Method and system for detecting a fermented sourdough for the manufacture of a fermented bakery product

Abstract

The present application relates to the field of dough fermentation spectrum detection, and particularly relates to a method and system for detecting fermented sour dough for making fermented food. The method obtains the abnormal fermentation time of the target dough and the abnormal wavelength in the spectrum graph at the abnormal fermentation time according to the absorbance difference of the same wavelength of the spectrum graph at the same time between the target dough and each other fermented sour dough, obtains the first fermentation abnormality degree according to the number of the abnormal fermentation time and the number change of the abnormal wavelength in the spectrum graph at each abnormal fermentation time, divides the target dough after fermentation into multiple sample blocks, clusters the sample blocks, obtains the second fermentation abnormality degree according to the number of the sample blocks in the clustering cluster, the number of the clustering cluster, and the absorbance difference of each wavelength of the spectrum graph of the sample blocks between each clustering cluster, and evaluates the fermentation quality of the target dough by combining the first and second fermentation abnormality degrees. The present application can improve the accuracy of the evaluation of the sour dough fermentation effect and quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of dough fermentation spectrum detection, and in particular to a fermentation sour dough detection method and system for fermented food manufacturing. BACKGROUND

[0002] In the industrial production and manufacturing of fermented food, the fermentation effect of sour dough directly affects the quality, taste, efficiency, safety and consistency of the final baked product. Since it is necessary to ensure the consistency of the taste of the same type of fermented food, the fermentation of the dough is one of the decisive factors affecting the taste, appearance and nutritional composition of the food. Therefore, it is crucial to detect the fermentation process of sour dough.

[0003] Traditional sour dough fermentation monitoring methods have many limitations, such as being destructive, having limited information acquisition, and being offline, etc. They may not fully reflect the complex changes that occur during the fermentation process. For example, traditional detection indicators such as protein content and acidity may not be sufficient to accurately assess the fermentation effect and quality. In addition, the quality of sour dough food often depends on sensory evaluation, such as taste, smell and appearance. These evaluations have strong subjectivity and are difficult to quantify, which leads to the inability of existing methods to accurately assess the fermentation effect and quality of sour dough. SUMMARY

[0004] In order to solve the technical problem that the existing method cannot accurately assess the fermentation effect and quality of sour dough, the purpose of the present application is to provide a fermentation sour dough detection method and system for fermented food manufacturing, and the technical solution adopted is as follows:

[0005] The present application provides a fermentation sour dough detection method for fermented food manufacturing, which comprises:

[0006] Obtaining the spectrum graph of each fermentation sour dough in the same production batch at each time during the fermentation process, the spectrum graph being composed of absorbance corresponding to different wavelengths;

[0007] Taking any one of the fermentation sour dough as a target dough, obtaining the fermentation state inconsistency degree of each wavelength in the spectrum graph of the target dough at each time according to the difference in absorbance of the same wavelength between the spectrum graph of the target dough and each other fermentation sour dough at the same time; based on the fermentation state inconsistency degree of each wavelength in the spectrum graph of the target dough at each time, screening out the fermentation abnormal time of the target dough and the abnormal wavelength in the spectrum graph of each fermentation abnormal time; according to the number of fermentation abnormal times and the number of abnormal wavelengths in the spectrum graph of each fermentation abnormal time, obtaining the first fermentation abnormality degree of the target dough;

[0008] The target dough is divided into blocks to obtain a plurality of sample blocks of the target dough, and a spectrum of each sample block is obtained. The sample blocks are clustered according to the differences in absorbance of the same wavelength between the spectrum of each sample block, and a plurality of clustering clusters are obtained. The second fermentation abnormality degree of the target dough is obtained according to the number of sample blocks in each clustering cluster, the number of clustering clusters, and the differences in absorbance of each wavelength of the spectrum of the sample blocks between the clustering clusters.

[0009] The fermentation quality of the target dough is evaluated in combination with the first fermentation abnormality degree and the second fermentation abnormality degree.

[0010] Further, the fermentation state inconsistency degree of each wavelength in the spectrum of the target dough at each time point includes:

[0011] The absolute value of the difference in absorbance of the same wavelength between the spectrum of the target dough and the spectrum of each other fermented sour dough at the same time is taken as the absorbance difference value of each wavelength between the spectrum of the target dough and the spectrum of each other fermented sour dough at each time point.

[0012] The cumulative value of the absorbance difference value of each wavelength between the spectrum of the target dough and the spectrum of all other fermented sour dough at each time point is normalized to obtain the fermentation state inconsistency degree of each wavelength in the spectrum of the target dough at each time point.

[0013] Further, the fermentation abnormal time point of the target dough and the abnormal wavelength in the spectrum at each fermentation abnormal time point are screened out.

[0014] The average value of the fermentation state inconsistency degree of all wavelengths in the spectrum of the target dough at each time point is taken as the fermentation state deviation degree of the target dough at each time point.

[0015] The time point at which the fermentation state deviation degree is greater than a preset deviation threshold is taken as the fermentation abnormal time point of the target dough.

[0016] In the spectrum of each fermentation abnormal time point of the target dough, the wavelength at which the fermentation state inconsistency degree is greater than a preset deviation threshold is taken as the abnormal wavelength in the spectrum of each fermentation abnormal time point of the target dough.

[0017] Further, the first fermentation abnormality degree of the target dough includes:

[0018] The first fermentation abnormal time point of the target dough is negatively correlated to obtain a fermentation abnormality occurrence evaluation value of the target dough.

[0019] The number of all fermentation abnormal time points of the target dough is the numerator, the number of all time points in the fermentation process is the denominator, and the ratio is the fermentation abnormal time point proportion of the target dough;

[0020] The fermentation abnormal severity of the target dough is obtained by synthesizing and normalizing the average value of the fermentation state deviation degree of the target dough at all time points, the fermentation abnormality occurrence evaluation value of the target dough, and the fermentation abnormal time point proportion;

[0021] According to the sequence, the number of abnormal wavelengths in the spectrum of the target dough at each fermentation abnormal time point is linearly fitted, and the slope of the fitted straight line is normalized to obtain the abnormal component quantity change trend value of the target dough;

[0022] The set of all abnormal wavelengths in the spectrum of the target dough at each fermentation abnormal time point is taken as the abnormal wavelength set of the target dough at each fermentation abnormal time point, and the abnormal component change evaluation value of the target dough is obtained according to the number of the same abnormal wavelengths in the abnormal wavelength set of the adjacent two fermentation abnormal time points of the target dough and the abnormal component quantity change trend value of the target dough;

[0023] The fermentation abnormal severity of the target dough and the abnormal component change evaluation value are synthesized and normalized to obtain the first fermentation abnormality degree of the target dough.

[0024] Further, the abnormal component change evaluation value of the target dough includes:

[0025] The Jaccard similarity coefficient between the abnormal wavelength sets of the adjacent two fermentation abnormal time points of the target dough is negatively correlated and normalized to obtain the abnormal component change degree of the target dough at the adjacent two fermentation abnormal time points;

[0026] The average value of the abnormal component change degree of the target dough at all adjacent two fermentation abnormal time points is taken as the overall abnormal component change degree of the target dough;

[0027] The overall abnormal component change degree of the target dough and the abnormal component quantity change trend value are synthesized and normalized to obtain the abnormal component change evaluation value of the target dough.

[0028] Further, the obtaining of the plurality of clustering clusters includes:

[0029] According to the order from small to large, the sequence of absorbance of all wavelengths in the spectrum of each sample block is taken as the absorbance sequence of each sample block;

[0030] Euclidean distance of absorbance sequences between any two sample blocks is taken as a distance metric between any two sample blocks;

[0031] Based on the distance metric between any two sample blocks, all sample blocks are clustered to obtain a plurality of clustering clusters.

[0032] Further, the second fermentation abnormality degree of the target dough includes:

[0033] The clustering cluster containing the largest number of sample blocks is taken as the standard clustering cluster of the target dough;

[0034] The number of sample blocks in the standard clustering cluster is taken as the numerator, and the number of all sample blocks of the target dough is taken as the denominator, and the ratio is taken as the sample block number ratio of the standard clustering cluster of the target dough;

[0035] The average value of the distance metric between any two sample blocks in the standard clustering cluster is negatively correlated, and the aggregation degree of the standard clustering cluster of the target dough is obtained;

[0036] The sample block number ratio and the aggregation degree are integrated and negatively correlated to obtain the first fermentation non-uniformity of the target dough;

[0037] The difference between the absorbance sequences of the sample blocks between the standard clustering cluster and other clustering clusters except the standard clustering cluster is obtained to obtain the second fermentation non-uniformity of the target dough;

[0038] The first fermentation non-uniformity, the second fermentation non-uniformity of the target dough and the number of all clustering clusters are integrated and normalized to obtain the second fermentation abnormality degree of the target dough.

[0039] Further, the second fermentation non-uniformity of the target dough includes:

[0040] The average value of the absorbance sequence of all sample blocks in each clustering cluster is taken as the clustering center of each clustering cluster;

[0041] The Euclidean distance of the clustering center between the standard clustering cluster and each other clustering cluster except the standard clustering cluster is taken as the clustering center distance value between the standard clustering cluster and each other clustering cluster;

[0042] The average value of the clustering center distance value between the standard clustering cluster and all other clustering clusters is taken as the second fermentation non-uniformity of the target dough.

[0043] Further, the evaluation of the fermentation quality of the target dough includes:

[0044] comprehensive and normalized, a fermentation abnormality evaluation value of the target dough is obtained;

[0045] If the fermentation abnormality evaluation value of the target dough is greater than a preset abnormality threshold value, the target dough is identified as unqualified in fermentation.

[0046] The present application also provides a fermentation sourdough detection system for fermented food manufacturing, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the fermentation sourdough detection methods for fermented food manufacturing when executing the computer program.

[0047] The present application has the following advantages:

[0048] The present application considers that the existing method cannot accurately evaluate the fermentation effect and quality of sourdough. The present application first collects the spectrum graph of each fermentation sourdough in the same production batch at each time during the fermentation process by using the spectrum detection technology, thereby realizing nondestructive detection of the sourdough. Considering that the absorbance of the same wavelength in the spectrum graph at the same time between the target dough and other fermentation sourdoughs is relatively close when no abnormality occurs during the fermentation process, the fermentation abnormality possibility of each wavelength in the spectrum graph of the target dough at each time can be first reflected by the obtained fermentation state inconsistency degree, and then the fermentation abnormal time of the target dough and the abnormal wavelength in the spectrum graph at each fermentation abnormal time are screened out. The abnormal wavelength can represent the abnormal component of the target dough during the fermentation process. Considering that the target dough has a high abnormality time ratio when an abnormality occurs during the fermentation process, and the number of abnormal wavelengths shows a growth trend in the time sequence, the degree of abnormality of the target dough during the fermentation process can be reflected by the first fermentation abnormality degree, and the target dough is divided into multiple sample blocks, and all the sample blocks are clustered to obtain multiple clustering clusters. The more the number of clustering clusters is, and the greater the difference between the absorbance of each wavelength in the spectrum graph of the sample blocks of each clustering cluster is, the poorer the uniformity of the fermentation of different regions of the target dough is. Therefore, the degree of abnormality of the target dough during the fermentation process can also be reflected from another angle by the second fermentation abnormality degree, and then the fermentation quality of the target dough is accurately evaluated by combining the first fermentation abnormality degree and the second fermentation abnormality degree, thereby improving the accuracy of the evaluation of the fermentation effect and quality of the sourdough. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below will briefly introduce the drawings required by the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0050] Figure 1 A flow chart of a fermentation sour dough detection method for fermented food manufacturing provided by an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purposes, below will combine the drawings and preferred embodiments to specifically describe the fermentation sour dough detection method and system for fermented food manufacturing according to the present application, the specific implementation, structure, features and effects thereof, in detail. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0053] Below will specifically describe the specific scheme of the fermentation sour dough detection method and system for fermented food manufacturing provided by the present application in combination with the drawings.

[0054] Please refer to Figure 1 which shows a flow chart of a fermentation sour dough detection method for fermented food manufacturing provided by an embodiment of the present application, and the method comprises:

[0055] Step S1: acquiring the spectrum diagram of each fermentation sour dough in the same production batch at each time during the fermentation process, the spectrum diagram is composed of absorbance corresponding to different wavelengths.

[0056] During the fermentation process of the fermentation sour dough, the content changes of the organic acid, volatile compounds, protein content and microbial community contained in the fermentation sour dough are the main factors affecting whether the fermentation proceeds normally, and the absorbance of different wavelengths in the spectrum diagram can reflect the content of each component in the fermentation sour dough. Therefore, in the industrial production process of the fermented food manufacturing, the present embodiment first uses a spectrometer to collect the spectrum diagram of each fermentation sour dough in the same production batch at each time during the fermentation process, and the collection frequency of the spectrum diagram is 1 minute collection once, that is, the spectrum diagram of each fermentation sour dough is collected every other minute.

[0057] It should be noted that, in order to ensure the taste consistency of the same type of fermented bread, the fermentation temperature, the fermentation humidity, the water content, the strain and the inoculation amount of each fermented sourdough in the same production batch need to be kept consistent, so as to accurately evaluate the fermentation effect and the fermentation quality of the fermented sourdough.

[0058] Step S2: Taking any one of the fermented sourdoughs as a target dough, obtaining the fermentation state inconsistency degree of each wavelength in the spectrum diagram of each time of the target dough according to the difference in the absorbance of the same wavelength in the spectrum diagram of the same time between the target dough and each other fermented sourdough; screening out the fermentation abnormal time of the target dough and the abnormal wavelength in the spectrum diagram of each fermentation abnormal time based on the fermentation state inconsistency degree of each wavelength in the spectrum diagram of each time of the target dough; and obtaining the first fermentation abnormality degree of the target dough according to the number of the fermentation abnormal time and the number change of the abnormal wavelength in the spectrum diagram of each fermentation abnormal time.

[0059] In the industrial production process of the fermented bread manufacturing, it is necessary to ensure the taste consistency of the same type of fermented bread, that is, to ensure that the ingredient content of each fermented sourdough in the same production batch remains consistent during the fermentation process. In the spectrum detection of the fermented sourdough, the main detected ingredients include organic acids, volatile compounds, protein content and microbial community structure, etc. Different wavelengths represent different ingredients. The greater the absorbance corresponding to the wavelength, the higher the content of the measured ingredient, for example: the absorbance of organic acids is usually measured at 570 nm wavelength, which is used to measure the content of lactic acid and acetic acid. The higher the absorbance, the higher the concentration of these organic acids. In the same fermentation environment, the greater the difference in the absorbance of the same wavelength in the spectrum diagram of the same time between the fermented sourdoughs, the more likely the ingredient corresponding to the wavelength of the fermented sourdough at that time is abnormal. Therefore, any one of the fermented sourdoughs can be taken as a target dough, and the difference in the absorbance of the same wavelength in the spectrum diagram of the same time between the target dough and each other fermented sourdough can be analyzed. The fermentation state inconsistency degree can reflect the possibility of fermentation abnormality of each wavelength in the spectrum diagram of each time of the target dough. Subsequently, the fermentation abnormal time and the abnormal wavelength in the spectrum diagram of each fermentation abnormal time can be screened out based on the fermentation state inconsistency degree, and the degree of abnormality of the target dough in the fermentation process can be accurately analyzed.

[0060] Preferably, in one embodiment of the present application, the method for obtaining the fermentation state inconsistency degree of each wavelength in the spectrum diagram of each time of the target dough specifically comprises:

[0061] The absolute value of the difference in absorbance at the same wavelength of the spectrum at the same moment between the target dough and each other fermented sourdough except the target dough is taken as the absorbance difference value at each wavelength of the spectrum at each moment between the target dough and each other fermented sourdough.

[0062] Normalize the accumulated absorbance difference values ​​of each wavelength of the spectra at each moment between the target dough and all other fermented sourdoughs, and limit the calculation results to range, thereby obtaining the fermentation state inconsistency of each wavelength in the spectrum graph of the target dough at each moment.

[0063] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of values, or activation functions and hyperbolic tangent functions can be used to implement normalization processing, which will not be repeated or limited.

[0064] As an example, in one embodiment of the present invention, the expression for the fermentation state inconsistency of each wavelength in the spectrum graph of the target dough at each moment can be specifically, for example, as follows:

[0065]

[0066] in, The target dough The first The inconsistency of fermentation state at each wavelength; Indicates that the target dough is The first Absorbance at each wavelength; Indicates the first Other sourdoughs in The first Absorbance at each wavelength; Indicates the target dough and other sourdoughs in the The spectrum of the moment The absorbance difference at each wavelength; represents the number of other fermented sourdoughs other than the target dough; Represents the normalization function, used for normalization processing.

[0067] The greater the fermentation state inconsistency degree of the target dough at a certain wavelength in the spectrum diagram at a certain moment in the fermentation process, the more likely it is that the component represented by the wavelength in the spectrum diagram at the moment of the target dough is abnormal. Therefore, the fermentation abnormal moment of the target dough and the abnormal wavelength in the spectrum diagram at each fermentation abnormal moment are screened based on the fermentation state inconsistency degree of each wavelength in the spectrum diagram of the target dough at each moment, so as to facilitate subsequent accurate analysis of the degree of abnormality of the target dough in the entire fermentation process.

[0068] Preferably, in an embodiment of the present application, the method for obtaining the fermentation abnormal moment of the target dough and the abnormal wavelength in the spectrum diagram at each fermentation abnormal moment specifically comprises:

[0069] The average value of the fermentation state inconsistency degree of all wavelengths in the spectrum diagram of the target dough at each moment is taken as the fermentation state deviation degree of the target dough at each moment. The greater the fermentation state deviation value at a certain moment, the greater the overall level of abnormality of the target dough represented by all wavelengths in the spectrum diagram at the moment, that is, the more likely it is that the moment is a fermentation abnormal moment. Therefore, the moment with a fermentation state deviation degree greater than a preset deviation threshold value can be taken as the fermentation abnormal moment of the target dough, wherein the preset deviation threshold value is in the range of 0 to 1. In an embodiment of the present application, the preset deviation threshold value is set to 0.7. The preset deviation threshold value can also be set by the implementer according to the specific implementation scene, which is not limited herein.

[0070] Then, in the spectrum diagram at each fermentation abnormal moment of the target dough, the wavelength with a fermentation state inconsistency degree greater than the preset deviation threshold value is taken as the abnormal wavelength in the spectrum diagram at each fermentation abnormal moment of the target dough. The abnormal wavelength in the spectrum diagram can represent the component with abnormal content of the target dough at the fermentation abnormal moment.

[0071] In the entire fermentation process, the greater the number of fermentation abnormal moments of the target dough and the greater the number of abnormal wavelengths showing a growth trend in the time sequence, the greater the degree of abnormality of the target dough in the fermentation process. Therefore, the number of fermentation abnormal moments of the target dough and the number of abnormal wavelengths in the spectrum diagram at each fermentation abnormal moment can be analyzed, and the first fermentation abnormality degree is obtained to reflect the degree of abnormality of the target dough in the fermentation process. Subsequently, the effect and quality of the fermentation of the target dough can be accurately evaluated based on the first fermentation abnormality degree.

[0072] Preferably, in an embodiment of the present application, the method for obtaining the first fermentation abnormality degree of the target dough specifically comprises:

[0073] First, the numerical value of the first fermentation abnormal time of the target dough is negatively correlated, and the fermentation abnormality evaluation value of the target dough is obtained. The greater the fermentation abnormality evaluation value, the more the target dough appears the fermentation abnormality phenomenon, and the greater the influence on the whole fermentation process, and the more serious the target dough fermentation abnormality.

[0074] The number of all fermentation abnormal times of the target dough is the numerator, and the number of all times in the fermentation process is the denominator. The ratio is the fermentation abnormal time proportion of the target dough. The greater the fermentation abnormal time proportion, the more frequent the target dough occurs in the whole fermentation process. The more serious the target dough fermentation abnormality.

[0075] At the same time, the greater the overall level of the fermentation state deviation degree of the target dough at all times, the more serious the abnormality of the target dough in the fermentation process, so the average value of the fermentation state deviation degree of the target dough at all times, the fermentation abnormality evaluation value of the target dough and the fermentation abnormal time proportion are comprehensively processed and normalized, and the calculation result is limited to the range of 0-1, so as to obtain the fermentation abnormality severity of the target dough.

[0076] In the embodiment of the present application, the average value of the fermentation state deviation degree of the target dough at all times, the fermentation abnormality evaluation value of the target dough and the fermentation abnormal time proportion can be calculated to realize the comprehensive processing of the three, and the subsequent step of comprehensive processing of two or more data can also use the same method.

[0077] As an example, in an embodiment of the present application, the expression of the fermentation abnormality severity of the target dough can be specifically, for example:

[0078]

[0079] Wherein, represents the fermentation abnormality severity of the target dough; represents the average value of the fermentation state deviation degree of the target dough at all times; represents the numerical value of the first fermentation abnormal time of the target dough; represents the fermentation abnormality evaluation value of the target dough; represents the number of all fermentation abnormal times of the target dough; represents the number of all times in the fermentation process; represents the fermentation abnormal time proportion of the target dough; represents a normalization function for normalization processing; represents a preset first adjustment parameter for preventing the denominator from being 0, the value range of​ In an embodiment of the present application, the is set to 0.01, The specific value can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0080] It should be noted that in other embodiments of the present application, negative correlation mapping can also be achieved through other basic mathematical operations, which are not described here.

[0081] Then, according to the time sequence, the number of abnormal wavelengths in the spectrum of the target dough at each fermentation abnormal time is linearly fitted, and the slope of the fitted straight line is normalized to limit the calculation result to the range of , thereby obtaining the abnormal component quantity trend value of the target dough. The greater the abnormal component quantity trend value, the more likely the number of abnormal wavelengths in the spectrum of the target dough at each fermentation abnormal time is in an upward trend in the time sequence, that is, the number of abnormal components contained in the target dough is more likely to be in an upward trend in the time sequence, and thus the greater the degree of fermentation abnormality of the target dough in the fermentation process. In an embodiment of the present application, the least squares method or other fitting methods can be used to achieve linear fitting, which is not limited and described here.

[0082] The set of all abnormal wavelengths in the spectrum of the target dough at each fermentation abnormal time is taken as the abnormal wavelength set of the target dough at each fermentation abnormal time. According to the number of the same abnormal wavelengths in the abnormal wavelength sets of the adjacent two fermentation abnormal times of the target dough, and the abnormal component quantity trend value of the target dough, the abnormal component change evaluation value of the target dough is obtained. The greater the abnormal component change evaluation value, the more likely the number of abnormal components contained in the target dough is in an upward trend in the time sequence, and the more obvious the change in the number of abnormal component types in the time sequence, and thus the greater the degree of fermentation abnormality of the target dough in the fermentation process.

[0083] Preferably, in an embodiment of the present application, the method for obtaining the abnormal component change evaluation value of the target dough specifically comprises:

[0084] The Jaccard similarity coefficient between the abnormal wavelength sets of the adjacent two fermentation abnormal times of the target dough is negatively correlated and normalized, and the calculation result is limited to the range of , thereby obtaining the abnormal component change degree of the target dough at the adjacent two fermentation abnormal times. The greater the abnormal component change degree, the greater the difference in the types of abnormal components of the target dough at the adjacent two fermentation abnormal times, and thus the average value of the abnormal component change degrees of the target dough at all adjacent two fermentation abnormal times can be taken as the overall abnormal component change degree of the target dough.

[0085] Then, the abnormal component overall change degree and the abnormal component quantity change trend value of the target dough are comprehensively processed and normalized, and the calculation result is limited in the range of 0 to 1, so as to obtain the abnormal component change evaluation value of the target dough.

[0086] As an example, in an embodiment of the present application, the expression of the abnormal component change evaluation value of the target dough can be specifically, for example:

[0087]

[0088] wherein, represents the abnormal component change evaluation value of the target dough; represents the abnormal component quantity change trend value of the target dough; represents the Jaccard similarity coefficient between the abnormal wavelength sets of the target dough at the adjacent two fermentation abnormal time points; represents the abnormal component change degree of the target dough at the adjacent two fermentation abnormal time points; represents the abnormal component change degree of the target dough at the adjacent two fermentation abnormal time points; represents the negative correlation normalization processing of the Jaccard similarity coefficient, and in other embodiments of the present application, the negative correlation normalization processing can also be realized by using, for example, a negative exponential function with a natural constant e as the base, which is not limited here; represents the number of fermentation abnormal time points of the target dough, and then represents the number of adjacent two fermentation abnormal time points; represents the abnormal component overall change degree of the target dough; represents the normalization function for normalization processing. Further, the fermentation abnormal severity and the abnormal component change evaluation value of the target dough can be comprehensively processed and normalized, and the calculation result is limited in the range of 0 to 1, so as to obtain the first fermentation abnormality degree of the target dough.

[0089] As an example, in an embodiment of the present application, the expression of the first fermentation abnormality degree of the target dough can be specifically, for example:

[0090]

[0091]

[0092] wherein, represents the first fermentation abnormality degree of the target dough; represents the fermentation abnormal severity of the target dough; represents the abnormal component change evaluation value of the target dough; ​​​​represents a normalization function for normalization processing.

[0093] At this point, the analysis of the degree of abnormality of the target dough during fermentation is completed.

[0094] Step S3: The target dough after fermentation is divided into blocks to obtain a plurality of sample blocks of the target dough, and the spectrum of each sample block is obtained. According to the difference in absorbance of the same wavelength between the spectrum of each sample block, all sample blocks are clustered to obtain a plurality of clustering clusters. According to the number of sample blocks in each clustering cluster, the number of clustering clusters, and the difference in absorbance of each wavelength of the spectrum of the sample blocks between the clustering clusters, a second fermentation abnormality degree of the target dough is obtained.

[0095] The above process is a detection analysis of the entire dough, and does not consider the influence of uneven fermentation of different regions of the same fermented sour dough on the fermentation effect. Therefore, in the embodiment of the present application, the target dough after fermentation is divided into blocks to obtain a plurality of sample blocks of the target dough, and the spectrum of each sample block is obtained by a spectrometer. In one embodiment of the present application, the target dough can be placed in a rectangular three-dimensional container, and the target dough is divided into blocks by, for example, a grid division method, that is, the dough is first cut into a surface layer, a middle layer and a bottom layer in equal amounts from top to bottom, and then each layer is cut into 9 small rectangular blocks in equal amounts, thereby obtaining a plurality of sample blocks. The division method of the target dough can be determined by the implementer according to the specific implementation scene, which is not limited here.

[0096] When the ingredients and ingredient contents of each sample block of the target dough are similar, it means that the fermentation of different regions of the target dough is more uniform, which means that each region of the target dough has experienced a similar fermentation process without local over-fermentation or under-fermentation. Therefore, it indicates that the fermentation effect of the target dough is good, otherwise, it indicates that the fermentation effect of the target dough is poor. Therefore, first, according to the difference in absorbance of the same wavelength between the spectrum of each sample block, all sample blocks are clustered to obtain a plurality of clustering clusters. Subsequently, based on the difference in absorbance of each wavelength between the spectrum of the sample blocks in each clustering cluster, the degree of abnormality of the target dough can be accurately analyzed.

[0097] Preferably, in one embodiment of the present application, the method for obtaining a plurality of clustering clusters specifically comprises:

[0098] First, according to the order of wavelength from small to large, the sequence of absorbance of all wavelengths in the spectrum of each sample block is taken as the absorbance sequence of each sample block.

[0099] The Euclidean distance of the absorbance sequence between any two sample blocks is taken as a distance measurement between any two sample blocks, all sample blocks are clustered based on the distance measurement between any two sample blocks, and a plurality of clustering clusters are obtained, then the fermentation effects of the sample blocks in the same clustering cluster are relatively similar, and the fermentation effects of the sample blocks in different clustering clusters are relatively different, in an embodiment of the present application, the existing K-means clustering algorithm can be selected to realize the clustering operation, and the number of clustering clusters can be determined by using the existing elbow method, in other embodiments of the present application, other clustering algorithms can also be used to realize the clustering operation, which is not limited here.

[0100] The more the number of clustering clusters and the greater the difference in absorbance of each wavelength of the spectral graph of the sample blocks between the clustering clusters, the poorer the uniformity of fermentation of different regions of the target dough, and the greater the degree of fermentation abnormality of the target dough, therefore, the second fermentation abnormality degree of the target dough can be obtained according to the number of sample blocks in each clustering cluster, the number of clustering clusters, and the difference in absorbance of each wavelength of the spectral graph of the sample blocks between the clustering clusters.

[0101] Preferably, in an embodiment of the present application, the method for obtaining the second fermentation abnormality degree of the target dough specifically comprises:

[0102] The clustering cluster containing the largest number of sample blocks is taken as a standard clustering cluster of the target dough, wherein the fermentation of the sample blocks in the standard clustering cluster can represent the fermentation of the whole target dough.

[0103] The number of sample blocks in the standard clustering cluster is taken as the numerator, the number of all sample blocks of the target dough is taken as the denominator, and the ratio is taken as the sample block number ratio of the standard clustering cluster of the target dough, the greater the sample block number ratio of the standard clustering cluster, the more similar the fermentation of most sample blocks of the target dough, and the more uniform the fermentation of the target dough, on the contrary, the more uneven the fermentation of the target dough.

[0104] The average value of the distance measurement between any two sample blocks in the standard clustering cluster is negatively correlated, and the aggregation degree of the standard clustering cluster of the target dough is obtained, the greater the aggregation degree of the standard clustering cluster, the more similar the fermentation of the sample blocks in the standard clustering cluster, and the more uniform the fermentation of the target dough, on the contrary, the more uneven the fermentation of the target dough.

[0105] Therefore, the sample block number ratio and the aggregation degree can be comprehensively mapped and negatively correlated to obtain the first fermentation unevenness of the target dough.

[0106] As an example, in an embodiment of the present application, the expression of the first fermentation unevenness of the target dough can be specifically, for example:

[0107]

[0108] wherein, represents the first fermentation unevenness of the target dough; represents the number of sample blocks in the standard clustering cluster; represents the number of all sample blocks of the target dough; represents the proportion of the number of sample blocks of the standard clustering cluster in the target dough; represents the average value of the distance metric between any two sample blocks in the standard clustering cluster; represents the degree of aggregation of the standard clustering cluster of the target dough; represents an exponential function with a natural constant as the base, for negative correlation mapping; represents a preset second adjustment parameter for preventing the denominator from being 0, the value range of in an embodiment of the present application, is set to 0.01, the specific numerical value of may also be set by the implementer according to the specific implementation scenario, which is not limited herein.

[0109] It should be noted that in other embodiments of the present application, other basic mathematical operations can also be used to achieve negative correlation mapping, which is not described herein.

[0110] Then, the difference in the absorbance sequence of the sample blocks between the standard clustering cluster and other clustering clusters except the standard clustering cluster is obtained, to obtain the second fermentation unevenness of the target dough.

[0111] Preferably, in an embodiment of the present application, the method for obtaining the second fermentation unevenness of the target dough specifically comprises:

[0112] The average value of the absorbance sequence of all sample blocks in each clustering cluster is taken as the clustering center of each clustering cluster.

[0113] The Euclidean distance between the clustering centers of the standard clustering cluster and each other clustering cluster except the standard clustering cluster is taken as the clustering center distance value between the standard clustering cluster and each other clustering cluster. The greater the clustering center distance value, the greater the difference in the fermentation of the sample blocks between the other clustering cluster and the standard clustering cluster, and the more uneven the fermentation of the target dough.

[0114] Therefore, the average value of the clustering center distance values between the standard clustering cluster and all other clustering clusters can be taken as the second fermentation unevenness of the target dough.

[0115] As an example, in an embodiment of the present application, the expression of the second fermentation unevenness of the target dough can be specifically, for example, as follows:

[0116]

[0117] wherein, represents the second fermentation unevenness of the target dough; represents the cluster center distance value between the standard cluster and the other cluster; represents the number of clusters, and represents the number of clusters other than the standard cluster.

[0118] Meanwhile, the more the number of clusters, the more uneven the fermentation of the target dough, and the greater the degree of abnormality of the fermentation of the target dough, so the first fermentation unevenness, the second fermentation unevenness and the number of all clusters of the target dough are comprehensively processed and normalized, and the calculation result is limited in the range of 0 to 1, so as to obtain the second fermentation abnormality of the target dough.

[0119] As an example, in an embodiment of the present application, the expression of the second fermentation abnormality of the target dough can be specifically, for example:

[0120]

[0121] wherein, represents the second fermentation abnormality of the target dough; represents the first fermentation unevenness of the target dough; represents the second fermentation unevenness of the target dough; represents the number of clusters; represents a normalization function for normalization processing.

[0122] Step S4: combining the first fermentation abnormality and the second fermentation abnormality to evaluate the fermentation quality of the target dough.

[0123] The greater the first fermentation abnormality and the second fermentation abnormality of the target dough, the greater the degree of abnormality in the fermentation process of the target dough, and the more likely the fermentation of the target dough is unqualified, so the first fermentation abnormality and the second fermentation abnormality of the target dough can be combined to accurately evaluate the fermentation quality of the target dough.

[0124] Preferably, in an embodiment of the present application, the method for evaluating the fermentation quality of the target dough specifically comprises:

[0125] The first fermentation abnormality and the second fermentation abnormality of the target dough are comprehensively processed and normalized, and the calculation result is limited in the range of 0 to 1. ​The fermentation abnormality evaluation value of the target dough is obtained in the range, and the greater the fermentation abnormality evaluation value, the greater the degree of abnormal fermentation of the target dough, and the more likely the fermentation of the target dough is unqualified. Therefore, if the fermentation abnormality evaluation value of the target dough is greater than the preset abnormal threshold, the target dough is identified as unqualified fermentation, wherein the preset abnormal threshold is in the range of In an embodiment of the present application, the preset abnormal threshold is set to 0.8, and the preset abnormal threshold can also be set by the implementer according to the specific implementation scene, which is not limited here.

[0126] The fermentation quality of each fermented sour dough in the same production batch can be evaluated by the same method, so as to screen out the dough with unqualified fermentation.

[0127] An embodiment of the present application provides a fermented sour dough detection system for fermented food manufacturing, which comprises a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and the computer program can realize the method described in steps S1-S4 when running in the processor.

[0128] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the 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 possible or can be advantageous.

[0129] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

Claims

1. A method for detecting a fermented sourdough for fermented bakery production, characterized in that, The method comprises: acquiring a spectrum diagram of each fermentation sourdough at each time in the same production batch in the fermentation process, the spectrum diagram being composed of absorbance corresponding to different wavelengths; taking any one of the fermentation sourdoughs as a target sourdough, obtaining a fermentation state inconsistency degree of each wavelength in the spectrum diagram of the target sourdough at each time according to the difference in absorbance of the same wavelength between the spectrum diagram of the target sourdough and the spectrum diagram of each other fermentation sourdough at the same time, screening out a fermentation abnormal time of the target sourdough and an abnormal wavelength in the spectrum diagram of each fermentation abnormal time of the target sourdough based on the fermentation state inconsistency degree of each wavelength in the spectrum diagram of the target sourdough at each time, and obtaining a first fermentation abnormality degree of the target sourdough according to the number of the fermentation abnormal times and the number change of the abnormal wavelengths in the spectrum diagram of each fermentation abnormal time; blocking the target sourdough after the fermentation is completed to obtain a plurality of sample blocks of the target sourdough, acquiring a spectrum diagram of each sample block, clustering all the sample blocks according to the difference in absorbance of the same wavelength between the spectrum diagrams of the sample blocks to obtain a plurality of clustering clusters, and obtaining a second fermentation abnormality degree of the target sourdough according to the number of the sample blocks in each clustering cluster, the number of the clustering clusters, and the difference in absorbance of each wavelength between the spectrum diagrams of the sample blocks in each clustering cluster; combining the first fermentation abnormality degree and the second fermentation abnormality degree to evaluate the fermentation quality of the target sourdough.

2. The method for detecting a fermented sourdough for fermented bakery production according to claim 1, characterized in that, The method comprises: taking the absolute value of the difference in absorbance of the same wavelength between the spectrum diagram of the target sourdough and the spectrum diagram of each other fermentation sourdough at the same time as the difference value of the absorbance of each wavelength between the spectrum diagram of the target sourdough and the spectrum diagram of each other fermentation sourdough at each time; normalizing the cumulative value of the difference values of the absorbance of each wavelength between the spectrum diagram of the target sourdough and the spectrum diagrams of all other fermentation sourdoughs at each time to obtain the fermentation state inconsistency degree of each wavelength in the spectrum diagram of the target sourdough at each time.

3. The method for detecting a fermented sourdough for fermented bakery product manufacturing according to claim 1, characterized in that, The method comprises: taking the average value of the fermentation state inconsistency degrees of all the wavelengths in the spectrum diagram of the target sourdough at each time as a fermentation state deviation degree of the target sourdough at each time; taking the time when the fermentation state deviation degree is greater than a preset deviation threshold as the fermentation abnormal time of the target sourdough; taking the wavelength whose fermentation state inconsistency degree is greater than the preset deviation threshold in the spectrum diagram of each fermentation abnormal time of the target sourdough as the abnormal wavelength in the spectrum diagram of each fermentation abnormal time of the target sourdough.

4. The method for detecting a fermented sourdough for fermented bakery production according to claim 3, characterized in that, The method comprises: performing a negative correlation mapping on the number of the first fermentation abnormal time of the target sourdough to obtain a fermentation abnormality occurrence evaluation value of the target sourdough; taking the number of all the fermentation abnormal times of the target sourdough as the numerator and the number of all the times in the fermentation process as the denominator, and taking the ratio as the proportion of the fermentation abnormal times of the target sourdough. comprehensive and normalized, to obtain a fermentation abnormality severity of the target dough. According to time sequence, the number of abnormal wavelengths in the spectral graph of the target dough at each fermentation abnormal time is linearly fitted, and the slope of the fitted straight line is normalized to obtain an abnormal component quantity change trend value of the target dough. The set of all abnormal wavelengths in the spectral graph of the target dough at each fermentation abnormal time is taken as the abnormal wavelength set of the target dough at each fermentation abnormal time, and according to the number of the same abnormal wavelengths in the abnormal wavelength sets of the adjacent two fermentation abnormal times of the target dough and the abnormal component quantity change trend value of the target dough, an abnormal component change evaluation value of the target dough is obtained. The fermentation abnormality severity of the target dough and the abnormal component change evaluation value are comprehensively processed and normalized to obtain a first fermentation abnormality degree of the target dough.

5. The method for detecting a fermented sourdough for fermented bakery production according to claim 4, characterized in that, The abnormal component change evaluation value of the target dough includes: The Jaccard similarity coefficient between the abnormal wavelength sets of the adjacent two fermentation abnormal times of the target dough is negatively correlated and normalized to obtain the abnormal component change degree of the target dough at the adjacent two fermentation abnormal times. The average value of the abnormal component change degree of the target dough at all adjacent two fermentation abnormal times is taken as the overall change degree of the abnormal components of the target dough. The overall change degree of the abnormal components of the target dough and the abnormal component quantity change trend value are comprehensively processed and normalized to obtain the abnormal component change evaluation value of the target dough.

6. The method for detecting a fermented sourdough for fermented bakery production according to claim 1, characterized in that, The multiple clustering clusters include: According to the order from small to large, the sequence of the absorbance of all wavelengths in the spectral graph of each sample block is taken as the absorbance sequence of each sample block. The Euclidean distance of the absorbance sequence between any two sample blocks is taken as the distance measure between any two sample blocks. Based on the distance measure between any two sample blocks, all sample blocks are clustered to obtain multiple clustering clusters.

7. The method for detecting a fermented sourdough for fermented bakery production according to claim 6, characterized in that, The second fermentation abnormality degree of the target dough includes: The clustering cluster containing the largest number of sample blocks is taken as the standard clustering cluster of the target dough. The number of sample blocks in the standard clustering cluster is taken as the numerator, the number of all sample blocks of the target dough is taken as the denominator, and the ratio is taken as the sample block number proportion of the standard clustering cluster of the target dough. The average value of the distance measure between any two sample blocks in the standard clustering cluster is negatively correlated and mapped to obtain the aggregation degree of the standard clustering cluster of the target dough. The sample block number proportion and the aggregation degree are comprehensively processed and negatively correlated to obtain a first fermentation non-uniformity of the target dough. The difference between the absorbance sequence of the sample blocks between the standard clustering cluster and other clustering clusters except the standard clustering cluster is obtained to obtain a second fermentation non-uniformity of the target dough. The first fermentation unevenness, the second fermentation unevenness and the number of all clustering clusters of the target dough are integrated and normalized to obtain a second fermentation abnormality of the target dough.

8. The method for detecting a fermented sourdough for fermented bakery production according to claim 7, characterized in that, The obtaining of the second fermentation unevenness of the target dough comprises: taking the average value of the absorbance sequence of all sample blocks in each clustering cluster as a clustering center of each clustering cluster; taking the Euclidean distance between the clustering centers of the standard clustering cluster and each other clustering cluster except the standard clustering cluster as a clustering center distance value between the standard clustering cluster and each other clustering cluster; taking the average value of the clustering center distance values between the standard clustering cluster and all other clustering clusters as the second fermentation unevenness of the target dough.

9. The method for detecting a fermented sourdough for fermented bakery production according to claim 1, characterized in that, The evaluation of the fermentation quality of the target dough comprises: integrating and normalizing the first fermentation abnormality and the second fermentation abnormality of the target dough to obtain a fermentation abnormality evaluation value of the target dough; if the fermentation abnormality evaluation value of the target dough is greater than a preset abnormality threshold, the target dough is identified as unqualified in fermentation.

10. A fermented sourdough detection system for fermented bakery production, the system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the steps of the method of any one of claims 1-9 when executing the computer program.

Images