Method and system for monitoring activity decay during storage of an egg-derived trophic factor product

Abstract

The application provides a method and system for monitoring activity attenuation of an egg-derived nutritional factor product during storage. The concentration of characteristic aldehyde and the ratio of protein oxidation fluorescence peak intensity are obtained by non-destructive headspace gas sensing and fluorescence detection. A response bias curve is constructed using calibration samples with the same activity level, and the two parameters are corrected according to temperature and water activity. Multiple speckle images are collected, and the average speckle contrast ratio of each local area is calculated to determine the spatial distribution characteristics. The range is compared with the standard sample with known caking grade, and the deterioration index of hygroscopic caking is determined. The corrected two values and the deterioration index are input into the activity prediction model to form an activity retention rate estimation sequence. The activity trend is determined according to the change slope, and the activity attenuation of the product is determined. The above scheme can improve the comparability of activity attenuation monitoring results under different storage conditions, thereby reducing the frequency of destructive standard activity detection during storage.

Description

Technical Field

[0001] This application relates to the field of food storage quality testing technology, specifically to a method and system for monitoring the activity decay of egg-derived nutritional factors during storage. Background Technology

[0002] Egg-derived nutritional factor products are products formed by extracting, separating, or formulating proteins, lipids, or other functional components from egg raw materials. Oxygen, temperature, and moisture conditions during storage can cause lipid oxidation, protein oxidative denaturation, and moisture absorption and clumping, thereby altering the standard activity test values ​​of the target components.

[0003] Current storage monitoring of egg-derived nutritional factor products typically involves offline testing of extracted samples using chromatographic, immunoassay, or corresponding standard activity detection methods. This makes it difficult to continuously obtain activity changes of the same monitored product. Furthermore, non-destructive sensing signals used to reflect oxidation and structural changes are affected by temperature, moisture state, laser power, exposure parameters, and imaging position, resulting in a lack of comparability in activity decay monitoring results under different storage conditions. Consequently, it is difficult to continuously and reliably confirm the activity decay of the product, and therefore, destructive standard activity detection is still frequently required for verification. Summary of the Invention

[0004] This application provides a method and system for monitoring the activity decay of egg-derived nutritional factor products during storage, which can improve the comparability of activity decay monitoring results under different storage conditions, thereby reducing the frequency of destructive standard activity testing during storage.

[0005] In a first aspect, this application provides a method for monitoring the activity decay of egg-derived nutrient factor products during storage, comprising: Non-destructive headspace gas sensing and fluorescence detection were used to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios for stored egg-derived nutrient products. Based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed between the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity are obtained according to the response deviation curve. Collect multi-frame speckle images of egg source nutritional factor products, calculate the multi-frame average speckle contrast value of each local area based on the multi-frame speckle images, and then determine the spatial distribution characteristics based on the multi-frame average speckle contrast value of each local area. Compare the spatial distribution characteristics with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption agglomeration. The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into an activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is a downward trend, it is determined that the product has experienced activity decay.

[0006] In some embodiments, obtaining the characteristic aldehyde concentration measurement value and the ratio of protein oxidation fluorescence peak intensity specifically includes: The gas in the headspace of the product packaging is selectively responded to by a gas sensor array to obtain the response signal of the target characteristic aldehyde. Using a pre-established working curve between the target characteristic aldehyde response signal and concentration, the response signal of the target characteristic aldehyde is converted into a characteristic aldehyde concentration measurement value; The product is irradiated sequentially with a preset reference excitation wavelength and a preset oxidation indication excitation wavelength using a fiber optic fluorescence probe, and the corresponding fluorescence spectra are collected. Extract the reference peak intensity and oxidation indicator peak intensity after normalization of the instrument response from the corresponding fluorescence spectrum, calculate the ratio of the oxidation indicator peak intensity to the reference peak intensity, and obtain the protein oxidation fluorescence peak intensity ratio.

[0007] In some embodiments, the gas sensor array is a plurality of semiconductor gas sensors with different aldehyde molecule selective sensitive membranes.

[0008] In some embodiments, based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed between the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity are obtained according to the response deviation curve. Specifically, this includes: Multiple temperature and water activity calibration conditions were set up. Calibration samples with the same activity level confirmed by standard activity testing were selected. After water balance was reached under each calibration condition, the characteristic aldehyde concentration and the ratio of protein oxidation fluorescence peak intensity were measured during the preset calibration period. Using the characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratio at reference temperature and reference water activity as references, the response deviations of characteristic aldehyde concentration and protein oxidation fluorescence peak intensity ratio under each calibration condition were calculated, and a response deviation curve indexed by temperature and water activity was constructed. The current temperature is obtained by a temperature sensor, and the current water activity value is obtained by testing parallel samples of the same batch and under the same storage conditions as the monitored product using a water activity meter according to a preset calibration cycle. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are found using the current temperature and current water activity value as indexes. The characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are then subtracted from the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio, respectively, to obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio.

[0009] In some embodiments, selecting calibration samples with the same activity level confirmed by standard activity testing specifically includes: Multiple calibration sample groups are set up from the same batch of products with the same product formula and packaging form. Each calibration sample group includes a sensor measurement sample and a corresponding parallel test sample. Before and after the calibration conditions reach moisture balance, the parallel test samples are subjected to standard activity tests. Only the sensor measurement samples in the calibration sample group whose test results are all within the same preset activity retention rate equivalent range are determined as the calibration samples.

[0010] In some embodiments, acquiring multi-frame speckle images of egg-derived nutritional factor products specifically includes: The product is illuminated by a laser light source through a transparent area or a reserved detection window of the product packaging, causing the product to scatter light and form a scattered light field carrying information about the product's microstructure. Keeping the laser power, camera exposure parameters, and product imaging position constant, multiple speckle images formed by the interference of the scattered light field are continuously acquired at fixed time intervals.

[0011] In some embodiments, the laser source is a semiconductor laser diode.

[0012] In some embodiments, the multi-frame average speckle contrast ratio of each local region is calculated based on the multi-frame speckle images, and then the spatial distribution characteristics are determined based on the multi-frame average speckle contrast ratio of each local region. The spatial distribution characteristics are compared with the calibration range of a standard sample with a known agglomeration level to determine the deterioration index of the moisture-absorbing agglomerates. Specifically, this includes: Image registration is performed on the multi-frame speckle images, and the registered speckle images are divided into multiple local regions corresponding to the positions, and the speckle contrast value of each local region is calculated frame by frame. Based on the positional correspondence of local regions, the average speckle contrast ratio of each local region is obtained by averaging the speckle contrast ratio of each local region across multiple frames of speckle images. The median and interquartile range of the multi-frame average speckle contrast ratio of each local region are calculated to form the spatial distribution characteristics, and the range of spatial distribution characteristics pre-calibrated by a standard sample with a known agglomeration level under the same acquisition conditions is obtained. The spatial distribution characteristics are compared with the range of the spatial distribution characteristics to determine the moisture absorption and clumping status level of the product, and the moisture absorption and clumping status level is converted into a moisture absorption and clumping deterioration index according to the preset level assignment rules.

[0013] In some embodiments, the corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into an activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is a downward trend, it is determined that the product has experienced activity decay. Specifically, this includes: The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at each monitoring time in the accelerated storage experiment were used as input features. The activity retention rate obtained by standard activity detection of the same sample was used as a label to train and validate the activity prediction model. The corrected characteristic aldehyde concentration measurements, corrected protein oxidation fluorescence peak intensity ratios, and deterioration index of moisture absorption and clumping of the product under monitoring at each continuous monitoring time are input into the validated activity prediction model to obtain the estimated activity retention rate at each monitoring time. The estimated values ​​of the activity retention rate at each monitoring time are sorted according to the monitoring time to form the activity retention rate estimation sequence. The activity retention rate estimation sequence is linearly fitted within the monitoring window with the actual monitoring time as the independent variable. The slope of the fitted line is calculated and used as the change slope. The activity trend is determined based on the sign of the slope of the change, and when the activity trend is a downward trend, it is determined that the product has experienced activity decay.

[0014] Secondly, this application provides a monitoring system for the activity decay of egg-derived nutrient factor products during storage, comprising: The acquisition module is used to perform non-destructive headspace gas sensing and fluorescence detection on stored egg-derived nutrient factor products to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios. The processing module is used to construct a response deviation curve of the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio relative to the reference temperature and reference water activity based on calibration samples with the same activity level confirmed by standard activity detection, obtain the current temperature and current water activity value, and when the current temperature and current water activity value are within the calibration range of the response deviation curve, obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio according to the response deviation curve. The processing module is also used to acquire multi-frame speckle images of egg source nutritional factor products, calculate the multi-frame average speckle contrast value of each local area based on the multi-frame speckle images, and then determine the spatial distribution characteristics based on the multi-frame average speckle contrast value of each local area. The spatial distribution characteristics are compared with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption agglomeration. The execution module is used to input the corrected characteristic aldehyde concentration measurement value, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring time into the activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence, calculate the slope of the activity retention rate estimation sequence within the monitoring window, determine the activity trend based on the slope, and determine that the product has experienced activity decay when the activity trend is a downward trend.

[0015] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for monitoring the activity decay of egg-derived nutritional factor products during storage.

[0016] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for monitoring the activity decay of egg-derived nutritional factor products during storage.

[0017] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The method and system for monitoring the activity decay of egg-derived nutrient factor products during storage provided in this application utilizes calibration samples with the same activity level to construct response deviation curves. Benchmark corrections are applied to the characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios under different temperatures and water activities. Multi-frame speckle images of the egg-derived nutrient factor products are acquired, and the average speckle contrast ratio of each local region is calculated based on these images. The spatial distribution characteristics of the average speckle contrast ratio of each local region and the calibration range of the standard sample determine the deterioration index of moisture absorption and clumping. The corrected characteristic aldehyde concentration measurements, corrected protein oxidation fluorescence peak intensity ratios, and the deterioration index of moisture absorption and clumping at continuous monitoring times are then input into a pre-trained activity prediction model to form an activity retention rate estimation sequence. The slope of the change determines the activity trend and whether activity decay has occurred. This scheme integrates condition correction, structural state calibration, and continuous time-series determination into the same monitoring process, enabling chemical and structural responses to be included in activity estimation under a unified benchmark. This improves the comparability of activity decay monitoring results under different storage conditions, thereby reducing the frequency of destructive standard activity testing during storage. Attached Figure Description

[0018] Figure 1 This is an exemplary flowchart of a method for monitoring the activity decay of egg-derived nutrient factor products during storage, according to some embodiments of this application. Figure 2 This is an exemplary flowchart illustrating the ratio of aldehyde concentration measurements to protein oxidation fluorescence peak intensity according to some embodiments of this application; Figure 3 This is an exemplary flowchart illustrating the determination of the deterioration index of moisture-absorbing caking according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a monitoring system for the activity decay of egg-derived nutrient factor products during storage, as shown in some embodiments of this application. Figure 5 This is a schematic diagram of the structure of a computer device for implementing a method for monitoring the activity decay of egg-derived nutrient factor products during storage, according to some embodiments of this application. Detailed Implementation

[0019] To better understand the above technical solutions, the following will provide a detailed description of the technical solutions in conjunction with the accompanying drawings and specific embodiments.

[0020] refer to Figure 1 The figure is an exemplary flowchart of a method for monitoring the activity decay of egg-derived nutrient products during storage, according to some embodiments of this application. This method mainly includes the following steps: In step 101, the stored egg-derived nutrient product is subjected to non-destructive headspace gas sensing and fluorescence detection to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios.

[0021] In some embodiments, obtaining the characteristic aldehyde concentration measurement and the ratio of protein oxidation fluorescence peak intensity can be achieved by the following steps: The gas in the headspace of the product packaging is selectively responded to by a gas sensor array to obtain the response signal of the target characteristic aldehyde. Using a pre-established working curve between the target characteristic aldehyde response signal and concentration, the response signal of the target characteristic aldehyde is converted into a characteristic aldehyde concentration measurement value; The product is irradiated sequentially with a preset reference excitation wavelength and a preset oxidation indication excitation wavelength using a fiber optic fluorescence probe, and the corresponding fluorescence spectra are collected. Extract the reference peak intensity and oxidation indicator peak intensity after normalization of the instrument response from the corresponding fluorescence spectrum, calculate the ratio of the oxidation indicator peak intensity to the reference peak intensity, and obtain the protein oxidation fluorescence peak intensity ratio.

[0022] The gas sensor array comprises multiple semiconductor gas sensors with different aldehyde molecule selectivity sensitive membranes. Specifically, this array is arranged in the packaging headspace or a sealed sensing interface connected to the packaging headspace. When target characteristic aldehydes such as hexanal and pentanal are adsorbed onto their corresponding sensitive membranes, they cause changes in the sensor's conductivity or resistance, and the gas sensor array outputs a response signal from the target characteristic aldehyde. A working curve between the response signal and concentration is established using a standard gas of the target characteristic aldehyde at a known concentration. During actual monitoring, the response signal of the target characteristic aldehyde is converted into a measured value of the characteristic aldehyde concentration based on this working curve. For fluorescence detection, a spectral pre-scan is performed using an unoxidized reference sample with the same formulation and calibration samples with different oxidation degrees to determine the reference excitation wavelength, oxidation indicator excitation wavelength, and corresponding emission peaks. For example, the intensity of the tryptophan reference peak at approximately 340 nm can be extracted under excitation at approximately 280 nm, and the intensity of the oxidized crosslinking product indicator peak at approximately 410 nm to 420 nm can be extracted under excitation at approximately 325 nm. After normalizing the excitation power and detector response using fluorescence calibration standards, the ratio of the oxidation indicator peak intensity to the reference peak intensity is calculated to obtain the protein oxidation fluorescence peak intensity ratio.

[0023] It should be noted that the characteristic aldehyde concentration measurement value in this application includes the concentration value of at least one target characteristic aldehyde such as hexanal or pentanal, used to characterize the level of lipid oxidation-related volatiles in the product; the protein oxidation fluorescence peak intensity ratio is the ratio of the oxidation indicator peak intensity to the reference peak intensity, used to characterize the change in protein oxidation product accumulation relative to the fluorescence of the natural protein. Headspace gas detection can be completed through a built-in sensing element in the packaging or a sealed sensing interface, and fluorescence detection can be completed through a transparent packaging area or a reserved detection window. The transparent packaging area or reserved detection window has light transmittance at both the reference excitation wavelength and the oxidation indicator excitation wavelength. For example, a reserved detection window made of quartz glass or other ultraviolet-transmitting materials can be used, thereby obtaining the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio without removing the product being tested.

[0024] In step 102, based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed between the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity are obtained according to the response deviation curve.

[0025] In some embodiments, based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed between the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity can be obtained according to the response deviation curve using the following steps: Multiple temperature and water activity calibration conditions were set up. Calibration samples with the same activity level confirmed by standard activity testing were selected. After water balance was reached under each calibration condition, the characteristic aldehyde concentration and the ratio of protein oxidation fluorescence peak intensity were measured during the preset calibration period. Using the characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratio at reference temperature and reference water activity as references, the response deviations of characteristic aldehyde concentration and protein oxidation fluorescence peak intensity ratio under each calibration condition were calculated, and a response deviation curve indexed by temperature and water activity was constructed. The current temperature is obtained by a temperature sensor, and the current water activity value is obtained by testing parallel samples of the same batch and under the same storage conditions as the monitored product using a water activity meter according to a preset calibration cycle. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are found using the current temperature and current water activity value as indexes. The characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are then subtracted from the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio, respectively, to obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio.

[0026] The selection of calibration samples with the same activity level confirmed by standard activity testing can be achieved by the following steps: multiple calibration sample groups are set up from the same batch of products with the same product formula and packaging form. Each calibration sample group includes a sensor measurement sample and a corresponding parallel test sample. Standard activity testing is performed on the parallel test samples before and after the calibration conditions reach moisture balance. Only the sensor measurement sample in the calibration sample group whose test results are all within the same preset equivalent range of activity retention rate is determined as the calibration sample.

[0027] Each calibration sample group is placed in parallel under target calibration conditions. Sensing samples are used to collect sensor responses, while parallel detection samples are used for standard activity testing before and after calibration. A correspondence is established through sample group numbers. Current temperature is collected by temperature sensors located near the product at preset time intervals. Current water activity values ​​are obtained by a water activity meter from parallel samples of the same batch, packaging, and storage conditions as the monitored product, according to a preset calibration cycle. The corresponding response deviation curve is reconstructed when the product batch, packaging, or storage conditions change. This method ensures that headspace gas, fluorescence, and speckle measurements of the monitored product remain non-destructive, using only parallel samples for periodic calibration.

[0028] In some embodiments, reference Figure 2 The figure is an exemplary flowchart illustrating the correction ratio of characteristic aldehyde concentration measurement and protein oxidation fluorescence peak intensity according to some embodiments of this application. The corrected characteristic aldehyde concentration measurement and the corrected protein oxidation fluorescence peak intensity ratio, obtained based on the response deviation curve, can be achieved through the following steps: In step 1021, when the current temperature and the current water activity value are within the calibration range of the response deviation curve, the characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are searched using the current temperature and the current water activity value as indexes. In step 1022, the characteristic aldehyde concentration response deviation is subtracted from the characteristic aldehyde concentration measurement value, and the protein oxidation fluorescence peak intensity ratio response deviation is subtracted from the protein oxidation fluorescence peak intensity ratio value to obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio value.

[0029] The system uses the current temperature and current water activity value as a two-dimensional index to locate the response deviation of characteristic aldehyde concentration and the response deviation of protein oxidation fluorescence peak intensity ratio in the response deviation curve, and subtracts them from the measured value of characteristic aldehyde concentration and the protein oxidation fluorescence peak intensity ratio, respectively. If the current temperature or current water activity value exceeds the calibration range, the system stops outputting the corrected measured value of characteristic aldehyde concentration and the corrected protein oxidation fluorescence peak intensity ratio, records the original measured value and the current operating condition, and prompts the user to use the standard activity detection result or expand the calibration range; extrapolated values ​​exceeding the range are not included in the activity prediction model.

[0030] It should be noted that the response deviation curve characterizes the difference in measurement response between calibration samples with the same activity level and those differing only in temperature and water activity. This curve is used to convert measurements under various storage conditions to the same baseline condition. The characteristic aldehyde concentration response deviation is the difference between the measured characteristic aldehyde concentration under each calibration condition and the measured characteristic aldehyde concentration at the baseline temperature and baseline water activity. The protein oxidation fluorescence peak intensity ratio response deviation is the difference between the protein oxidation fluorescence peak intensity ratio under each calibration condition and the protein oxidation fluorescence peak intensity ratio at the baseline temperature and baseline water activity. Each response deviation has the same dimension as the corresponding measured value and can therefore be directly subtracted. The actual activity change caused by storage is estimated using a model relationship between continuously monitored characteristics and standard activity detection labels.

[0031] In step 103, multi-frame speckle images of egg source nutritional factor products are acquired. The average speckle contrast ratio of each local area is calculated based on the multi-frame speckle images. Then, the spatial distribution characteristics are determined based on the average speckle contrast ratio of each local area. The spatial distribution characteristics are compared with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption and agglomeration.

[0032] In some embodiments, acquiring multi-frame speckle images of egg-derived nutritional factor products can be achieved using the following steps: The product is illuminated by a laser light source through a transparent area or a reserved detection window of the product packaging, causing the product to scatter light and form a scattered light field carrying information about the product's microstructure. Keeping the laser power, camera exposure parameters, and product imaging position constant, multiple speckle images formed by the interference of the scattered light field are continuously acquired at fixed time intervals.

[0033] The laser source is a semiconductor laser diode, which can be used to illuminate the product surface. Each monitoring session uses a fixed laser wavelength, laser power, spot size, incident angle, camera exposure parameters, camera aperture, packaging window, and product imaging position, continuously acquiring multiple frames of speckle images within the same acquisition period. Each frame is first registered, and then the speckle contrast ratio of the same local area is averaged across multiple frames to reduce random imaging noise. The moisture absorption and clumping state of the product is determined by the correspondence between the spatial distribution of the average speckle contrast ratio across multiple frames and the calibration range of the standard sample. It should be noted that the speckle image is a granular image of light and dark intensity distribution formed by the interference of scattered light from within or on the surface of the product; the speckle contrast ratio is the ratio of the standard deviation of grayscale to the average grayscale value of a local area.

[0034] In some embodiments, reference Figure 3 The figure is an exemplary flowchart illustrating the determination of the deterioration index of moisture-absorbing agglomerates according to some embodiments of this application. The deterioration index of moisture-absorbing agglomerates is determined by calculating the multi-frame average speckle contrast value of each local region based on the multi-frame speckle images, and then determining the spatial distribution characteristics based on the multi-frame average speckle contrast value of each local region. The spatial distribution characteristics are then compared with the calibration range of a standard sample with known agglomeration levels. The following steps can be used to determine the deterioration index of moisture-absorbing agglomerates: In step 1031, image registration is performed on the multi-frame speckle images, and the registered speckle images are divided into multiple local regions corresponding to the positions, and the speckle contrast value of each local region is calculated frame by frame. In step 1032, according to the positional correspondence of local regions, the speckle contrast ratio of each local region is averaged in multiple frames of speckle images to obtain the multi-frame average speckle contrast ratio of each local region. In step 1033, the median and interquartile range of the multi-frame average speckle contrast ratio of each local region are calculated to form the spatial distribution characteristics, and the range of spatial distribution characteristics pre-calibrated by a standard sample with a known agglomeration level under the same acquisition conditions is obtained. In step 1034, the spatial distribution characteristics are compared with the range of the spatial distribution characteristics to determine the moisture absorption and clumping state level of the product, and the moisture absorption and clumping state level is converted into a moisture absorption and clumping deterioration index according to a preset level assignment rule.

[0035] In practice, each frame of the registered speckle image is divided into local regions with consistent locations. The ratio of the grayscale standard deviation to the grayscale average value is calculated for each local region. When the grayscale average value of a local region is lower than a preset effective grayscale lower limit or higher than a preset saturated grayscale upper limit, the local region is marked as an invalid region. The speckle contrast ratio values ​​of the same effective local region in multiple frames are averaged to obtain the multi-frame average speckle contrast ratio value of that local region.

[0036] Then, the average speckle contrast ratio of all effective local regions across multiple frames is statistically analyzed, and the median and interquartile range are taken as spatial distribution characteristics. Standard samples with no agglomeration, slight agglomeration, moderate agglomeration, and severe agglomeration can be used to establish the spatial distribution characteristic range corresponding to each agglomeration level under the same laser power, exposure parameters, packaging window, and imaging position as the actual monitoring. The agglomeration level of the standard samples is determined by sieving results, compaction measurement results, or manual verification results.

[0037] The spatial distribution characteristics of the current product are compared with the calibration range of the standard sample. A matching moisture absorption and clumping state level is selected and converted into a moisture absorption and clumping deterioration index according to a preset level assignment rule. When the spatial distribution characteristics do not fall within any calibration range or the number of effective local areas is lower than a preset lower limit, the moisture absorption and clumping deterioration index is not output and a re-collection prompt is given. At this time, the monitoring moment is not included in the activity retention rate estimation sequence.

[0038] It should be noted that the deterioration index of moisture absorption and clumping represents the matching level between the spatial distribution characteristics of the current product and the standard sample with known clumping level under the same collection conditions, and is used as the structural state input of the activity prediction model.

[0039] In step 104, the corrected characteristic aldehyde concentration measurement value, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into the activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is a downward trend, it is determined that the product has experienced activity decay.

[0040] In some embodiments, the corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into an activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is downward, it can be determined that the product has experienced activity decay, which can be achieved by the following steps: The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at each monitoring time in the accelerated storage experiment were used as input features. The activity retention rate obtained by standard activity detection of the same sample was used as a label to train and validate the activity prediction model. The corrected characteristic aldehyde concentration measurements, corrected protein oxidation fluorescence peak intensity ratios, and deterioration index of moisture absorption and clumping of the product under monitoring at each continuous monitoring time are input into the validated activity prediction model to obtain the estimated activity retention rate at each monitoring time. The estimated values ​​of the activity retention rate at each monitoring time are sorted according to the monitoring time to form the activity retention rate estimation sequence. The activity retention rate estimation sequence is linearly fitted within the monitoring window with the actual monitoring time as the independent variable. The slope of the fitted line is calculated and used as the change slope. The activity trend is determined based on the sign of the slope of the change, and when the activity trend is a downward trend, it is determined that the product has experienced activity decay.

[0041] The training samples for the activity prediction model use the same product formulation and packaging as the products to be monitored. The activity retention rate is the ratio of the standard activity detection value at a certain monitoring time to the initial standard activity detection value of the same batch; the standard activity detection items are determined based on the target components to be monitored. The model can use a support vector regression model or a random forest regression model, with the corrected characteristic aldehyde concentration measurement value, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping as inputs, and the activity retention rate as the label. Training data is divided into training and validation sets according to sample batches. After validation, the model parameters are frozen for storage monitoring of the same product formulation and packaging. The temperature and water activity conditions for accelerated storage experiments are set within the calibration range of the response deviation curve to ensure that the input features used for training are under the same baseline conditions as the input features used in actual monitoring.

[0042] During actual monitoring, the corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at each monitoring time can be fed into the validated activity prediction model to obtain an estimated activity retention rate, which is then used to form an estimated activity retention rate sequence based on the actual monitoring time. In this application, the estimated value within the monitoring window can be linearly fitted with the actual monitoring time, and the slope of the fitted line is the slope of change. Therefore, even when the monitoring intervals are unequal, the calculation is still based on the actual time.

[0043] Furthermore, the activity trend can be determined by the sign of the slope of change: when the slope is less than zero, the activity trend is downward, and the product is determined to have experienced activity decay; when the slope is greater than or equal to zero, the product is not determined to have experienced activity decay. The activity decay determination can be used to trigger warehouse personnel to review or adjust subsequent monitoring arrangements; product release is still based on the corresponding standard activity test results. When it is necessary to suppress misjudgments caused by random fluctuations in model estimation, the product can also be determined to have experienced activity decay when the slope is less than a preset negative threshold and the number of estimated activity retention rates within the monitoring window is not less than a preset number.

[0044] In addition, this application provides a system for monitoring the activity decay of egg-derived nutritional factors during storage. (Reference) Figure 4The figure is a schematic diagram of a system for monitoring the activity decay of egg-derived nutrient factor products during storage, according to some embodiments of this application. The system includes an acquisition module 401, a processing module 402, and an execution module 403. The modules are described below: The acquisition module 401 is used to perform non-destructive headspace gas sensing and fluorescence detection on stored egg-derived nutrient factor products to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios. The processing module 402 is used to construct a response deviation curve of the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio relative to the reference temperature and reference water activity based on the calibration sample with the same activity level confirmed by standard activity detection, obtain the current temperature and current water activity value, and when the current temperature and current water activity value are within the calibration range of the response deviation curve, obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio according to the response deviation curve. In specific implementation, the processing module 402 is also used to acquire multi-frame speckle images of egg source nutritional factor products, calculate the multi-frame average speckle contrast ratio of each local area based on the multi-frame speckle images, and then determine the spatial distribution characteristics based on the multi-frame average speckle contrast ratio of each local area. The spatial distribution characteristics are compared with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption agglomeration. The execution module 403 is used to input the corrected characteristic aldehyde concentration measurement value, the corrected protein oxidation fluorescence peak intensity ratio and the deterioration index of moisture absorption and clumping at continuous monitoring time into the activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence, calculate the slope of the activity retention rate estimation sequence within the monitoring window, determine the activity trend based on the slope of the change, and determine that the product has experienced activity decay when the activity trend is a downward trend.

[0045] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for monitoring the activity decay of egg-derived nutritional factor products during storage.

[0046] In some embodiments, reference Figure 5 The figure is a schematic diagram of a computer device for monitoring the activity decay of egg-derived nutrient factor products during storage, according to some embodiments of this application. The method for monitoring the activity decay of egg-derived nutrient factor products during storage in the above embodiments can be implemented through... Figure 5 The computer device shown is used to implement this, and the computer device includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0047] The processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0048] The communication bus 502 can be used to transmit information between the aforementioned components.

[0049] The memory 503 may be a read-only memory (ROM), random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), other optical disc storage, disk storage, or other media capable of storing instructions or data structures and accessible by a computer. The memory 503 may be independently configured and connected to the processor 501 via the communication bus 502, or it may be integrated with the processor 501.

[0050] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0051] The communication interface 504 uses a transceiver-type device to communicate with other devices or communication networks, including Ethernet, radio access network (RAN), or wireless local area network (WLAN).

[0052] In a specific implementation, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor is one or more devices, circuits, and / or processing cores used to process data or computer program instructions.

[0053] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0054] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for monitoring the activity decay of egg-derived nutrient factor products during storage.

[0055] The above embodiments are used to illustrate the technical solutions of this application, and are not intended to limit its scope; those skilled in the art can combine or make equivalent substitutions to the embodiments without departing from the technical concept of this application.

[0056] All equivalent substitutions or improvements made within the technical concept of this application shall have their applicable scope determined in conjunction with the specific technical solution.

Claims

1. A method for monitoring the activity decay of egg-derived nutritional factor products during storage, characterized in that, include: Non-destructive headspace gas sensing and fluorescence detection were used to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios for stored egg-derived nutrient products. Based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed between the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity are obtained according to the response deviation curve. Collect multi-frame speckle images of egg source nutritional factor products, calculate the multi-frame average speckle contrast value of each local area based on the multi-frame speckle images, and then determine the spatial distribution characteristics based on the multi-frame average speckle contrast value of each local area. Compare the spatial distribution characteristics with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption agglomeration. The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into an activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is a downward trend, it is determined that the product has experienced activity decay.

2. The method as described in claim 1, characterized in that, Obtaining the characteristic aldehyde concentration measurement value and the ratio of protein oxidation fluorescence peak intensity specifically includes: The gas in the headspace of the product packaging is selectively responded to by a gas sensor array to obtain the response signal of the target characteristic aldehyde. Using a pre-established working curve between the target characteristic aldehyde response signal and concentration, the response signal of the target characteristic aldehyde is converted into a characteristic aldehyde concentration measurement value; The product is irradiated sequentially with a preset reference excitation wavelength and a preset oxidation indication excitation wavelength using a fiber optic fluorescence probe, and the corresponding fluorescence spectra are collected. Extract the reference peak intensity and oxidation indicator peak intensity after normalization of the instrument response from the corresponding fluorescence spectrum, calculate the ratio of the oxidation indicator peak intensity to the reference peak intensity, and obtain the protein oxidation fluorescence peak intensity ratio.

3. The method as described in claim 2, characterized in that, The gas sensor array consists of multiple semiconductor gas sensors with different aldehyde molecule selectivity sensitive membranes.

4. The method as described in claim 1, characterized in that, Based on calibration samples with the same activity level confirmed by standard activity testing, a response deviation curve is constructed for the measured value of the characteristic aldehyde concentration and the ratio of the protein oxidation fluorescence peak intensity relative to the reference temperature and reference water activity. The current temperature and current water activity value are obtained. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the corrected measured value of the characteristic aldehyde concentration and the corrected ratio of the protein oxidation fluorescence peak intensity are obtained according to the response deviation curve, specifically including: Multiple temperature and water activity calibration conditions were set up. Calibration samples with the same activity level confirmed by standard activity testing were selected. After water balance was reached under each calibration condition, the characteristic aldehyde concentration and the ratio of protein oxidation fluorescence peak intensity were measured during the preset calibration period. Using the characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratio at reference temperature and reference water activity as references, the response deviations of characteristic aldehyde concentration and protein oxidation fluorescence peak intensity ratio under each calibration condition were calculated, and a response deviation curve indexed by temperature and water activity was constructed. The current temperature is obtained by a temperature sensor, and the current water activity value is obtained by testing parallel samples of the same batch and under the same storage conditions as the monitored product using a water activity meter according to a preset calibration cycle. When the current temperature and current water activity value are within the calibration range of the response deviation curve, the characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are found using the current temperature and current water activity value as indexes. The characteristic aldehyde concentration response deviation and the protein oxidation fluorescence peak intensity ratio response deviation are then subtracted from the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio, respectively, to obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio.

5. The method as described in claim 4, characterized in that, The calibration samples selected, which have been confirmed by standard activity testing to have the same activity level, specifically include: Multiple calibration sample groups are set up from the same batch of products with the same product formula and packaging form. Each calibration sample group includes a sensor measurement sample and a corresponding parallel test sample. Before and after the calibration conditions reach moisture balance, the parallel test samples are subjected to standard activity tests. Only the sensor measurement samples in the calibration sample group whose test results are all within the same preset activity retention rate equivalent range are determined as the calibration samples.

6. The method as described in claim 1, characterized in that, The acquisition of multi-frame speckle images of egg-derived nutritional factor products specifically includes: The product is illuminated by a laser light source through a transparent area or a reserved detection window of the product packaging, causing the product to scatter light and form a scattered light field carrying information about the product's microstructure. Keeping the laser power, camera exposure parameters, and product imaging position constant, multiple speckle images formed by the interference of the scattered light field are continuously acquired at fixed time intervals.

7. The method as described in claim 6, characterized in that, The laser source is a semiconductor laser diode.

8. The method as described in claim 1, characterized in that, The average speckle contrast ratio of each local region is calculated based on the multi-frame speckle images. Then, the spatial distribution characteristics are determined based on the average speckle contrast ratio of each local region. These spatial distribution characteristics are compared with the calibration range of a standard sample with known agglomeration levels to determine the deterioration index of the moisture-absorbing agglomerates. Specifically, this includes: Image registration is performed on the multi-frame speckle images, and the registered speckle images are divided into multiple local regions corresponding to the positions, and the speckle contrast value of each local region is calculated frame by frame. Based on the positional correspondence of local regions, the average speckle contrast ratio of each local region is obtained by averaging the speckle contrast ratio of each local region across multiple frames of speckle images. The median and interquartile range of the multi-frame average speckle contrast ratio of each local region are calculated to form the spatial distribution characteristics, and the range of spatial distribution characteristics pre-calibrated by a standard sample with a known agglomeration level under the same acquisition conditions is obtained. The spatial distribution characteristics are compared with the range of the spatial distribution characteristics to determine the moisture absorption and clumping status level of the product, and the moisture absorption and clumping status level is converted into a moisture absorption and clumping deterioration index according to the preset level assignment rules.

9. The method as described in claim 1, characterized in that, The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring times are input into an activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence. The slope of the activity retention rate estimation sequence within the monitoring window is calculated, and the activity trend is determined based on the slope. When the activity trend is downward, it is determined that the product has experienced activity decay. Specifically, this includes: The corrected characteristic aldehyde concentration measurements, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at each monitoring time in the accelerated storage experiment were used as input features. The activity retention rate obtained by standard activity detection of the same sample was used as a label to train and validate the activity prediction model. The corrected characteristic aldehyde concentration measurements, corrected protein oxidation fluorescence peak intensity ratios, and deterioration index of moisture absorption and clumping of the product under monitoring at each continuous monitoring time are input into the validated activity prediction model to obtain the estimated activity retention rate at each monitoring time. The estimated values ​​of the activity retention rate at each monitoring time are sorted according to the monitoring time to form the activity retention rate estimation sequence. The activity retention rate estimation sequence is linearly fitted within the monitoring window with the actual monitoring time as the independent variable. The slope of the fitted line is calculated and used as the change slope. The activity trend is determined based on the sign of the slope of the change, and when the activity trend is a downward trend, it is determined that the product has experienced activity decay.

10. A monitoring system for the activity decay of egg-derived nutritional factor products during storage, characterized in that, include: The acquisition module is used to perform non-destructive headspace gas sensing and fluorescence detection on stored egg-derived nutrient factor products to obtain characteristic aldehyde concentration measurements and protein oxidation fluorescence peak intensity ratios. The processing module is used to construct a response deviation curve of the characteristic aldehyde concentration measurement value and the protein oxidation fluorescence peak intensity ratio relative to the reference temperature and reference water activity based on calibration samples with the same activity level confirmed by standard activity detection, obtain the current temperature and current water activity value, and when the current temperature and current water activity value are within the calibration range of the response deviation curve, obtain the corrected characteristic aldehyde concentration measurement value and the corrected protein oxidation fluorescence peak intensity ratio according to the response deviation curve. The processing module is also used to acquire multi-frame speckle images of egg source nutritional factor products, calculate the multi-frame average speckle contrast value of each local area based on the multi-frame speckle images, and then determine the spatial distribution characteristics based on the multi-frame average speckle contrast value of each local area. The spatial distribution characteristics are compared with the calibration range of standard samples with known agglomeration grades to determine the deterioration index of moisture absorption agglomeration. The execution module is used to input the corrected characteristic aldehyde concentration measurement value, the corrected protein oxidation fluorescence peak intensity ratio, and the deterioration index of moisture absorption and clumping at continuous monitoring time into the activity prediction model pre-trained based on standard activity detection data to form an activity retention rate estimation sequence, calculate the slope of the activity retention rate estimation sequence within the monitoring window, determine the activity trend based on the slope, and determine that the product has experienced activity decay when the activity trend is a downward trend.

Images