Intelligent interaction packaging box for food and information acquisition method

Abstract

The invention relates to the technical field of food packaging, in particular to an intelligent interaction packaging box for food and an information acquisition method, the packaging box comprises a box body, an NFC interaction chip, a built-in acquisition module, an identification module, a related determination module, a correction module, a quality determination module and a control component of an output module. According to the method, the correlation degree between the parameters reflecting the apparent state and the pH value and the correlation degree between the parameters reflecting the apparent state and the overturning parameters are calculated and compared, and the correlation strength between the abnormal event and food internal chemical deterioration or external physical disturbance is quantified, so that the abnormal source is distinguished on the mechanism level; when it is judged that abnormity is mainly caused by physical disturbance, a preset apparent threshold value used for apparent state judgment is dynamically corrected, and a terminal consumer can obtain related information such as product quality through the NFC chip on the premise that sealing is not opened; the problem that the food quality misjudgment rate is high due to the fact that physical disturbance and real deterioration are difficult to distinguish on apparent signals is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of food packaging technology, and in particular to an intelligent interactive packaging box for food and a method for acquiring information. Background Technology

[0002] With the rapid development of logistics technology, the cross-regional distribution of high-end semi-solid foods such as sauces and pastes has become commonplace. However, during transportation and handling, reversible apparent changes caused by drastic physical disturbances and irreversible intrinsic deterioration caused by spoilage chemical reactions exhibit highly similar signals in the unopened state, making it difficult for existing monitoring technologies to accurately distinguish between them. This not only leads to consumer misjudgment and food waste but also hinders precise quality management of the supply chain. Therefore, intelligently identifying physical disturbances and chemical spoilage has become a key challenge in solving food safety issues and optimizing supply chain management.

[0003] Chinese Patent Application Publication No. CN108382729A discloses a packaging box with an NFC chip and its control method. The packaging box includes a sealed box body and further includes: a substrate adhered to the inner surface of the box body; a sensing module installed in the substrate, including a gas sensor and a humidity sensor, wherein the gas sensor is used to detect the gas concentration inside the box body and transmit it to the NFC chip, and the humidity sensor is used to detect the humidity inside the box body and transmit it to the NFC chip; an NFC chip installed in the substrate and connected to the sensing module; the NFC chip includes a detection module for detecting the temperature inside the box body, and the detection module calculates the temperature inside the box body using a thermistor with a positive temperature coefficient and its corresponding first weight and a thermistor with a negative temperature coefficient and its corresponding second weight.

[0004] Therefore, the packaging box with NFC chip and its control method have the following problems: the packaging box relies on a flexible battery for power, which is prone to unstable operation of the sensor and chip due to battery aging; the packaging box relies on monitoring general environmental parameters, which can easily overlook the core direct indicators of product deterioration; the packaging box simply analyzes environmental factors such as temperature and humidity and odor recognition, which can easily fail to distinguish between external disturbances and product deterioration, leading to misjudgments of food quality. Summary of the Invention

[0005] To address this, the present invention provides an intelligent interactive packaging box and information acquisition method for food, which overcomes the problem of high misjudgment rate and poor reliability of food quality in the prior art due to the difficulty in distinguishing between physical disturbances and actual deterioration in apparent signals through a multi-parameter collaborative judgment and self-correction mechanism.

[0006] To achieve the above objectives, in one aspect, the present invention provides an intelligent interactive packaging box for food, comprising: The container is used to store condiments and similar food items; An NFC interaction chip, which is set on the surface of the box, is used to store food quality information; The control unit, located on the side of the housing, includes an acquisition module, an identification module, a correlation determination module, a correction module, a quality determination module, and an output module. The acquisition module is used to acquire in real time the height difference of the free liquid level in the box, the flipping angle of the box, the number of flipping times, the pH value of the food, and the proportion of food adhering to the wall. The identification module is used to identify apparent anomalies based on the comparison result between the apparent index determined by the height difference of the free liquid surface and the wall coverage ratio and a preset apparent threshold. The relevant determination module is used to determine the quality correlation and disturbance correlation based on the apparent anomaly identification results and according to the synergistic characteristics of the changes in the free liquid surface height difference and the wall-mounted coverage ratio with the pH value, the flipping angle, and the number of flipping times, respectively. The correction module is used to determine the apparent anomaly type based on the comparison results of the quality correlation and the perturbation correlation, and to correct the preset apparent threshold by combining the temporal correlation features of the flip angle and the number of flips. The quality determination module is used to determine a comprehensive quality index based on the deviation of the corrected preset apparent threshold and the pH value. The output module is used to output the food quality information determined based on the threshold comparison result of the comprehensive quality index to the NFC interactive chip.

[0007] Furthermore, the identification module includes: An index calculation unit is used to perform a weighted summation of the normalized free liquid surface height difference and the wall coverage ratio based on a threshold comparison result of the free liquid surface height difference and the wall coverage ratio, so as to determine the apparent index. An appearance recognition unit is used to identify appearance anomalies when the appearance index is greater than the preset appearance threshold.

[0008] Furthermore, the relevant determination module includes: A quality-related calculation unit is used to determine the pH height correlation and pH wall adhesion correlation based on the correlation characteristics between the pH value and the free liquid surface height difference and the wall coverage ratio, respectively. The disturbance-related calculation unit is used to determine the angle-height correlation, angle-wall-attachment correlation, number-height correlation, and number-wall-attachment correlation based on the correlation characteristics between the flipping angle and the number of flipping cycles and the free liquid surface height difference and the wall-attachment coverage ratio, respectively. The relevant determination unit is used to determine the quality correlation based on the pH high correlation and the pH wall adhesion correlation, and to determine the disturbance correlation based on the number high correlation, the number wall adhesion correlation, the angle high correlation, and the angle wall adhesion correlation.

[0009] Furthermore, the relevant determination unit includes: The first determining subunit is used to determine the quality correlation based on the average of the pH height correlation and the pH wall adhesion correlation; The second determining subunit is used to determine the disturbance correlation based on the average of the frequency high correlation, the frequency wall-hanging correlation, the angle high correlation, and the angle wall-hanging correlation.

[0010] Furthermore, the correction module includes: The synchronization determination unit is used to determine the disturbance-related anomaly based on the comparison result of the quality correlation degree and the disturbance correlation degree, the comparison result of the thresholds of the quality correlation degree and the disturbance correlation degree, and to determine the anomaly duration based on the time difference between the anomaly flip time determined by the flip angle and the flip number and the apparent anomaly time. The correction unit is used to determine the apparent anomaly type as the physical disturbance type based on the threshold comparison result of the abnormal duration, and to correct the preset apparent threshold according to the temporal correlation characteristics of the flip angle and the number of flips.

[0011] Furthermore, the synchronization determination unit includes: An anomaly determination subunit is used to determine the disturbance-related anomaly when the quality correlation is less than the disturbance correlation, the quality correlation is less than a preset quality correlation threshold, and the disturbance correlation is greater than a preset disturbance correlation threshold. The flip determination subunit is used to determine the historical moment when the flip angle is greater than a preset angle threshold and the number of flips is greater than a preset number threshold as the abnormal flip moment based on the disturbance-related anomaly. A synchronous determination subunit is used to calculate the time difference between the abnormal reversal time and the apparent abnormal time to obtain the abnormal duration.

[0012] Furthermore, the correction unit includes: A type determination subunit is used to determine the apparent anomaly type as the physical disturbance type when the anomaly duration is greater than a preset anomaly threshold. The influence quantization subunit is used to determine the flipping comprehensive factor based on the physical disturbance type, according to the average value of the flipping angle and the average value of the flipping number within the abnormal duration; A correction subunit is used to correct the preset appearance threshold based on the coupling relationship between the flip synthesis factor and the preset appearance threshold.

[0013] Furthermore, the quality determination module includes: A deviation calculation unit is used to determine the apparent deviation degree based on the difference between the apparent index and the corrected preset apparent threshold, and to determine the pH deviation degree based on the difference between the pH value and the preset pH threshold. A quality determination unit is used to determine the overall quality index based on a weighted fusion result of the apparent deviation and the pH deviation.

[0014] Furthermore, the output module includes: The grade determination unit is used to determine the quality grade as Level 1, Level 2, or Level 3 based on the threshold comparison result of the comprehensive quality index. An output unit is used to output the quality level to the NFC interaction chip.

[0015] On the other hand, the present invention also provides a method for obtaining information about food, comprising: The height difference of the free liquid level inside the box and the proportion of food adhering to the wall are obtained; Apparent anomalies are identified by comparing the apparent index determined by the height difference of the free liquid surface and the wall coverage ratio with a preset apparent threshold. Obtain the box's flipping angle, the number of flips, and the food's pH value; Based on the apparent anomaly identification results, the quality correlation and disturbance correlation are determined according to the synergistic characteristics of the changes in the free liquid surface height difference and the wall-mounted coverage ratio with the pH value, the flipping angle, and the number of flipping times, respectively. The apparent anomaly type is determined based on the comparison results of the quality correlation and the perturbation correlation, and the preset apparent threshold is corrected by combining the temporal correlation features of the flip angle and the number of flips. The overall quality index is determined based on the degree of deviation between the modified preset apparent threshold and the pH value; The food quality information determined based on the threshold comparison results of the comprehensive quality index is output to the NFC interactive chip.

[0016] Compared with existing technologies, the beneficial effects of this invention lie in its ability to quantify the correlation between abnormal events and inherent chemical deterioration or external physical disturbances in food by calculating and comparing the correlation between parameters reflecting apparent state and pH value and inversion parameters, thereby distinguishing the root cause of the abnormality at the mechanistic level. Based on this, when the abnormality is determined to be mainly caused by physical disturbance, the preset apparent threshold used for apparent state determination is dynamically adjusted according to the intensity and duration of the disturbance event, enabling the packaging to learn and adapt to normal physical noise generated during transportation and handling, thus significantly reducing the subsequent false alarm rate. Furthermore, by converting the comprehensive quality index obtained after accurate diagnosis and dynamic correction into an intuitive quality level and specific action suggestions, and outputting it through a passive and convenient medium such as an NFC chip, end consumers can obtain product quality and other relevant information without intrusion or opening the package. This effectively solves the problem of high false alarm rates and poor reliability in food quality assessment caused by the difficulty in distinguishing between physical disturbances and actual deterioration in apparent signals.

[0017] Furthermore, by setting independent preset height and ratio thresholds, the comprehensive calculation and anomaly determination of the apparent index are only initiated when both the free liquid surface height difference and the wall coverage ratio, two key physical parameters, simultaneously exceed their respective thresholds. This avoids misjudgments caused by random factors affecting a single parameter, thus improving the specificity and reliability of apparent anomaly identification. Simultaneously, by weighted summation, the degree of exceeding the limits in both dimensions is integrated into a comprehensive apparent index, providing accurate and quantitative input for subsequent collaborative analysis distinguishing between physical disturbances and quality deterioration.

[0018] Furthermore, by systematically calculating multiple sets of Pearson correlation coefficients between apparent parameters and chemical and physical disturbance indicators within a unified, pre-defined time period, a data-driven diagnosis of the causes of apparent anomalies was achieved. By transforming the two key judgments—whether the anomaly is related to chemical deterioration and whether it is related to physical disturbance—from subjective experience or single-signal analysis into quantitative assessment based on statistical correlation, erroneous decisions caused by subjective misjudgment or single-indicator analysis are avoided.

[0019] Furthermore, by averaging multiple independent correlation coefficients into a single quality correlation and perturbation correlation index, the dimensionality reduction and quantification of complex synergistic relationships are achieved. The quality correlation integrates the correlation between pH value and two apparent parameters, reliably capturing abnormal patterns driven by chemical deterioration. The perturbation correlation integrates the multiple correlations between the angle and number of flips and two apparent parameters, which can characterize the intensity and pattern of the impact of physical perturbation on the apparent state. This provides a clear and reliable decision-making basis for accurately distinguishing whether the root cause of the anomaly is quality deterioration or physical perturbation and initiating corresponding intelligent correction strategies.

[0020] Furthermore, by comparing the strength of the correlation between quality and disturbance and using threshold screening, the causes of anomalies are initially distinguished. Based on this, the specific violent reversal events that triggered the anomalies are precisely traced back and identified. The causal relationship is rigorously verified by calculating the time difference between these events and the appearance of the anomalies. In addition, an anomaly is only classified as a physical disturbance type when it is confirmed that it was directly caused by an adjacent preceding physical disturbance. The preset appearance threshold is then adjusted accordingly, giving the system's judgment criteria context awareness and adaptive capabilities. This ensures that the final output quality information truly reflects the chemical safety status of the food, rather than a temporary change in physical state.

[0021] Furthermore, by utilizing the relative strength of the correlation between quality and disturbance and comparing it with their respective thresholds, the anomaly can be macroscopically identified as potentially originating from the disturbance. Based on this, by using angle and frequency thresholds to trace back and locate the moment of dramatic reversal in history and calculating the time difference between that moment and the moment the apparent anomaly occurred, the abstract strong correlation is transformed into the specific timing of what dramatic action occurred and whether the anomaly followed closely behind. This achieves the verification from statistical correlation to causal association of specific events, enhancing the certainty and reliability of attributing the apparent anomaly to physical disturbance.

[0022] Furthermore, by calculating the average angle and number of flips within the confirmed abnormal duration, the average intensity of the disturbance is quantified. By fusing multi-dimensional disturbance information into a flipping comprehensive factor and dynamically adjusting the preset apparent threshold using the flipping comprehensive factor, the system's judgment criteria can be intelligently relaxed based on the intensity of the most recently confirmed physical disturbance. This avoids misjudging normal physical state changes caused by non-quality factors such as transportation and handling as food spoilage, ensuring that the quality assessment results truly reflect the inherent chemical safety status of the food.

[0023] Furthermore, by defining a preset pH threshold as an objective benchmark for chemical deterioration, and by using preset apparent weight coefficients and preset pH weight coefficients to differentiate the weighting of apparent deviation and pH deviation, a comprehensive evaluation model with chemical indicators as the core and corrected apparent indicators as an auxiliary is constructed. This model can balance the physical disturbance information and intrinsic chemical information after preliminary diagnosis and correction, ensuring that the final output comprehensive quality index is not excessively affected by non-deteriorating physical interference, thereby transforming the relatively accurate root cause analysis results into a stable and reliable quality grade.

[0024] Furthermore, by setting clear upper and lower limits for preset index ranges, the system converts continuous comprehensive quality indices into well-defined quality levels. This achieves an efficient mapping from complex multi-sensor data analysis results to clear conclusions that can be intuitively understood and acted upon. By establishing a secondary quality level as an intermediate level, precise operational guidance is provided for products that are in a critical quality state, have undergone certain changes but have not yet reached the level of spoilage, suggesting their use in the short term. This eliminates safety hazards while significantly reducing unnecessary food losses caused by absolute judgment standards, thus achieving a good balance between safety control and user experience.

[0025] Furthermore, by analyzing the coordinated changes in apparent parameters, pH value, and flipping data over time, the system intelligently distinguishes between physical disturbances and chemical deterioration. Based on this, for non-deterioration anomalies caused by transportation and handling, the system can automatically learn and relax the appearance judgment threshold, significantly reducing the false alarm rate. Simultaneously, by weightedly fusing the corrected appearance information with the core chemical indicator pH value, a stable and reliable comprehensive quality index is output. This ensures that the quality level and usage recommendations obtained by users through the NFC chip under non-invasive conditions are accurate guidance that eliminates transportation interference and truly reflects the internal state of the food. This enhances user experience and the credibility of supply chain quality traceability while ensuring consumer safety. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the control component for the intelligent interactive packaging box used for food in this embodiment; Figure 2 This is a logic diagram for the appearance recognition unit to identify appearance anomalies in this embodiment; Figure 3 This is a logic diagram for determining the quality level as Level 1 by the quality determination unit in this embodiment. Figure 4 This is a flowchart of the information acquisition method for food in this embodiment. Detailed Implementation

[0027] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0028] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0029] Please see Figure 1As shown, this is a structural schematic diagram of the control component for the intelligent interactive packaging box for food in this embodiment. On one hand, this embodiment provides an intelligent interactive packaging box for food, including: The container is used to store condiments and similar food items; An NFC interaction chip, which is set on the surface of the box, is used to store food quality information; The control unit, located on the side of the housing, includes an acquisition module, an identification module, a correlation determination module, a correction module, a quality determination module, and an output module. The acquisition module is used to acquire in real time the height difference of the free liquid level in the box, the flipping angle of the box, the number of flipping times, the pH value of the food, and the proportion of food adhering to the wall. The identification module, which is connected to the acquisition module, is used to identify apparent anomalies based on the comparison result between the apparent index determined by the height difference of the free liquid surface and the wall coverage ratio and a preset apparent threshold. The relevant determination module is connected to the acquisition module and the identification module respectively, and is used to determine the quality correlation and disturbance correlation based on the appearance anomaly identification result and the synergistic characteristics of the changes in the free liquid surface height difference and the wall coverage ratio with the pH value, the flipping angle and the number of flipping, respectively. The correction module is connected to the acquisition module and the correlation determination module respectively, and is used to determine the apparent anomaly type based on the comparison results of the quality correlation and the perturbation correlation, and to correct the preset apparent threshold by combining the temporal correlation features of the flip angle and the number of flips. The quality determination module is connected to the acquisition module and the correction module respectively, and is used to determine the comprehensive quality index based on the degree of deviation between the corrected preset apparent threshold and the pH value. The output module, which is connected to the quality determination module, is used to output the food quality information determined based on the threshold comparison result of the comprehensive quality index to the NFC interactive chip.

[0030] In this embodiment, the smart interactive packaging box for food is used for quality monitoring during the supply chain logistics and last-mile delivery of semi-solid or high-viscosity liquid seasonings. During this process, the product experiences continuous vibration and bumps during long-distance transportation, frequent handling and turning during delivery and when the user receives the seasoning. These complex physical disturbances can easily cause changes in the product's internal structure, such as oil-water separation, component sedimentation, and uneven properties. These changes can easily be mistaken by sensors as food spoilage, leading to misjudgments when the user has not opened the packaging and accessed the product quality information via the NFC interactive chip.

[0031] In this embodiment, multi-dimensional parameters are captured in real time through built-in sensors. The control component is the core control unit of the intelligent interactive packaging box, which typically uses a high-performance microprocessor or microcontroller, such as the STM32 series or Arduino series. It is responsible for receiving raw data from various sensors such as accelerometers, gyroscopes, pH sensors, and optical sensors, and for collecting, analyzing, and processing this data in real time according to a preset algorithm program. Among them, the free liquid level height difference refers to the vertical height difference between the liquid levels on both sides of the seasoning inside the packaging box, reflecting the uniformity and fluidity of the seasoning. This difference is calculated using an ultrasonic ranging sensor installed at the bottom or side wall of the packaging box, combined with the current flip angle information of the box. The flip angle refers to the spatial angle of the packaging box relative to its initial upright state. This is obtained using an accelerometer or gyroscope installed inside the box near the center. The accelerometer calculates the tilt angle by measuring the changes in the box's acceleration, while the gyroscope determines the tilt angle relative to the initial upright state by detecting changes in the box's angular velocity and direction. The control component calculates the flip angle based on the data output from the sensors. The number of flips refers to the number of times the packaging box flips from its initial upright state within a preset time period. The number of complete cycles after one or more flips and returning to an upright position is determined by monitoring the acceleration of an accelerometer or the angular velocity obtained from a gyroscope. Each time a flip event is detected, it is counted once, and the final flip count is obtained. pH value is a basic chemical indicator for measuring acidity and alkalinity. A pH sensor installed at the contact point on the bottom or side wall of the box reacts with hydrogen ions in the liquid and converts it into an electrical signal. The control component reads this signal in real time to obtain the pH value. The wall-mounted coverage ratio refers to the percentage of the area of ​​the seasoning attached to the inner side wall of the packaging box relative to the total inner wall area. A distributed capacitive sensor installed on the inner side wall of the packaging box identifies and counts the area of ​​the seasoning attached based on the changes in the sensor signal, and then calculates the wall-mounted coverage ratio.

[0032] The preset appearance threshold is a benchmark value used to measure the appearance state of food. It depends on the physical characteristics and quality standards of the food and is usually set between 0.1 and 1.0. In this embodiment, it is set to 0.5, which can accurately distinguish between normal and abnormal states.

[0033] By calculating and comparing the correlation between parameters reflecting apparent state and pH value and inversion parameters, the strength of the association between abnormal events and the inherent chemical deterioration or external physical disturbances in food can be quantified, thus distinguishing the root cause of the abnormality at the mechanistic level. Based on this, when the abnormality is determined to be mainly caused by physical disturbance, the preset apparent threshold used for apparent state determination is dynamically adjusted according to the intensity and duration of the disturbance event. This allows the packaging to learn and adapt to the normal physical noise generated during transportation and handling, significantly reducing subsequent false alarm rates. Furthermore, by converting the comprehensive quality index obtained after precise diagnosis and dynamic correction into an intuitive quality level and specific action suggestions, and outputting it through a passive and convenient medium like an NFC chip, end consumers can obtain product quality information non-invasively and without opening the package. This effectively solves the problem of high false alarm rates and poor reliability in food quality assessment caused by the difficulty in distinguishing between physical disturbances and actual deterioration in apparent signals.

[0034] Please continue reading. Figure 2 As shown, this is a logic diagram for the appearance recognition unit to identify appearance anomalies in this embodiment. In this embodiment, the recognition module includes: An index calculation unit is used to perform a weighted summation of the ratio of the free liquid surface height difference to the preset height threshold and the ratio of the wall-mounted coverage ratio to the preset ratio threshold when the free liquid surface height difference is greater than a preset height threshold and the wall-mounted coverage ratio is greater than a preset ratio threshold, in order to determine the apparent index. An appearance recognition unit, connected to the index calculation unit, is used to identify the appearance anomaly when the appearance index is greater than the preset appearance threshold.

[0035] The preset height threshold is a benchmark value used to determine whether the height difference of the free liquid surface is abnormal. It depends on the viscosity, flowability and packaging size of the food, and is usually set between 5mm and 20mm. In this embodiment, it is set to 10mm, which can effectively filter out normal liquid surface fluctuations caused by slight shaking or temperature changes. The preset proportion threshold is a benchmark value used to determine whether the wall-mounted coverage ratio is abnormal. It depends on the adhesion characteristics of the food and the inner wall surface process, and is usually set between 10% and 30%. In this embodiment, it is set to 15%, which can distinguish between normal wall-mounted adhesion and abnormal adhesion caused by deterioration or severe disturbance.

[0036] In this embodiment, during the weighted summation of the ratio of the free liquid surface height difference to the preset height threshold and the ratio of the wall-attached coverage ratio to the preset ratio threshold, the weight corresponding to the ratio of the free liquid surface height difference to the preset height threshold reflects the contribution of the free liquid surface height difference to the apparent index. It depends on the physical characteristics of the food and its sensitivity to anomalies such as stratification and sedimentation, and is usually set between 0.5 and 0.8. In this embodiment, it is set to 0.6, which can highlight the role of liquid surface instability as a core apparent anomaly indicator. The weight corresponding to the ratio of the wall-attached coverage ratio to the preset ratio threshold depends on the adhesiveness of the food and the significance of the wall-attached phenomenon, and is usually set between 0.2 and 0.5. In this embodiment, it is set to 0.4, which can reasonably reflect the contribution of the wall-attached phenomenon to the overall appearance.

[0037] By setting independent preset height and ratio thresholds, the comprehensive calculation and anomaly determination of the apparent index are only initiated when both the free liquid surface height difference and the wall coverage ratio, two key physical parameters, simultaneously exceed their respective thresholds. This avoids misjudgments caused by random factors affecting a single parameter, thus improving the specificity and reliability of apparent anomaly identification. Furthermore, by using weighted summation, the degree of exceeding the limits in both dimensions is integrated into a comprehensive apparent index, providing accurate and quantitative input for subsequent collaborative analysis distinguishing between physical disturbances and quality deterioration.

[0038] Specifically, the relevant determination module includes: The quality-related calculation unit is used to calculate the Pearson correlation coefficient between the pH value and the height difference of the free liquid surface based on a preset time period, so as to obtain the pH height correlation, and to calculate the Pearson correlation coefficient between the pH value and the wall coverage ratio, so as to obtain the pH wall adhesion correlation. The disturbance correlation calculation unit is used to calculate the Pearson correlation coefficients of the flipping angle with respect to the height difference of the free liquid surface and the wall coverage ratio based on the preset time duration, so as to obtain the angle height correlation and the angle wall coverage correlation; and to calculate the Pearson correlation coefficients of the number of flippings with respect to the height difference of the free liquid surface and the wall coverage ratio, so as to obtain the number height correlation and the number wall coverage correlation. A determining unit, which is connected to the mass correlation calculation unit and the disturbance correlation calculation unit respectively, is used to determine the mass correlation based on the pH height correlation and the pH wall adhesion correlation, and to determine the disturbance correlation based on the number height correlation, the number wall adhesion correlation, the angle height correlation and the angle wall adhesion correlation.

[0039] The preset time frame is the length of the time window used to calculate the Pearson correlation coefficient. It depends on the typical time scale of food state changes and the real-time requirements of the system. It is usually set between 15 minutes and 1 hour. In this embodiment, it is set to 30 minutes, which can effectively capture the short-term or medium-term synergistic change patterns between pH value, apparent parameters and flipping events.

[0040] By systematically calculating multiple sets of Pearson correlation coefficients between apparent parameters and chemical and physical disturbance indicators within a unified, pre-defined time period, a data-driven diagnosis of the causes of apparent anomalies was achieved. By transforming the two key judgments—whether the anomaly is related to chemical deterioration and whether it is related to physical disturbance—from subjective experience or single-signal analysis into a quantitative assessment based on statistical correlation, erroneous decisions caused by subjective misjudgment or single-indicator analysis are avoided.

[0041] Specifically, the relevant determination unit includes: The first determining subunit is used to determine the quality correlation based on the average of the pH height correlation and the pH wall adhesion correlation; The second determining subunit is used to determine the disturbance correlation based on the average of the frequency high correlation, the frequency wall-hanging correlation, the angle high correlation, and the angle wall-hanging correlation.

[0042] By averaging multiple independent correlation coefficients into single quality correlation and perturbation correlation indices, the dimensionality reduction and quantification of complex synergistic relationships are achieved. The quality correlation integrates the correlation between pH value and two apparent parameters, reliably capturing abnormal patterns driven by chemical deterioration. The perturbation correlation integrates the multiple correlations between the angle and number of flips and two apparent parameters, which can characterize the intensity and pattern of the impact of physical perturbation on the apparent state. This provides a clear and reliable decision-making basis for accurately distinguishing whether the root cause of the anomaly is quality deterioration or physical perturbation and initiating corresponding intelligent correction strategies.

[0043] Specifically, the correction module includes: A synchronization determination unit is used to determine the abnormal duration based on the comparison result of the quality correlation and the disturbance correlation, the threshold comparison result of the quality correlation and the disturbance correlation, and the time difference between the abnormal flip time determined by the flip angle and the number of flips and the apparent abnormal time. A correction unit, connected to the related determination unit, is used to determine the apparent anomaly type as the physical disturbance type based on the threshold comparison result of the abnormal duration, and to correct the preset apparent threshold according to the temporal correlation characteristics of the flip angle and the number of flips.

[0044] By comparing the strength of the correlation between quality and disturbance and using threshold screening, the causes of anomalies are initially distinguished. Based on this, the specific violent reversal events that triggered the anomalies are precisely traced back and identified. The causal relationship is rigorously verified by calculating the time difference between these events and the appearance of the anomalies. Furthermore, an anomaly is only classified as a physical disturbance when it is confirmed that it was directly caused by an adjacent preceding physical disturbance. The preset appearance threshold is then adjusted accordingly, giving the system's judgment criteria context awareness and adaptive capabilities. This ensures that the final output quality information truly reflects the chemical safety status of the food, rather than a temporary change in physical state.

[0045] Specifically, the synchronization determination unit includes: An anomaly determination subunit is used to determine a disturbance-related anomaly when the quality correlation is less than the disturbance correlation, the quality correlation is less than a preset quality correlation threshold, and the disturbance correlation is greater than a preset disturbance correlation threshold. The flip determination subunit is used to determine the historical moment when the flip angle is greater than a preset angle threshold and the number of flips is greater than a preset number threshold as the abnormal flip moment based on the disturbance-related anomaly. A synchronous determination subunit is used to calculate the time difference between the abnormal reversal time and the apparent abnormal time to obtain the abnormal duration.

[0046] The preset quality correlation threshold is a threshold value used to determine whether the quality correlation is significant. It depends on the stringency requirements for attributing quality anomalies and the signal-noise level, and is typically set between 0.2 and 0.4. In this embodiment, it is set to 0.3, ensuring that anomalies are determined to be related to quality only when pH and apparent parameters show a clear positive or negative co-change. The preset perturbation correlation threshold is a threshold value used to determine whether the perturbation correlation is significant. It depends on the required significance level of physical perturbation events, and is typically set between 0.5 and 0.8. In this embodiment, it is set to 0.6, reliably identifying a statistically strong or higher correlation between the flipping parameter and the apparent parameter. The angle threshold is used to define the lower limit of the angle of a violent flipping event. It depends on the engineering definition of an effective physical disturbance action and common use cases, and is usually set between 30 degrees and 60 degrees. In this embodiment, it is set to 45 degrees, which can effectively distinguish between everyday slight tilting and shaking and effective flipping actions that may cause significant changes in the structure of the contents. The preset number threshold is used to define the lower limit of the frequency of frequent flipping. It depends on the flipping frequency that is sufficient to produce a cumulative disturbance effect on semi-solid or high-viscosity fluids. It is usually set between 3 times / minute and 8 times / minute. In this embodiment, it is set to 5 times / minute, which can filter out flipping patterns that are continuous and may cause cumulative changes in apparent parameters.

[0047] By utilizing the relative strength of the correlation between quality and disturbance and comparing it with their respective thresholds, the anomaly can be macroscopically identified as potentially originating from the disturbance. Based on this, the moment of dramatic reversal in history can be traced back by using angle and frequency thresholds, and the time difference between this moment and the moment the apparent anomaly occurred can be calculated. This transforms the abstract strong correlation into the specific timing of what dramatic action occurred and whether the anomaly followed closely behind. This achieves the verification from statistical correlation to causal association of specific events, enhancing the certainty and reliability of attributing apparent anomalies to physical disturbances.

[0048] Specifically, the correction unit includes: A type determination subunit is used to determine the apparent anomaly type as the physical disturbance type when the time synchronization degree is greater than a preset synchronization threshold. An influence quantification subunit, connected to the type determination subunit, is used to determine a flipping comprehensive factor based on the physical disturbance type, according to the average value of the flipping angle and the average value of the flipping number within the abnormal duration, where Z = a × (D / D0) + c × (S / S0), where Z is the flipping comprehensive factor, a is a preset angle coefficient, D is the average value of the flipping angle, D0 is a preset angle threshold, c is a preset number coefficient, S is the average value of the flipping number, and S0 is a preset number threshold. A correction subunit, connected to the influence quantization subunit, is used to correct the preset appearance threshold based on the coupling relationship between the flip synthesis factor and the preset appearance threshold, wherein G C =G0×(1+y×Z), where G C G0 is the modified preset apparent threshold, G0 is the original preset apparent threshold, y is the preset correction coefficient, and Z is the flipping comprehensive factor.

[0049] The preset synchronization threshold is the maximum allowable time difference used to determine whether there is a temporal causal relationship between the time of anomalous flip and the time of apparent anomaly. It depends on the typical delay time of a detectable apparent response caused by a physical disturbance, and is usually set between 1 and 5 minutes. In this embodiment, it is set to 2 minutes, which can accurately identify a violent flip event immediately before the occurrence of apparent anomaly as a potential trigger. The preset angle coefficient is a coefficient used to weight the angle influence component when calculating the flip comprehensive factor. It depends on the relative importance of the flip angle change on the apparent parameter, and is usually set between 0.5 and 0.8. In this embodiment, it is set to 0.7, which can make the angle mean have a dominant weight in the comprehensive factor. The preset angle threshold is a reference angle value used to normalize the average flip angle, and is usually set between 30 and 60 degrees. In this embodiment, it is set to 45 degrees, which can convert the measured average angle into a dimensionless ratio. The preset frequency coefficient is a weighting factor used to calculate the flipping comprehensive factor, depending on the cumulative effect of the flipping frequency on the apparent parameters. It is usually set between 0.2 and 0.5, and in this embodiment, it is set to 0.3, which can reasonably characterize the continuous and cumulative disturbance effect caused by high-frequency shaking and complement the angle influence. The preset frequency threshold is a reference frequency value used to normalize the average flipping frequency, usually set between 3 times / minute and 8 times / minute, and in this embodiment, it is set to 5 times / minute, which can quantify the persistence and density of the disturbance. The preset correction coefficient is an intensity coefficient used to control the adjustment of the preset apparent threshold according to the flipping comprehensive factor. It depends on the design target of the system's tolerance to physical disturbances, and is usually set between 0.1 and 0.5, and in this embodiment, it is set to 0.3, which can ensure effective reduction of false alarms while avoiding the underreporting of real quality anomalies due to overcorrection.

[0050] By calculating the average angle and number of flips within the confirmed anomaly duration, the average intensity of the disturbance is quantified. By fusing multi-dimensional disturbance information into a flipping comprehensive factor and dynamically adjusting the preset apparent threshold using the flipping comprehensive factor, the system's judgment criteria can be intelligently relaxed based on the intensity of the most recently confirmed physical disturbance. This avoids misjudging normal physical state changes caused by non-quality factors such as transportation and handling as food spoilage, ensuring that the quality assessment results truly reflect the food's intrinsic chemical safety status.

[0051] Specifically, the quality determination module includes: A deviation calculation unit is used to calculate the relative deviation between the apparent index and the corrected preset apparent threshold to obtain the apparent deviation degree, where R B =max[0,(G t -G C ) / G C ], where R BIt is the apparent deviation, G t It is the apparent index, and the relative deviation between the pH value and the preset pH threshold is calculated to obtain the pH deviation, R. H =max[0,(H t -H0) / H0], where R H It is the pH deviation, H t It is the pH value, and H0 is the preset pH threshold; A quality determination unit, connected to the deviation calculation unit, is used to determine a comprehensive quality index based on a weighted fusion result of the apparent deviation and the pH deviation, where U = 1 - (α × R) B +β×R H ), where U is the comprehensive quality index, α is the preset apparent weight coefficient, and β is the preset pH weight coefficient.

[0052] The preset pH threshold is a benchmark value used to determine whether the pH value is abnormal. It depends on the acidity or alkalinity range of the seasoning and is usually set between 4.0 and 7.0. In this embodiment, it is set to 5.0, which can effectively distinguish between normal and spoiled acidity or alkalinity ranges. The preset apparent weight coefficient is the weight of the apparent index's contribution to the comprehensive quality index. It is usually set between 0.5 and 0.8. In this embodiment, it is set to 0.6, which can highlight the importance of the apparent index in the comprehensive quality assessment. The preset pH weight coefficient is the weight of the pH value's contribution to the comprehensive quality index. It is usually set between 0.2 and 0.5. In this embodiment, it is set to 0.4, which can reasonably reflect the impact of pH value on changes in food quality.

[0053] By defining a preset pH threshold as an objective benchmark for chemical deterioration, and using preset apparent weight coefficients and preset pH weight coefficients to differentiate the weighting of apparent deviation and pH deviation, a comprehensive evaluation model with chemical indicators as the core and corrected apparent indicators as an auxiliary is constructed. This model can balance the physical disturbance information and intrinsic chemical information after preliminary diagnosis and correction, ensuring that the final output quality comprehensive index is not excessively affected by non-deteriorating physical interference, thereby transforming the relatively accurate root cause analysis results into a stable and reliable quality grade.

[0054] Please see Figure 3 As shown, this is the logic diagram for determining the quality level as Level 1 by the quality determination unit in this embodiment. In this embodiment, the output module includes: A grade determination unit is used to determine the quality grade as Level 1 when the comprehensive quality index is greater than the upper limit of a preset index range, and to determine the quality grade as Level 2 when the comprehensive quality index is greater than the lower limit of a preset index range and less than the upper limit of a preset index range, and to determine the quality grade as Level 3 when the comprehensive quality index is less than the lower limit of a preset index range. An output unit, connected to the grade determination unit, is used to output the quality grade to the NFC interactive chip on the box.

[0055] In this embodiment, the food quality information includes not only the quality grade but also corresponding usage recommendations generated based on the quality grade. Specifically, when the quality grade is Level 1, the corresponding usage recommendation is that it can be used normally; when the quality grade is Level 2, the corresponding usage recommendation is that it should be used within 2 weeks; and when the quality grade is Level 3, the corresponding usage recommendation is that it is not recommended to consume and after-sales service should be requested.

[0056] The upper limit of the preset index range is the highest boundary value for distinguishing quality grades. It depends on the strict requirements for the minimum quality score that high-quality products need to achieve, and is usually set between 0.85 and 0.95. In this embodiment, it is set to 0.9, which can ensure that only products with extremely high comprehensive quality index can be classified as Grade 1 quality. The lower limit of the preset index range is the lowest boundary value for distinguishing quality grades. It depends on the minimum tolerance for the quality and safety of seasonings and the level of risk control, and is usually set between 0.6 and 0.75. In this embodiment, it is set to 0.7, which can effectively intercept products that no longer meet the basic safety and edibility requirements and prevent them from being mistakenly judged as edible.

[0057] By setting clear upper and lower limits for preset index ranges, the system converts continuous comprehensive quality indices into well-defined quality levels. This achieves an efficient mapping from complex multi-sensor data analysis results to clear conclusions that can be intuitively understood and acted upon. By establishing a secondary quality level as an intermediate level, it provides precise operational guidance for products that are in a critical quality state and have undergone certain changes but have not yet reached the level of spoilage. This eliminates safety hazards and significantly reduces unnecessary food losses caused by absolute judgment standards, thus achieving a good balance between safety control and user experience.

[0058] On the other hand, please see Figure 4 As shown, it is a flowchart of the information acquisition method for food in this embodiment; This embodiment also provides a method for obtaining information about food, including: The height difference of the free liquid level inside the box and the proportion of food adhering to the wall are obtained; Apparent anomalies are identified by comparing the apparent index determined by the height difference of the free liquid surface and the wall coverage ratio with a preset apparent threshold. Obtain the box's flipping angle, the number of flips, and the food's pH value; Based on the apparent anomaly identification results, the quality correlation and disturbance correlation are determined according to the synergistic characteristics of the changes in the free liquid surface height difference and the wall-mounted coverage ratio with the pH value, the flipping angle, and the number of flipping times, respectively. The apparent anomaly type is determined based on the comparison results of the quality correlation and the perturbation correlation, and the preset apparent threshold is corrected by combining the temporal correlation features of the flip angle and the number of flips. The overall quality index is determined based on the degree of deviation between the modified preset apparent threshold and the pH value; The food quality information determined based on the threshold comparison results of the comprehensive quality index is output to the NFC interactive chip.

[0059] By analyzing the coordinated changes in apparent parameters, pH value, and flipping data over time, the system intelligently distinguishes between physical disturbances and chemical deterioration. Based on this, for non-deterioration anomalies caused by transportation and handling, the system can automatically learn and relax the appearance judgment threshold, significantly reducing the false alarm rate. Simultaneously, by weightedly fusing the corrected appearance information with the core chemical indicator pH value, the system ultimately outputs a stable and reliable comprehensive quality index. This ensures that the quality level and usage recommendations obtained by users through the NFC chip under non-invasive conditions are accurate guidance that eliminates transportation interference and truly reflects the internal state of the food. This enhances user experience and the credibility of supply chain quality traceability while ensuring consumer safety.

[0060] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart interactive packaging box for food, characterized in that, include: The container is used to store condiments and similar food items; An NFC interaction chip, which is set on the surface of the box, is used to store food quality information; The control unit, located on the side of the housing, includes an acquisition module, an identification module, a correlation determination module, a correction module, a quality determination module, and an output module. The acquisition module is used to acquire in real time the height difference of the free liquid level in the box, the flipping angle of the box, the number of flipping times, the pH value of the food, and the proportion of food adhering to the wall. The identification module is used to identify appearance anomalies based on the comparison result between the appearance index determined by the height difference of the free liquid surface and the wall coverage ratio and a preset appearance threshold. The relevant determination module is used to determine the quality correlation and disturbance correlation based on the apparent anomaly identification results and according to the synergistic characteristics of the changes in the free liquid surface height difference and the wall coverage ratio with the pH value, the flipping angle, and the number of flipping times, respectively. The correction module is used to determine the apparent anomaly type based on the comparison result of the quality correlation and the perturbation correlation, and to correct the preset apparent threshold by combining the temporal correlation features of the flip angle and the number of flips. The quality determination module is used to determine a comprehensive quality index based on the deviation of the corrected preset apparent threshold and the pH value. The output module is used to output the food quality information determined based on the threshold comparison result of the comprehensive quality index to the NFC interactive chip.

2. The intelligent interactive packaging box for food according to claim 1, characterized in that, The identification module includes: An index calculation unit is used to perform a weighted summation of the normalized free liquid surface height difference and the wall coverage ratio based on a threshold comparison result of the free liquid surface height difference and the wall coverage ratio, so as to determine the apparent index. An appearance recognition unit is used to identify appearance anomalies when the appearance index is greater than the preset appearance threshold.

3. The intelligent interactive packaging box for food according to claim 2, characterized in that, The relevant determination module includes: A quality-related calculation unit is used to determine the pH height correlation and pH wall adhesion correlation based on the correlation characteristics between the pH value and the free liquid surface height difference and the wall coverage ratio, respectively. The disturbance-related calculation unit is used to determine the angle-height correlation, angle-wall-attachment correlation, number-height correlation, and number-wall-attachment correlation based on the correlation characteristics between the flipping angle and the number of flipping cycles and the height difference of the free liquid surface and the wall-attachment coverage ratio, respectively. The relevant determination unit is used to determine the quality correlation based on the pH high correlation and the pH wall adhesion correlation, and to determine the disturbance correlation based on the number high correlation, the number wall adhesion correlation, the angle high correlation, and the angle wall adhesion correlation.

4. The intelligent interactive packaging box for food according to claim 3, characterized in that, The relevant determination unit includes: The first determining subunit is used to determine the quality correlation based on the average of the pH height correlation and the pH wall adhesion correlation; The second determining subunit is used to determine the disturbance correlation based on the average of the frequency high correlation, the frequency wall-hanging correlation, the angle high correlation, and the angle wall-hanging correlation.

5. The intelligent interactive packaging box for food according to claim 4, characterized in that, The correction module includes: The synchronization determination unit is used to determine the disturbance-related anomaly based on the comparison result of the quality correlation degree and the disturbance correlation degree, the comparison result of the thresholds of the quality correlation degree and the disturbance correlation degree, and to determine the anomaly duration based on the time difference between the anomaly flip time determined by the flip angle and the flip number and the apparent anomaly time. The correction unit is used to determine the apparent anomaly type as the physical disturbance type based on the threshold comparison result of the abnormal duration, and to correct the preset apparent threshold according to the temporal correlation characteristics of the flip angle and the number of flips.

6. The intelligent interactive packaging box for food according to claim 5, characterized in that, The synchronization determination unit includes: An anomaly determination subunit is used to determine the disturbance-related anomaly when the quality correlation is less than the disturbance correlation, the quality correlation is less than a preset quality correlation threshold, and the disturbance correlation is greater than a preset disturbance correlation threshold. The flip determination subunit is used to determine the historical moment when the flip angle is greater than a preset angle threshold and the number of flips is greater than a preset number threshold as the abnormal flip moment based on the disturbance-related anomaly. A synchronous determination subunit is used to calculate the time difference between the abnormal reversal time and the apparent abnormal time to obtain the abnormal duration.

7. The intelligent interactive packaging box for food according to claim 6, characterized in that, The correction unit includes: A type determination subunit is used to determine the apparent anomaly type as the physical disturbance type when the anomaly duration is greater than a preset anomaly threshold. The influence quantization subunit is used to determine the flipping comprehensive factor based on the physical disturbance type, according to the average value of the flipping angle and the average value of the flipping number within the abnormal duration; A correction subunit is used to correct the preset appearance threshold based on the coupling relationship between the flip synthesis factor and the preset appearance threshold.

8. The intelligent interactive packaging box for food according to claim 7, characterized in that, The quality determination module includes: A deviation calculation unit is used to determine the apparent deviation degree based on the difference between the apparent index and the corrected preset apparent threshold, and to determine the pH deviation degree based on the difference between the pH value and the preset pH threshold. A quality determination unit is used to determine the overall quality index based on a weighted fusion result of the apparent deviation and the pH deviation.

9. The intelligent interactive packaging box for food according to claim 8, characterized in that, The output module includes: The grade determination unit is used to determine the quality grade as Level 1, Level 2, or Level 3 based on the threshold comparison result of the comprehensive quality index. An output unit is used to output the quality level to the NFC interaction chip.

10. A method for acquiring information about food, applied to the intelligent interactive packaging box for food as described in any one of claims 1-9, characterized in that, include: The height difference of the free liquid level inside the box and the proportion of food adhering to the wall are obtained; Apparent anomalies are identified by comparing the apparent index determined by the height difference of the free liquid surface and the wall coverage ratio with a preset apparent threshold. Obtain the box's flipping angle, the number of flips, and the food's pH value; Based on the apparent anomaly identification results, the quality correlation and disturbance correlation are determined according to the synergistic characteristics of the changes in the free liquid surface height difference and the wall-mounted coverage ratio with the pH value, the flipping angle, and the number of flipping times, respectively. The apparent anomaly type is determined based on the comparison results of the quality correlation and the perturbation correlation, and the preset apparent threshold is corrected by combining the temporal correlation features of the flip angle and the number of flips. The overall quality index is determined based on the degree of deviation between the modified preset apparent threshold and the pH value; The food quality information determined based on the threshold comparison results of the comprehensive quality index is output to the NFC interactive chip.

Images