A method for detecting the sterilization effect of fermented food based on pulsed electric field

Abstract

The present invention relates to the field of data processing technology, and in particular to a method for detecting the bactericidal effect of fermented food based on a pulsed electric field, the method comprising: obtaining high-spectral data of multiple positions in each fermented food; determining the core characteristic wavelength of each fermented food and the representative position points under the core characteristic wavelength according to the spectral reflectance of each position point of each fermented food at different characteristic wavelengths; clustering each fermented food according to the core characteristic wavelength of each fermented food and the representative position points under the core characteristic wavelength to obtain multiple clusters; using the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster, detecting the bactericidal effect of the fermented food in each cluster. The present invention improves the detection accuracy of the bactericidal effect of fermented food.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a method for detecting the sterilization effect of fermented food based on a pulse electric field. Background Art

[0002] Pulsed Electric Field (PEF) is a technology that uses high-voltage electric pulses to treat liquid or semi-solid foods. It is mainly used to sterilize, preserve and improve the quality of fermented foods. Compared with traditional heat treatment methods, PEF technology can better preserve the nutrients and flavor of fermented foods. In the process of using PEF technology to sterilize fermented foods, fermented foods will produce fermentation liquid during the production process, and the fermentation liquid will not be completely discharged during packaging. Therefore, there are differences in the fermentation liquid content between packaged fermented foods, and this difference will affect the sterilization effect of PEF.

[0003] In some scenarios, in order to evaluate the bactericidal effect of fermented foods after PEF treatment, hyperspectral technology is usually used to evaluate the bactericidal effect of fermented foods. However, due to the difference in the content of fermentation liquid in fermented foods, the bactericidal effect of fermented foods is uneven, and the data shown by fermented foods with different fermentation liquid contents in hyperspectral detection are inconsistent. It is difficult to establish a detection standard for bactericidal effect detection using hyperspectral technology, resulting in low detection accuracy of the bactericidal effect of fermented foods. Summary of the invention

[0004] In order to solve the technical problem of low accuracy in detecting the bactericidal effect of fermented food, the present invention aims to provide a method for detecting the bactericidal effect of fermented food based on a pulsed electric field. The technical solution adopted is as follows:

[0005] The embodiment of the present invention provides a method for detecting the sterilization effect of fermented food based on a pulse electric field, comprising: obtaining hyperspectral data of multiple position points in each fermented food, the fermented food being food sterilized by a pulse electric field, and the hyperspectral data including spectral reflectances at different characteristic wavelengths; determining a core characteristic wavelength of each fermented food and a representative position point at the core characteristic wavelength according to the spectral reflectances of each position point of each fermented food at different characteristic wavelengths, the spectral emissivity of each position point at the core characteristic wavelength being consistent and the fermentation liquid in the fermented food being evenly distributed, and the representativeness of the spectral reflectance intensity of the representative position point at the core characteristic wavelength being greater than a first threshold; clustering each fermented food according to the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength to obtain multiple clusters; detecting the sterilization effect of the fermented food in each cluster by using the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster.

[0006] Optionally, determining the core characteristic wavelength of each fermented food and the representative position points at the core characteristic wavelength based on the spectral reflectance of each position point of each fermented food at different characteristic wavelengths includes: determining the first average spectral reflectance and standard deviation of the spectral reflectance of each position point at the current characteristic wavelength, and the absolute value of the first difference between the spectral reflectance of each position point at the current characteristic wavelength and the first average spectral reflectance; determining the core characteristic wavelength of the fermented food based on the first average spectral reflectance, the absolute value of the first difference, the number of position points and the standard deviation; determining the neighborhood window of each position point at the current core characteristic wavelength; determining the prominence of the position point belonging to the neighborhood window based on the spectral reflectance of the neighborhood position points within the neighborhood window and the spectral reflectance of the position points belonging to the neighborhood window; determining the first representativeness of the spectral reflectance intensity of the position points belonging to the neighborhood window based on the spectral reflectance of the neighborhood position points within the neighborhood window and the second average spectral reflectance and prominence of the spectral reflectance of the position points belonging to the neighborhood window; and determining the representative position point at the current core characteristic wavelength using the first representativeness.

[0007] Optionally, determining the core characteristic wavelength of the fermented food based on the first average spectral reflectance, the absolute value of the first difference, the number of position points and the standard deviation includes: determining the information expression degree of the current characteristic wavelength based on the first average spectral reflectance, the absolute value of the first difference and the number of position points; determining the uniformity of the fermentation liquid distribution of the current characteristic wavelength based on a preset value and the standard deviation; normalizing the first product of the information expression degree and the uniformity of the fermentation liquid distribution to obtain the core expression degree of the current characteristic wavelength; and determining the characteristic wavelength whose core expression degree is greater than or equal to a second threshold as the core characteristic wavelength of the fermented food.

[0008] Optionally, after using the first representativeness to determine the representative position point at the current core characteristic wavelength, the method also includes: determining the core characteristic wavelength with the smallest core expression difference from the current core characteristic wavelength as the adjacent core characteristic wavelength; determining the second representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs at the adjacent core characteristic wavelength; using the first representativeness to determine the representative position point at the current core characteristic wavelength includes: determining the comprehensive representativeness of the position point to which the neighborhood window belongs at the current core characteristic wavelength based on the first representativeness and the second representativeness; determining the position point whose comprehensive representativeness is greater than a third threshold as the representative position point.

[0009] Optionally, determining the comprehensive representativeness of the position point to which the neighborhood window belongs at the current core characteristic wavelength based on the first representativeness and the second representativeness includes: calculating the absolute value of the second difference between the first representativeness and the second representativeness and the sum of a preset value; normalizing the first ratio of the first representativeness and the sum to obtain the comprehensive representativeness of the position point to which the neighborhood window belongs.

[0010] Optionally, determining the information expression degree of the current characteristic wavelength based on the first average spectral reflectance, the absolute value of the first difference and the number of position points includes: calculating a second ratio of a first superposition value of the absolute values ​​of each first difference to the number of position points, and a second product of the second ratio and the first average spectral reflectance; normalizing the inverse of the sum of the preset value and the second product to obtain the information expression degree of the current characteristic wavelength.

[0011] Optionally, determining the prominence of the position points to which the neighborhood window belongs based on the spectral reflectance of the neighborhood position points within the neighborhood window and the spectral reflectance of the position points to which the neighborhood window belongs includes: determining a third ratio of the spectral reflectance of each neighborhood position point within the neighborhood window to the spectral reflectance of the position points to which the neighborhood window belongs; and determining the ratio of the second superposition value of each third ratio to the number of neighborhood position points as the prominence of the position points to which the neighborhood window belongs.

[0012] Optionally, determining the first representativeness of the spectral reflectance intensity of the position point belonging to the neighborhood window based on the spectral reflectance of the neighborhood position point within the neighborhood window and the second average spectral reflectance and prominence of the spectral reflectance of the position point belonging to the neighborhood window includes: determining a fourth ratio of the prominence to the second average spectral reflectance as the first representativeness.

[0013] Optionally, clustering each fermented food according to its core characteristic wavelength and representative position points under the core characteristic wavelength to obtain multiple clusters includes: determining the second difference between the corresponding core characteristic wavelengths in each fermented food and the Euclidean distance between the corresponding representative position points; calculating the third product between the second difference and the Euclidean distance; normalizing the third superposition value of each third product to obtain the hyperspectral performance distance between each fermented food; clustering each fermented food according to the hyperspectral performance distance to obtain multiple clusters.

[0014] Optionally, utilizing the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster to detect the sterilization effect of the fermented foods in each cluster includes: calculating a fifth ratio of the number of fermented foods in the current cluster to the maximum number; when the fifth ratio is less than or equal to a fourth threshold, determining that the sterilization effect of the fermented foods in the current cluster is poor, indicating that the sterilization treatment of the fermented foods in the current cluster is incomplete.

[0015] The present invention has the following beneficial effects: by acquiring hyperspectral data of multiple position points in each fermented food, the fermented food is food sterilized by a pulse electric field, and the hyperspectral data includes spectral reflectances at different characteristic wavelengths; then, according to the spectral reflectances of each position point of each fermented food at different characteristic wavelengths, the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength are determined, the spectral emissivity of each position point at the core characteristic wavelength is consistent and the fermentation liquid in the fermented food is evenly distributed, and the representativeness of the spectral reflectance intensity of the representative position point at the core characteristic wavelength is greater than a first threshold value; secondly, according to the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength, each fermented food is clustered to obtain multiple clusters; finally, the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster are used to detect the sterilization effect of the fermented food in each cluster.

[0016] In this way, the embodiment of the present invention determines the core characteristic wavelength in the hyperspectral data and the representative position points under the core characteristic wavelength based on the characteristics of the fermented food itself. The fermented foods are clustered according to the core characteristic wavelength and the representative position points. The bactericidal effects of the fermented foods in each cluster are similar. The number of fermented foods in the cluster and the maximum number are used to evaluate the bactericidal effects of the fermented foods in the cluster. In this way, the bactericidal effects of fermented foods are classified and evaluated, which solves the problem of difficulty in establishing detection standards for bactericidal effect detection using hyperspectral technology due to the uneven bactericidal effects of fermented foods and the inconsistent data shown by fermented foods with different fermentation liquid contents in hyperspectral detection, thereby improving the detection accuracy of the bactericidal effects of fermented foods. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flow chart of a method for detecting the sterilization effect of fermented food based on a pulse electric field provided by an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of the module composition of a pulsed electric field-based fermented food sterilization effect detection system provided by an embodiment of the present invention;

[0020] Figure 3 A schematic diagram of the structure of a fermented food sterilization effect detection system based on a pulse electric field according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, characteristics and effects of a fermented food sterilization effect detection method based on a pulsed electric field proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0022] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0023] The specific scheme of the method for detecting the sterilization effect of fermented food based on pulse electric field provided by the present invention is described in detail below in conjunction with the accompanying drawings.

[0024] Embodiment 1:

[0025] See also Figure 1 , which shows a flow chart of a method for detecting the sterilization effect of fermented food based on a pulse electric field provided by an embodiment of the present invention, comprising:

[0026] S101, obtaining hyperspectral data of multiple locations in each fermented food.

[0027] Among them, fermented food is food that has been sterilized by a pulsed electric field, and the hyperspectral data includes spectral reflectance at different characteristic wavelengths.

[0028] Specifically, fermented foods will be packaged after production and processing, such as putting fermented foods into bottles or sealed packaging bags, and then they will be sterilized by pulsed electric fields. After the pulsed electric field sterilization is completed, the hyperspectral data of all fermented foods in the current production batch can be obtained through a hyperspectral imager. Hyperspectral data includes spectral reflectance at different characteristic wavelengths. Different characteristic wavelengths include wavelengths in the visible light band, wavelengths in the near-infrared band, wavelengths in the short-wave infrared band, and wavelengths in the mid-infrared band, etc. The wavelength ranges of different bands are different. For example, the wavelength range in the visible light band is approximately between 400 nanometers and 700 nanometers, the wavelength range in the near-infrared band is approximately between 700 nanometers and 1000 nanometers, the wavelength range in the short-wave infrared band is approximately between 1000 nanometers and 2500 nanometers, and the wavelength range in the mid-infrared band is approximately between 3 microns and 5 microns or wider.

[0029] Further, for each fermented food, there are differences in the fermentation liquid content between the fermented foods, which will also cause differences in their hyperspectral data. This is because when packaging fermented foods, the fermented foods are generally packaged into bottles or plastic packaging by machines and then packaged. This process will ignore the amount of fermentation liquid content in the fermented food. In this way, although there are differences in the content between the fermented liquids, the fermented liquid content will usually tend to be within a certain range. When the packaged fermented food is sterilized by PEF, although the fermentation liquid content will affect the current uniformity of PEF, for fermented foods with the same or similar fermentation liquid content, it will correspond to similar hyperspectral data. When the fermentation liquid content deviates from the normal range or is in the normal range but its hyperspectral data deviates, the corresponding incomplete sterilization problem occupies a minority. Therefore, the embodiment of the present invention extracts and clusters the hyperspectral data of fermented foods. The number of fermented foods in the cluster is large, indicating that the sterilization effect is good. The number of fermented foods in the cluster is small, and there may be an incomplete sterilization problem. Therefore, the sterilization effect of fermented foods can be detected by the number of fermented foods in the cluster.

[0030] S102, determining the core characteristic wavelength of each fermented food and the representative position point under the core characteristic wavelength according to the spectral reflectance of each position point of each fermented food under different characteristic wavelengths.

[0031] Among them, the spectral emissivity of each position point at the core characteristic wavelength is consistent and the fermentation liquid is evenly distributed in the fermented food, and the representativeness of the spectral reflectivity intensity of the representative position point at the core characteristic wavelength is greater than the first threshold.

[0032] Specifically, the embodiment of the present invention obtains the spectral reflectance of each position point of the fermented food at different characteristic wavelengths through the above steps, and the lower the spectral reflectance, the stronger the absorption degree of the material composition and content in the fermented food to the light at the current characteristic wavelength, so that the inherent characteristics of the fermented food, such as the component information in the fermented food, can be more reflected. For fermented foods, due to the difference in the content of their fermented liquid and the incomplete sterilization of the pulsed electric field, the spectral reflectance of each fermented food at different characteristic wavelengths will be different. Therefore, the embodiment of the present invention analyzes the spectral reflectance at the characteristic wavelength of each fermented food, and determines the core characteristic wavelength that can better represent the characteristics of the current fermented food. Further, each characteristic wavelength of each fermented food is analyzed, and each position point based on the fermented food has a corresponding spectral reflectance at each characteristic wavelength. Since fermented foods usually contain the same food raw material, the spectral reflectance of each position point of the fermented food at the same characteristic wavelength is also consistent, so when the spectral reflectance is lower and the spectral reflectance of each position point is more consistent, the component information of the fermented food at the current characteristic wavelength can be more reflected. At the same time, when the spectral reflectance of each position point under the same characteristic wavelength is more consistent, it also reflects the uniformity of the distribution of the fermentation liquid in the fermented food. In this way, the pulsed electric field can be evenly distributed in the liquid, ensuring that each position of the fermented food is subjected to a similar electric field strength. In this way, the sterilization effect will be more consistent and effective. Therefore, the embodiment of the present invention determines the core characteristic wavelength of each fermented food by the consistency of the spectral reflectance of each position point and the uniformity of the fermentation liquid in the fermented food.

[0033] Furthermore, the core characteristic wavelength of the fermented food is determined based on the characteristic performance of each fermented food itself, and the embodiment of the present invention here selects representative position points that can better reflect the characteristics of the fermented food itself based on all position points of each fermented food, so as to reduce the interference of unnecessary factors in subsequent analysis. Therefore, here, the embodiment of the present invention compares the representativeness of the reflectivity intensity of each position point with that of the neighboring position points, and analyzes the representative continuity of the reflectivity intensity of the same position point under different core characteristic wavelengths, so as to select representative position points.

[0034] Based on the above analysis process, as an optional embodiment of the present invention, when determining the core characteristic wavelength of each fermented food and the representative position point under the core characteristic wavelength, first determine the first average spectral reflectance and standard deviation of the spectral reflectance of each position point at the current characteristic wavelength, and the absolute value of the first difference between the spectral reflectance of each position point at the current characteristic wavelength and the first average spectral reflectance; then determine the core characteristic wavelength of the fermented food according to the first average spectral reflectance, the absolute value of the first difference, the number of position points and the standard deviation; then determine the neighborhood window of each position point at the current core characteristic wavelength; secondly, determine the prominence of the position point belonging to the neighborhood window according to the spectral reflectance of the neighborhood position point within the neighborhood window and the spectral reflectance of the position point belonging to the neighborhood window; and determine the first representativeness of the spectral reflectance intensity of the position point belonging to the neighborhood window according to the spectral reflectance of the neighborhood position point within the neighborhood window and the second average spectral reflectance and prominence of the spectral reflectance of the position point belonging to the neighborhood window; finally, determine the representative position point under the current core characteristic wavelength using the first representativeness.

[0035] Specifically, for each fermented food, the first average spectral reflectance is obtained by superimposing the spectral reflectances of each position point of the fermented food at the current characteristic wavelength and then averaging them. The calculation process of the standard deviation is: first calculate the square of the difference between the spectral reflectance of each position point at the current characteristic wavelength and the first average spectral reflectance and sum them to obtain the sum of squared deviations. Then calculate the ratio of the sum of squared deviations to the number of each position point to obtain the variance. Finally, take the square root of the variance to obtain the standard deviation.

[0036] Furthermore, the neighborhood window of the position point at the current core characteristic wavelength refers to a window of D*D selected as the neighborhood window based on the current position point as the center, where D is a natural number greater than 1. Among them, D can be taken according to the actual scene, and the value is 9 in the embodiment of the present invention. The neighborhood window includes the position point at the center and other position points of the fermented food. Among them, the second average spectral reflectance is obtained by superimposing the spectral reflectance of all neighborhood position points in the neighborhood window and the spectral reflectance of the position point to which the neighborhood window belongs and then averaging them.

[0037] Furthermore, the prominence of the position point to which the neighborhood window belongs refers to the degree of difference between the spectral reflectance of the position point and the spectral reflectance of each neighborhood position point in its neighborhood window. Among them, the smaller the spectral reflectance of the position point to which the neighborhood window belongs, the higher the prominence of the position point to which the neighborhood window belongs, and the more it can reflect the inherent characteristics of the fermented food. The first representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs refers to whether there are more important component expressions or structural features in the current neighborhood window. The higher the first representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs, the more important component expressions or structural features are in the current neighborhood window.

[0038] Furthermore, when determining the core characteristic wavelength of the fermented food, as an optional embodiment of the present invention, the information expression degree of the current characteristic wavelength is first determined based on the first average spectral reflectance, the absolute value of the first difference, and the number of position points; then the uniformity of the fermentation liquid distribution of the current characteristic wavelength is determined based on the preset value and the standard deviation; then the first product of the information expression degree and the uniformity of the fermentation liquid distribution is normalized to obtain the core expression degree of the current characteristic wavelength; finally, the characteristic wavelength whose core expression degree is greater than or equal to the second threshold is determined as the core characteristic wavelength of the fermented food.

[0039] Specifically, the information expression degree can reflect the degree of embodiment of the inherent characteristics of the fermented food at the current characteristic wavelength. The higher the information expression degree, the more it can reflect the inherent characteristics of the fermented food. Among them, the lower the spectral reflectance, the stronger the absorption degree of the material components and content in the fermented food to the light at the current characteristic wavelength, which can better reflect the inherent characteristics of the fermented food. The uniformity of the fermentation liquid distribution can reflect the uniformity of the fermentation liquid distribution of the fermented food. The greater the uniformity of the fermentation liquid distribution, the higher the uniformity of the fermentation liquid distribution of the fermented food. The second threshold value can be determined according to the actual scenario. In the embodiment of the present invention, the value is 0.75. In the embodiment of the present invention, the characteristic wavelength with a core expression degree greater than or equal to 0.75 is used as the core characteristic wavelength.

[0040] Furthermore, the preset value may be 1, and the uniformity of the fermentation liquid distribution at the current characteristic wavelength may be calculated using the following formula:

[0041]

[0042] In the upper middle part, W represents the uniformity of the distribution of the fermentation liquid in the fermented food at the current characteristic wavelength. Indicates the number of locations in fermented food. Indicates the first location points. Indicates the first Spectral reflectance at each location. express The standard deviation of . Represents the normalization function, which is used to Normalize it so that its value is in the range of 0 to 1.

[0043] Furthermore, when determining the information expression degree of the current characteristic wavelength, firstly, the second ratio of the first superposition value of the absolute values ​​of each first difference and the number of position points, and the second product of the second ratio and the first average spectral reflectivity are calculated; then, the inverse of the sum of the preset value and the second product is normalized to obtain the information expression degree of the current characteristic wavelength.

[0044] Specifically, the preset value can be 1, and the information expression of the current characteristic wavelength can be calculated using the following formula:

[0045]

[0046] In the above formula, Q represents the information expression degree of the current characteristic wavelength. represents the first average spectral reflectance. Indicates the number of locations in fermented food. Indicates the first location points. Indicates the first Spectral reflectance at each location. Indicates the average difference between the spectral reflectance of the current characteristic wavelength and the first average spectral reflectance. Represents the normalization function, which is used to Normalize it so that its value is in the range of 0 to 1.

[0047] Furthermore, the core performance can be calculated using the following formula:

[0048]

[0049] In the above formula, Indicates the core expression of the current characteristic wavelength. Indicates the information expression of the current characteristic wavelength. W indicates the uniformity of the distribution of the fermentation liquid in the fermented food at the current characteristic wavelength. Represents the normalization function, which is used to Normalization is performed so that its value is within the range of 0 to 1. The higher the core expression of the current characteristic wavelength and the higher the information expression of the current characteristic wavelength, the higher the core expression of the current characteristic wavelength.

[0050] Furthermore, when determining the prominence of the position point to which the neighborhood window belongs, first determine the third ratio of the spectral reflectance of each neighborhood position point within the neighborhood window to the spectral reflectance of the position point to which the neighborhood window belongs; then determine the ratio of the second superposition value of each third ratio to the number of neighborhood position points as the prominence of the position point to which the neighborhood window belongs.

[0051] Specifically, the prominence of the location point to which the neighborhood window belongs can be calculated using the following formula:

[0052]

[0053] In the above formula, Indicates the current core characteristic wavelength in fermented food The prominence of a location point. Indicates the current core characteristic wavelength in fermented food The number of neighboring locations in the neighborhood window of a location. It is the first fermented food under the current core characteristic wavelength. Neighborhood location points. It is the first fermented food under the current core characteristic wavelength. The first position in the neighborhood window The spectral reflectance of the neighboring points. Indicates the current core characteristic wavelength in fermented food The spectral reflectance of each location point and its value is not 0.

[0054] Furthermore, the prominence of the location point to which the neighborhood window belongs can only reflect the The outstanding performance of a position point also needs to consider whether the neighborhood window in which it is located is in a more characteristic area. That is, whether each position point can be used as a representative position point is not only related to its own outstanding performance in the neighborhood window, but also to the neighborhood window in which it is located. Therefore, the representative position point can be determined by determining the first representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs. When determining the first representativeness, the fourth ratio of the prominence to the second average spectral reflectance can be determined as the first representativeness.

[0055] Specifically, the first representativeness can be calculated using the following formula:

[0056]

[0057] in, Indicates the current core characteristic wavelength The first representation of the spectral reflectance intensity at each location. Indicates the current core characteristic wavelength in fermented food The second average spectral reflectance of the neighborhood window of the location point and the value is not 0. Indicates the current core characteristic wavelength in fermented food The higher the prominence of the first position point, the greater the possibility that it is a representative position point.

[0058] Furthermore, when the first representativeness is used to determine the representative position point at the current core characteristic wavelength, the position point whose first representativeness is greater than the threshold value can be used as the representative position point. Alternatively, the representative position point can be determined by combining the first representativeness and the second representativeness of the spectral reflectance intensity of the position point belonging to the neighborhood window at the adjacent core characteristic wavelength.

[0059] As an optional embodiment of the present invention, after determining the representative position point at the current core characteristic wavelength using the first representativeness of the spectral reflectance intensity of the position point belonging to the neighborhood window, the method also includes: determining the core characteristic wavelength with the smallest core expression difference from the current core characteristic wavelength as the adjacent core characteristic wavelength; determining the second representativeness of the spectral reflectance intensity of the position point belonging to the neighborhood window at the adjacent core characteristic wavelength; determining the representative position point at the current core characteristic wavelength using the first representativeness includes: determining the comprehensive representativeness of the position point belonging to the neighborhood window at the current core characteristic wavelength based on the first representativeness and the second representativeness; determining the position point with a comprehensive representativeness greater than a third threshold as the representative position point.

[0060] Specifically, the representativeness of the spectral reflectance intensity of each position point at the current core characteristic wavelength is compared with the representativeness of the spectral reflectance intensity of the same position point at its adjacent core characteristic wavelength. If the position point has a relatively high representativeness of the spectral reflectance intensity at different core characteristic wavelengths, it means that the position point can embody representativeness based on multi-level analysis. Among them, the second representativeness of the spectral reflectance intensity at the adjacent core characteristic wavelength can be calculated by the representativeness calculation formula recorded in the above-mentioned embodiment of the present invention, and the embodiment of the present invention will not be repeated here. The third threshold value can be determined according to actual conditions. The embodiment of the present invention takes a value of 0.85, and the position point with a comprehensive representativeness greater than 0.85 is regarded as a representative position point.

[0061] Furthermore, when determining the comprehensive representativeness, as an optional embodiment of the present invention, the absolute value of the second difference between the first representativeness and the second representativeness and the sum of the preset values ​​are first calculated; then the first ratio of the first representativeness and the sum is normalized to obtain the comprehensive representativeness of the position point to which the neighborhood window belongs.

[0062] Specifically, the preset value can be set to 1, and the comprehensive representativeness can be calculated using the following formula:

[0063]

[0064] In the above formula, Indicates the current core characteristic wavelength The comprehensive representativeness of each location point. Indicates the current core characteristic wavelength The first representation of the spectral reflectance intensity at each location. No. The second representative of the spectral reflectance intensity at the adjacent core characteristic wavelength at each location point. Represents the normalization function, which is used to Normalize it so that its value is in the range of 0 to 1.

[0065] Through the above embodiments, the present invention obtains the comprehensive representativeness of each position point under each core characteristic wavelength, and sets the threshold to 0.85 based on the comprehensive representativeness of each position point under each core characteristic wavelength to screen out representative position points under each core characteristic wavelength.

[0066] S103, clustering each fermented food according to the core characteristic wavelength of each fermented food and the representative position points under the core characteristic wavelength to obtain a plurality of clusters.

[0067] Specifically, the core characteristic wavelength of each fermented food and the representative position point under the core characteristic wavelength are obtained through the above embodiment of the present invention. The number of core characteristic wavelengths of each fermented food and the number of representative position points under the core characteristic wavelength are the same, but the wavelength size of the specific core characteristic wavelength and the specific position of the representative position point in the fermented food are different. Therefore, the embodiment of the present invention determines the hyperspectral performance distance between fermented foods by comparing the difference in core characteristic wavelengths between fermented foods and the position difference of representative position points under the same core characteristic wavelength, because the core characteristic wavelength of each fermented food and its representative position point are determined based on their own characteristics. When the hyperspectral performance between fermented foods is more similar, that is, the hyperspectral performance distance is smaller, then its core characteristic wavelength and its representative position point will also be in a similar state. Therefore, the embodiment of the present invention clusters each fermented food based on the hyperspectral performance distance, and clusters fermented foods with similar hyperspectral performance between fermented foods into a cluster.

[0068] When clustering fermented foods, as an optional embodiment of the present invention, first determine the second difference between the corresponding core characteristic wavelengths in each fermented food and the Euclidean distance between the corresponding representative position points; then calculate the third product between the second difference and the Euclidean distance; secondly, normalize the third superposition value of each third product to obtain the hyperspectral performance distance between the two fermented foods; finally, cluster the fermented foods according to the hyperspectral performance distance to obtain multiple clusters.

[0069] Specifically, the hyperspectral performance distance can be calculated using the following formula:

[0070]

[0071] In the above formula, Indicates fermented food and fermented foods The hyperspectral representation distance between them. Indicates the number of core characteristic wavelengths present in fermented foods; Indicates the number of representative location points present in fermented foods. Indicates fermented food and fermented foods No. The second difference between the core characteristic wavelengths. Indicates fermented food No. The coordinates of the location point. Indicates fermented food No. The coordinates of the location point. Indicates fermented food and fermented foods No. represents the Euclidean distance between location points. Represents the normalization function, which is used to Normalize it so that its value is in the range of 0 to 1.

[0072] After obtaining the hyperspectral performance distances between the fermented foods, they are clustered by k-means clustering, wherein the K value in the k-means clustering can be determined according to actual conditions. In the embodiment of the present invention, the K value is 10, so that 10 clusters are obtained in the embodiment of the present invention, and the hyperspectral performance distances of the fermented foods in each cluster are similar.

[0073] S104: Detecting the sterilization effect of the fermented foods in each cluster by using the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster.

[0074] Specifically, when the packaged fermented food is sterilized by PEF, although the fermentation liquid content will affect the current uniformity of PEF, fermented foods with the same or similar fermentation liquid content will correspond to similar hyperspectral data. When the fermentation liquid content deviates from the normal range or is within the normal range but its hyperspectral data deviates, the corresponding problem of incomplete sterilization is in the minority. Because the problem of incomplete sterilization is in the minority, the smaller the size of the cluster corresponding to the fermented food with incomplete sterilization, the corresponding hyperspectral performance is similar to that of the fermented food with normal sterilization. Therefore, the larger the size of a cluster, the more it conforms to normal fermented food. Therefore, the embodiment of the present invention can determine the sterilization effect of the fermented food in the cluster based on the size of the cluster.

[0075] As an optional embodiment of the present invention, when detecting the sterilization effect of fermented foods in each cluster, the fifth ratio of the number of fermented foods in the current cluster to the maximum number is calculated first; then, when the fifth ratio is less than or equal to the fourth threshold, it is determined that the sterilization effect of the fermented food in the current cluster is poor, indicating that the sterilization treatment of the fermented food in the current cluster is incomplete.

[0076] Specifically, in the embodiment of the present invention, the number of fermented foods in each cluster is compared with the maximum number of fermented foods in the current 10 clusters to obtain a fourth ratio. The fourth threshold can be determined according to actual conditions. In the embodiment of the present invention, the value is 0.01. When the fourth ratio corresponding to a cluster is less than or equal to 0.01, it is considered that the sterilization treatment of the fermented food in the cluster is incomplete and there are quality problems.

[0077] The embodiment of the present invention determines the core characteristic wavelength in the hyperspectral data and the representative position points under the core characteristic wavelength based on the characteristics of the fermented food itself. The fermented foods are clustered according to the core characteristic wavelength and the representative position points. The bactericidal effects of the fermented foods in each cluster are similar. The number of fermented foods in the cluster and the maximum number are used to evaluate the bactericidal effects of the fermented foods in the cluster. In this way, the bactericidal effects of fermented foods are classified and evaluated, which solves the problem of difficulty in establishing detection standards for bactericidal effect detection using hyperspectral technology due to the uneven bactericidal effects of fermented foods and the inconsistent data shown by fermented foods with different fermentation liquid contents in hyperspectral detection, thereby improving the detection accuracy of the bactericidal effects of fermented foods.

[0078] Embodiment 2:

[0079] Corresponding to the method for detecting the sterilization effect of fermented food based on pulse electric field provided in the above embodiment, based on the same technical concept, the embodiment of the present invention also provides a system for detecting the sterilization effect of fermented food based on pulse electric field, such as Figure 2 As shown, Figure 2A schematic diagram of the module composition of a fermented food sterilization effect detection system based on a pulse electric field provided in an embodiment of the present invention, the fermented food sterilization effect detection system 200 based on a pulse electric field comprises: an acquisition module 201, used to acquire hyperspectral data of multiple positions in each fermented food, the fermented food is food sterilized by a pulse electric field, and the hyperspectral data includes spectral reflectance at different characteristic wavelengths; a determination module 202, used to determine the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength according to the spectral reflectance of each position point of each fermented food at different characteristic wavelengths, the spectral emissivity of each position point at the core characteristic wavelength is consistent and the fermentation liquid in the fermented food is evenly distributed, and the representativeness of the spectral reflectance intensity of the representative position point at the core characteristic wavelength is greater than a first threshold; a clustering module 203, used to cluster each fermented food according to the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength to obtain multiple clusters; a detection module 204, used to detect the sterilization effect of the fermented food in each cluster by using the number of fermented foods in each cluster and the maximum number of fermented foods in each cluster.

[0080] The embodiment of the present invention determines the core characteristic wavelength in the hyperspectral data and the representative position points under the core characteristic wavelength based on the characteristics of the fermented food itself. The fermented foods are clustered according to the core characteristic wavelength and the representative position points. The bactericidal effects of the fermented foods in each cluster are similar. The number of fermented foods in the cluster and the maximum number are used to evaluate the bactericidal effects of the fermented foods in the cluster. In this way, the bactericidal effects of fermented foods are classified and evaluated, which solves the problem of difficulty in establishing detection standards for bactericidal effect detection using hyperspectral technology due to the uneven bactericidal effects of fermented foods and the inconsistent data shown by fermented foods with different fermentation liquid contents in hyperspectral detection, thereby improving the detection accuracy of the bactericidal effects of fermented foods.

[0081] Optionally, the determination module 202 is also used to determine the first average spectral reflectance and standard deviation of the spectral reflectance of each position point at the current characteristic wavelength, and the absolute value of the first difference between the spectral reflectance of each position point at the current characteristic wavelength and the first average spectral reflectance; determine the core characteristic wavelength of the fermented food according to the first average spectral reflectance, the absolute value of the first difference, the number of position points and the standard deviation; determine the neighborhood window of each position point at the current core characteristic wavelength; determine the prominence of the position point belonging to the neighborhood window according to the spectral reflectance of the neighborhood position points within the neighborhood window and the spectral reflectance of the position points belonging to the neighborhood window; determine the first representativeness of the spectral reflectance intensity of the position points belonging to the neighborhood window according to the spectral reflectance of the neighborhood position points within the neighborhood window and the second average spectral reflectance and prominence of the spectral reflectance of the position points belonging to the neighborhood window; and determine the representative position point at the current core characteristic wavelength using the first representativeness.

[0082] Optionally, the determination module 202 is also used to determine the information expression degree of the current characteristic wavelength based on the first average spectral reflectance, the absolute value of the first difference and the number of position points; determine the uniformity of the fermentation liquid distribution of the current characteristic wavelength based on a preset value and a standard deviation; normalize the first product of the information expression degree and the uniformity of the fermentation liquid distribution to obtain the core expression degree of the current characteristic wavelength; and determine that the characteristic wavelength whose core expression degree is greater than or equal to a second threshold is the core characteristic wavelength of the fermented food.

[0083] Optionally, the determination module 202 is also used to determine the core characteristic wavelength with the smallest core expression difference from the current core characteristic wavelength as the adjacent core characteristic wavelength; determine the second representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs at the adjacent core characteristic wavelength; and determine the position point whose comprehensive representativeness is greater than a third threshold as the representative position point.

[0084] Optionally, the determination module 202 is further used to calculate the absolute value of the second difference between the first representative and the second representative and the sum of a preset value; normalize the first ratio of the first representative and the sum to obtain the comprehensive representativeness of the position point to which the neighborhood window belongs.

[0085] Optionally, the determination module 202 is also used to calculate the second ratio of the first superposition value of the absolute values ​​of each first difference to the number of position points, and the second product of the second ratio and the first average spectral reflectance; the inverse of the sum of the preset value and the second product is normalized to obtain the information expression of the current characteristic wavelength.

[0086] Optionally, the determination module 202 is also used to determine a third ratio of the spectral reflectance of each neighborhood position point in the neighborhood window to the spectral reflectance of the position point to which the neighborhood window belongs; and determine the ratio of the second superposition value of each third ratio to the number of neighborhood position points as the prominence of the position point to which the neighborhood window belongs.

[0087] Optionally, the determination module 202 is further configured to determine a fourth ratio of the prominence to the second average spectral reflectance as the first representativeness.

[0088] Optionally, the clustering module 203 is also used to determine the second difference between the corresponding core characteristic wavelengths in each fermented food and the Euclidean distance between the corresponding representative position points; calculate the third product between the second difference and the Euclidean distance; normalize the third superposition value of each third product to obtain the hyperspectral performance distance between the two fermented foods; cluster the fermented foods according to the hyperspectral performance distance to obtain multiple clusters.

[0089] Optionally, the detection module 204 is also used to calculate the fifth ratio of the number of fermented foods in the current cluster to the maximum number; when the fifth ratio is less than or equal to the fourth threshold, it is determined that the sterilization effect of the fermented food in the current cluster is poor, indicating that the sterilization treatment of the fermented food in the current cluster is incomplete.

[0090] Embodiment three:

[0091] Corresponding to the fermented food sterilization effect detection method based on pulse electric field provided in the above embodiment, based on the same technical concept, the embodiment of the present invention also provides another fermented food sterilization effect detection system based on pulse electric field, which is used to execute the above fermented food sterilization effect detection method based on pulse electric field. Figure 3 A schematic diagram of a pulse electric field-based fermented food sterilization effect detection system according to an embodiment of the present invention is shown in FIG. Figure 3 The fermented food sterilization effect detection system based on pulse electric field may have relatively large differences due to different configurations or performances, and may include one or more processors 301 and memory 302, the memory 302 is used to store computer programs that can be run on the processor, and the processor 301 is used to execute the program stored in the memory 302 to achieve the above Figure 1 The memory 302 may be a temporary storage or a permanent storage. The application program stored in the memory 302 may include one or more modules (not shown in the figure), each of which may include a series of computer executable instructions in the pulse electric field-based fermented food sterilization effect detection system.

[0092] Furthermore, the processor 301 can be configured to communicate with the memory 302, and execute a series of computer executable instructions in the memory 302 on the fermented food sterilization effect detection system based on pulse electric field. The fermented food sterilization effect detection system based on pulse electric field can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input and output interfaces 305, and one or more keyboards 306.

[0093] Specifically in this embodiment, the fermented food sterilization effect detection system based on pulse electric field includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 The various steps in the method embodiment are similar to those in the method embodiment, and have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described again here.

[0094] It should be noted that the fermented food sterilization effect detection system based on pulse electric field provided in the embodiment of the present invention and the fermented food sterilization effect detection method based on pulse electric field provided in the embodiment of the present invention are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned fermented food sterilization effect detection method based on pulse electric field, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0095] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0096] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for detecting the bactericidal effect of fermented food based on pulse electric field, characterized in that: The method for detecting the bactericidal effect of fermented food based on pulse electric field comprises: Acquire hyperspectral data of multiple positions in each fermented food, wherein the fermented food is food sterilized by a pulsed electric field, and the hyperspectral data includes spectral reflectance at different characteristic wavelengths; Determine the core characteristic wavelength of each fermented food and the representative position point at the core characteristic wavelength according to the spectral reflectance of each position point of each fermented food at different characteristic wavelengths, the spectral emissivity of each position point at the core characteristic wavelength is consistent and the fermentation liquid in the fermented food is evenly distributed, and the representativeness of the spectral reflectance intensity of the representative position point at the core characteristic wavelength is greater than a first threshold value; Clustering each of the fermented foods according to the core characteristic wavelength of each of the fermented foods and representative position points under the core characteristic wavelength to obtain a plurality of clusters; Using the number of fermented foods in each of the clusters and the maximum number of fermented foods in each of the clusters, detecting the sterilization effect of the fermented foods in each of the clusters; Determining the core characteristic wavelength of each fermented food and the representative position point under the core characteristic wavelength according to the spectral reflectance of each position point of each fermented food under different characteristic wavelengths comprises: Determine a first average spectral reflectance and a standard deviation of the spectral reflectance of each of the position points at the current characteristic wavelength, and an absolute value of a first difference between the spectral reflectance of each of the position points at the current characteristic wavelength and the first average spectral reflectance; determining a core characteristic wavelength of the fermented food according to the first average spectral reflectance, the absolute value of the first difference, the number of the position points, and the standard deviation; Determine a neighborhood window of each of the position points at the current core characteristic wavelength; Determine the prominence of the position point to which the neighborhood window belongs according to the spectral reflectance of the neighborhood position point within the neighborhood window and the spectral reflectance of the position point to which the neighborhood window belongs; Determine a first representativeness of the intensity of the spectral reflectance of the position point to which the neighborhood window belongs according to the spectral reflectance of the neighborhood position point within the neighborhood window and the second average spectral reflectance of the spectral reflectance of the position point to which the neighborhood window belongs and the prominence; The first representativeness is used to determine a representative position point at the current core characteristic wavelength.

2. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 1, characterized in that: Determining the core characteristic wavelength of the fermented food according to the first average spectral reflectance, the absolute value of the first difference, the number of the position points, and the standard deviation comprises: Determining the information expression degree of the current characteristic wavelength according to the first average spectral reflectance, the absolute value of the first difference, and the number of the position points; Determining the uniformity of the fermentation liquid distribution of the current characteristic wavelength according to a preset value and the standard deviation; Normalizing a first product of the information expression degree and the uniformity of the fermentation liquid distribution to obtain a core expression degree of the current characteristic wavelength; The characteristic wavelength whose core expression degree is greater than or equal to a second threshold is determined as the core characteristic wavelength of the fermented food.

3. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 2, characterized in that: After determining the representative position point at the current core characteristic wavelength by using the first representativeness, the method further includes: Determine the core characteristic wavelength with the smallest core expression difference from the current core characteristic wavelength as the adjacent core characteristic wavelength; Determine a second representativeness of the spectral reflectance intensity of the position point to which the neighborhood window belongs at the adjacent core characteristic wavelength; The determining of the representative position point at the current core characteristic wavelength by using the first representativeness comprises: Determine the comprehensive representativeness of the position point to which the neighborhood window belongs under the current core characteristic wavelength according to the first representativeness and the second representativeness; The position point whose comprehensive representativeness is greater than a third threshold is determined as the representative position point.

4. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 3, characterized in that: Determining the comprehensive representativeness of the position point to which the neighborhood window belongs under the current core characteristic wavelength according to the first representativeness and the second representativeness includes: Calculating a sum of an absolute value of a second difference between the first representative value and the second representative value and a preset value; A first ratio of the first representative value to the sum value is normalized to obtain a comprehensive representativeness of the position point to which the neighborhood window belongs.

5. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 2, characterized in that: The determining the information expression degree of the current characteristic wavelength according to the first average spectral reflectance, the absolute value of the first difference, and the number of the position points includes: Calculating a second ratio of a first superposition value of the absolute values ​​of each of the first differences to the number of the position points, and a second product of the second ratio and the first average spectral reflectance; The inverse of the sum of the preset value and the second product is normalized to obtain the information expression degree of the current characteristic wavelength.

6. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 1, characterized in that: Determining the prominence of the position point to which the neighborhood window belongs according to the spectral reflectance of the neighborhood position point within the neighborhood window and the spectral reflectance of the position point to which the neighborhood window belongs comprises: Determine a third ratio of the spectral reflectance of each of the neighborhood position points within the neighborhood window to the spectral reflectance of the position point to which the neighborhood window belongs; The ratio of the second superposition value of each of the third ratios to the number of the neighborhood position points is determined as the prominence of the position point to which the neighborhood window belongs.

7. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to claim 1, characterized in that: Determining the first representativeness of the intensity of the spectral reflectance of the position point to which the neighborhood window belongs according to the spectral reflectance of the neighborhood position point within the neighborhood window and the second average spectral reflectance of the spectral reflectance of the position point to which the neighborhood window belongs and the prominence includes: A fourth ratio of the prominence to the second average spectral reflectance is determined as the first representativeness.

8. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to any one of claims 1 to 7, characterized in that: The plurality of clusters obtained by clustering the fermented foods according to the core characteristic wavelength of the fermented foods and the representative position points under the core characteristic wavelength include: Determine the difference between the corresponding core characteristic wavelengths in the fermented foods and the Euclidean distance between the corresponding representative position points; Calculating a third product between the difference between the corresponding core characteristic wavelengths in the pairwise fermented foods and the Euclidean distance; Normalizing the third superposition values ​​of the third products to obtain the hyperspectral distances between any two fermented foods; The fermented foods are clustered according to the hyperspectral representation distance to obtain a plurality of clusters.

9. The method for detecting the bactericidal effect of fermented food based on pulse electric field according to any one of claims 1 to 7, characterized in that: The detecting the sterilization effect of the fermented food in each cluster by using the number of the fermented food in each cluster and the maximum number of the fermented food in each cluster comprises: calculating a fifth ratio of the number of fermented foods in the current cluster to the maximum number; When the fifth ratio is less than or equal to the fourth threshold, it is determined that the sterilization effect of the fermented food in the current cluster is poor, indicating that the sterilization treatment of the fermented food in the current cluster is incomplete.

Images