic wavelengths associated with the target food ingredient; Based on the spectral reflectance characteristics corresponding to each of the multiple spectral characteristic wavelengths, a spectral feature vector is determined. Based on the image data, multiple grayscale texture features corresponding to the target food ingredient are determined; Based on the aforementioned multiple grayscale texture features, an image feature vector is determined; The multidimensional fusion feature data is obtained by stitching together the spectral feature vector, the image feature vector, the temperature and humidity data, and the food type data.

3. The method according to claim 1, characterized in that, The step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: If the current freshness data of the target ingredient is less than the first freshness threshold, the current temperature and humidity data in the target area shall be maintained.

4. The method according to claim 1, characterized in that, The step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: When the current freshness data of the target ingredient is greater than or equal to a first freshness threshold and the current freshness data of the target ingredient is less than or equal to a second freshness threshold, a first freshness deviation is determined based on the first freshness threshold and the freshness data. Both the first freshness threshold and the second freshness threshold correspond to the type of the target ingredient, and the second freshness threshold is greater than the first freshness threshold. Based on the first freshness deviation and the first preset relationship, the first temperature and humidity adjustment data is determined. The first preset relationship is the mapping relationship between the food type, the first freshness deviation and the first temperature and humidity adjustment data. Based on the first temperature and humidity adjustment data and the current temperature and humidity data in the target area, the first temperature and humidity control data is obtained; Adjust the temperature and humidity in the target area to the first temperature and humidity control data.

5. The method according to claim 1 or 4, characterized in that, The step of dynamically adjusting the temperature and humidity data in the target area based on the current freshness data of the target ingredient includes: If the current freshness data of the target ingredient is greater than the second freshness threshold, a second freshness deviation is determined based on the second freshness threshold and the freshness data. The second freshness threshold corresponds to the type of ingredient of the target ingredient. Based on the second freshness deviation and the second preset relationship, the second temperature and humidity adjustment data are determined. The second preset relationship is the mapping relationship between the food type, the second freshness deviation and the second temperature and humidity adjustment data. Based on the second temperature and humidity adjustment data and the current temperature and humidity data in the target area, the second temperature and humidity control data is obtained; The temperature and humidity in the target area are adjusted to the second temperature and humidity control data, and a freshness warning message is generated. The second temperature and humidity control data is greater than the first temperature and humidity control data.

6. The method according to claim 1, characterized in that, The process of obtaining the current freshness data of the target ingredient based on the multi-dimensional fusion feature data and the preset correspondence includes: The multidimensional fusion feature data is input into the freshness prediction model to predict the freshness, thereby obtaining the current freshness data of the target ingredient. The freshness prediction model is obtained by training a preset model based on the reference multidimensional fusion feature data and the reference freshness data.

7. The method according to claim 6, characterized in that, Before acquiring the food type data, spectral data of the food surface, image data of the food, and current temperature and humidity data of the target area within the intelligent preservation device, the method further includes: Acquire reference food type data, reference spectral data of the food surface, reference image data of the food, and reference temperature and humidity data of the food in the simulation chamber; The food ingredients in the simulation chamber were subjected to destructive testing, and reference freshness data of the food ingredients after the destructive testing was obtained; Based on the reference spectral data, the reference image data, the reference temperature and humidity data, and the reference food type data, data fusion is performed to obtain reference multidimensional fusion feature data; The freshness prediction model is obtained by training the model based on the reference multidimensional fusion feature data and the reference freshness data, and the preset correspondence is realized based on the freshness prediction model.

8. The method according to claim 2, characterized in that, The step of determining the spectral feature vector based on the spectral reflectance characteristics corresponding to each of the multiple spectral feature wavelengths includes: The spectral reflectance features corresponding to each of the multiple spectral feature wavelengths are normalized and interpolated to obtain the spectral feature vector.

9. An intelligent food preservation device, characterized in that, The intelligent preservation device includes a controller, which is used to execute the control method of the intelligent preservation device as described in any one of claims 1 to 8.

10. An intelligent control system, characterized in that, The intelligent control system includes a server and an intelligent preservation device. The server is communicatively connected to the intelligent preservation device and is capable of data interaction with the intelligent preservation device. The intelligent preservation device is used to execute the control method of the intelligent preservation device as described in any one of claims 1 to 8.

Images