CN113127552B

CN113127552B - Food safety identification method and system based on big data

Info

Publication number: CN113127552B
Application number: CN202110452251.3A
Authority: CN
Inventors: 楼超群; 周骏贵; 胡飞杰; 杨婷婷; 顾慧丹; 申蓉; 郭楠歆
Original assignee: NANJING INSTITUTE OF PRODUCT QUALITY INSPECTION
Current assignee: NANJING INSTITUTE OF PRODUCT QUALITY INSPECTION
Priority date: 2021-04-26
Filing date: 2021-04-26
Publication date: 2021-10-29
Anticipated expiration: 2041-04-26
Also published as: CN113127552A

Abstract

According to the food safety identification method and system based on big data, the correlation coefficient between each piece of production environment monitoring data and the target food production line environment state can be obtained, and the corresponding production environment monitoring data queue is obtained by combining the food safety index characteristics of each piece of production environment monitoring data, so that the target food safety identification information label queue aiming at the target food production line environment state can be generated based on the production environment monitoring data queue. The production environment monitoring data queue is obtained through sequencing and sorting, the importance degrees of the target food safety identification information tag contents in the target food safety identification information tag queue can be correspondingly arranged in a descending order, the target food safety identification information tag contents in the target food safety identification information tag queue at the front of the sequence can be accurately positioned, and the food safety identification can meet the actual requirement through the target food safety identification information tag contents in the front of the sequence.

Description

Food safety identification method and system based on big data

Technical Field

The application relates to the technical field of big data food safety identification, in particular to a food safety identification method and system based on big data.

Background

Food safety means that food is non-toxic and harmless, meets the existing nutritional requirements, and does not cause any acute, subacute or chronic harm to human health. Along with the improvement of living standard of people, people pay more and more high attention to food safety, and pay more attention to the environmental safety in the food production and processing process.

At present, with the continuous development of big data technology, food production monitoring based on big data is gradually put into use, and monitoring and control on the food production and processing process can be realized to ensure food safety. However, the related big data food safety identification technology still has some places to be optimized.

Disclosure of Invention

In view of the foregoing, the present application provides the following.

The scheme of one embodiment of the application provides a food safety identification method based on big data, which is applied to a food safety identification system, and the method at least comprises the following steps:

acquiring a production environment monitoring data set aiming at the environmental state of a target food production line, wherein the production environment monitoring data set comprises at least two pieces of production environment monitoring data;

obtaining a correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the target food production line environment state;

sorting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristics of the production environment monitoring data to obtain a corresponding production environment monitoring data queue;

and generating a target food safety identification information tag queue aiming at the environmental state of the target food production line based on the production environment monitoring data queue, wherein the target food safety identification information tag queue comprises at least two target food safety identification information tag contents.

Alternatively, the sorting of the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristic of the production environment monitoring data to obtain the corresponding production environment monitoring data queue specifically includes:

splitting each piece of production environment monitoring data according to the correlation coefficient corresponding to each piece of production environment monitoring data and the food safety index characteristics of each piece of production environment monitoring data to obtain at least two production environment monitoring data subsets;

and sorting the production environment monitoring data subsets, and sorting the production environment monitoring data in the production environment monitoring data subsets to obtain the production environment monitoring data queue.

Alternatively, the splitting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristic of the production environment monitoring data to obtain at least two production environment monitoring data subsets specifically includes:

globally processing the food safety index characteristics of the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data to obtain the key food safety index characteristics of the production environment monitoring data;

and carrying out multi-dimensional characteristic identification processing on each piece of production environment monitoring data according to the key food safety index characteristics of each piece of production environment monitoring data to obtain at least two production environment monitoring data subsets.

Alternatively, the sorting the production environment monitoring data subsets, and sorting the production environment monitoring data in the production environment monitoring data subsets to obtain the production environment monitoring data queue specifically includes:

sorting the production environment monitoring data subsets according to the quantity of the production environment monitoring data contained in the production environment monitoring data subsets;

and, for each production environment monitoring data subset, performing the following operations:

sorting the production environment monitoring data in the production environment monitoring data subset according to the correlation detection result of the food safety index characteristic of the production environment monitoring data in the production environment monitoring data subset and the production environment monitoring data subset;

and generating the production environment monitoring data queue based on the arrangement result among the production environment monitoring data subsets and the arrangement result of the production environment monitoring data in the production environment monitoring data subsets.

Alternatively to this, the first and second parts may be,

the obtaining of the correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the target food production line environment state specifically includes:

respectively inputting the production environment monitoring data into a trained information tag identification network, extracting key description features of the production environment monitoring data based on a key description extraction network layer of data block dimensions in the trained information tag identification network, and obtaining correlation coefficients corresponding to the production environment monitoring data output by the key description extraction network layer;

the method comprises the following steps of sorting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristics of the production environment monitoring data to obtain a corresponding production environment monitoring data queue, and specifically comprises the following steps:

respectively inputting the production environment monitoring data and the correlation coefficient corresponding to the production environment monitoring data into a feature processing cascade network layer in the trained information label identification network, carrying out multi-dimensional feature identification processing and sorting on the production environment monitoring data based on the feature processing cascade network layer, and obtaining first local description contents of segment dimensions output by the feature processing cascade network layer, wherein the production environment monitoring data segments in the first local description contents are combined to form the production environment monitoring data queue;

the generating of the target food safety identification information tag queue for the target food production line environment state based on the production environment monitoring data queue specifically includes:

inputting the local description content into a description feature extraction layer in the trained information label identification network, and extracting dynamic key description features based on the description feature extraction layer to obtain the target food safety identification information label queue output by the description feature extraction layer; the trained information label recognition network is obtained by training according to a sample training set, sample data in the sample training set comprises sample production environment monitoring data with calibrated correlation coefficient identifications, and the correlation coefficient identifications show whether the sample production environment monitoring data are related to the environmental state of a sample food production line.

Alternatively, the respectively inputting the production environment monitoring data into the trained information tag identification network, performing key description feature extraction on the production environment monitoring data based on the key description extraction network layer of the data block dimension in the trained information tag identification network, and obtaining the correlation coefficient corresponding to the production environment monitoring data output by the key description extraction network layer specifically includes:

respectively inputting the production environment monitoring data into the key description extraction network layer, and mapping the production environment monitoring data to a preset vector space based on a data block transformation network layer in the key description extraction network layer to obtain the description content of the production environment monitoring data;

respectively converting the description contents of the monitoring data of each production environment into corresponding food safety index vectors through a global feature processing strategy;

based on the key description extraction network layer, respectively extracting dynamic characteristics between the food safety index vector of each piece of production environment monitoring data and the food safety index vectors of other pieces of production environment monitoring data except the piece of production environment monitoring data;

and obtaining a correlation coefficient between each piece of production environment monitoring data and the target food production line environment state based on the dynamic characteristics corresponding to each piece of production environment monitoring data.

Alternatively, the performing, on the basis of the feature processing cascade network layer, multi-dimensional feature identification processing and sorting on the monitoring data of each production environment to obtain the first partial description content of the segment dimension output by the feature processing cascade network layer specifically includes:

based on the trained information label identification network, performing feature processing cascade network layer, mapping each piece of production environment monitoring data to a preset vector space to obtain a fragment vector queue corresponding to each piece of production environment monitoring data;

extracting the description content of the segment vector queue corresponding to each piece of production environment monitoring data through continuous pooling treatment to obtain the food safety index description content of each piece of production environment monitoring data;

globally processing the food safety index description content of each production environment monitoring data according to the correlation coefficient corresponding to each production environment monitoring data to obtain the key food safety index description content of each production environment monitoring data;

performing multi-dimensional characteristic identification processing on the key food safety index description content based on each piece of production environment monitoring data to obtain at least two production environment monitoring data subsets;

and sorting all the production environment monitoring data subsets, sorting all the production environment monitoring data in each production environment monitoring data subset, splicing the key food safety index description contents of all the production environment monitoring data, and performing segment dimension conversion to obtain the first local description content.

Alternatively, the inputting the local description content into a description feature extraction layer in the trained information tag identification network, and performing dynamic key description feature extraction based on the description feature extraction layer to obtain the target food safety identification information tag queue output by the description feature extraction layer specifically includes:

sequentially generating each food safety identification information label content in the target food safety identification information label queue by adopting a set processing mode, wherein one food safety identification information label in the target food safety identification information queue at least comprises one food safety identification information label content;

wherein, in one setting process, the following operations are executed:

inputting the label content of the target food safety identification information output in the previous round into the description feature extraction layer, wherein the input in the first round into the description feature extraction layer is a preset initial reference segment;

analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data fragment in the sample queue through a dynamic identification strategy, wherein the matching probability represents the matching degree between the production environment monitoring data fragment and the food safety identification information label content output in the previous round;

globally processing the matching probability and the food safety index vector queue of the production environment monitoring data fragments in the production environment monitoring data queue, and inputting the processed data into a long-term and short-term neural network to obtain the target food safety index description content of the production environment monitoring data queue output in the current round;

and generating the target food safety identification information label content output in the current round based on the target food safety identification information label content output in the previous round and the target food safety index description content.

Alternatively, before analyzing the matching probability of the target food safety identification information tag content output in the previous round and each production environment monitoring data segment in the sample queue through the dynamic identification strategy, the method further comprises the following steps:

taking the target production environment monitoring data subsets selected in the current round and the associated production environment monitoring data subsets of the target production environment monitoring data subsets as significant production environment monitoring data subsets, and taking other production environment monitoring data subsets as potential production environment monitoring data subsets, wherein the target production environment monitoring data subsets selected each time are determined based on the sequence among the production environment monitoring data subsets;

adding a first matching feature to a production environment monitoring data segment in a significant production environment monitoring data subset in the production environment monitoring data queue, and adding a second matching feature to a production environment monitoring data segment in a potential production environment monitoring data subset in the production environment monitoring data queue to obtain a first matching food safety index vector corresponding to each production environment monitoring data segment in the sample queue;

adding the first matching features to the target food safety identification information label content output in the previous round to obtain a corresponding second matching food safety index vector;

the analyzing, by a dynamic identification strategy, matching probabilities of the contents of the target food safety identification information tag output in the previous round and the monitoring data segments of each production environment in the sample queue includes:

and analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data fragment in the sample queue based on a dynamic identification strategy by combining the first matching vector corresponding to each production environment monitoring data fragment in the sample queue and the second matching food safety index vector corresponding to the target food safety identification information label content output in the previous round.

The scheme of one embodiment of the application provides a food safety identification system, which comprises a processing engine, a network module and a memory; the processing engine and the memory communicate through the network module, and the processing engine reads the computer program from the memory and operates to perform the above-described method.

In the description that follows, additional features will be set forth, in part, in the description. These features will be in part apparent to those skilled in the art upon examination of the following and the accompanying drawings, or may be learned by production or use. The features of the present application may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations particularly pointed out in the detailed examples that follow.

Drawings

The present application will be further explained by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not intended to be limiting, and in these embodiments like numerals are used to indicate like structures, wherein:

FIG. 1 is a flow diagram of an exemplary big data based food safety identification method and/or process, according to some embodiments of the present application;

FIG. 2 is a block diagram of an exemplary big data based food safety identification device according to some embodiments of the present application;

fig. 3 is a schematic diagram of the hardware and software components of an exemplary food safety identification system according to some embodiments of the present application.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only examples or embodiments of the application, from which the application can also be applied to other similar scenarios without inventive effort for a person skilled in the art. Unless otherwise apparent from the context, or otherwise indicated, like reference numbers in the figures refer to the same structure or operation.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, other words may be substituted by other expressions if they accomplish the same purpose.

As used in this application and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.

Flow charts are used herein to illustrate operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, the various steps may be processed in reverse order or simultaneously. Meanwhile, other operations may be added to the processes, or a certain step or several steps of operations may be removed from the processes.

In order to better understand the technical solutions of the present invention, the following detailed descriptions of the technical solutions of the present invention are provided with the accompanying drawings and the specific embodiments, and it should be understood that the specific features in the embodiments and the examples of the present invention are the detailed descriptions of the technical solutions of the present invention, and are not limitations of the technical solutions of the present invention, and the technical features in the embodiments and the examples of the present invention may be combined with each other without conflict.

First, an exemplary big data based food safety identification method is described, referring to fig. 1, which is a flowchart illustrating an exemplary big data based food safety identification method and/or process according to some embodiments of the present application, and the big data based food safety identification method may include the following steps 100-400.

Step 100, the food safety identification system acquires a production environment monitoring data set aiming at the environmental state of a target food production line.

In an embodiment of the present application, the production environment monitoring data set includes at least two pieces of production environment monitoring data.

By way of example, a target food production line environmental state may be understood as an environmental state of a target food production line, such as an environmental state within a plant. Different production environment monitoring data correspond to different monitoring emphasis points. Generally, the production environment monitoring data set of the environmental status of the target food production line can be acquired by the relevant environment sensors, which are not described herein.

200, the food safety identification system obtains a correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the target food production line environment state.

In the embodiment of the present application, the correlation coefficient may be understood as a correlation degree or a degree of association between each piece of production environment monitoring data and the target food production line environment state, and may be, for example, a pearson correlation coefficient or a spearman correlation coefficient.

Step 300, the food safety identification system sorts the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristics of the production environment monitoring data to obtain a corresponding production environment monitoring data queue.

In the embodiment of the application, the food safety index features are used for evaluating food safety from different dimensions, such as the dimensions of mold proportion, food deterioration and the like. The sorting of the production environment monitoring data may be understood as sorting the production environment monitoring data, and accordingly, the production environment monitoring data queue is a sorting queue of the production environment monitoring data.

In some possible embodiments, the sorting of the production environment monitoring data according to the correlation coefficient corresponding to each production environment monitoring data and the food safety index characteristic of each production environment monitoring data described in the above step 300 to obtain the corresponding production environment monitoring data queue may be implemented by the following technical solutions described in the following steps 310 and 320.

And 310, splitting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristics of the production environment monitoring data to obtain at least two production environment monitoring data subsets.

In a related embodiment, the splitting of the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristic of the production environment monitoring data described in the above step 310 to obtain at least two production environment monitoring data subsets may include the following technical solutions described in steps 311 and 312.

And 311, performing global processing on the food safety index characteristics of the production environment monitoring data according to the correlation coefficients corresponding to the production environment monitoring data to obtain the key food safety index characteristics of the production environment monitoring data.

For example, the global processing may be understood as performing weighted processing on the food safety index characteristics of the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data.

And step 312, performing multi-dimensional feature recognition processing on each piece of production environment monitoring data according to the key food safety index features of each piece of production environment monitoring data to obtain at least two production environment monitoring data subsets.

For example, the multidimensional feature identification process may be understood as clustering the production environment monitoring data according to the key food safety index features of the production environment monitoring data.

Therefore, accurate clustering of the production environment monitoring data can be realized, and repeated data cannot occur among production environment monitoring data subsets.

Step 320, sorting the production environment monitoring data subsets, and sorting the production environment monitoring data in the production environment monitoring data subsets to obtain the production environment monitoring data queue.

In some possible embodiments, the sorting of the production environment monitoring data subsets described in the above step S320, and the sorting of the production environment monitoring data in the production environment monitoring data subsets to obtain the production environment monitoring data queue may be implemented by the following technical solutions described in the following steps 321 and 322.

Step 321, sorting the production environment monitoring data subsets according to the quantity of the production environment monitoring data included in the production environment monitoring data subsets.

Step 322, for each production environment monitoring data subset, respectively executing the following operations: sorting the production environment monitoring data in the production environment monitoring data subset according to the correlation detection result of the food safety index characteristic of the production environment monitoring data in the production environment monitoring data subset and the production environment monitoring data subset; and generating the production environment monitoring data queue based on the arrangement result among the production environment monitoring data subsets and the arrangement result of the production environment monitoring data in the production environment monitoring data subsets.

It is understood that the correlation detection result is used to characterize the degree of correlation between the food safety index characteristic of each piece of the production environment monitoring data in the production environment monitoring data subset and the production environment monitoring data subset. Correspondingly, through twice sequencing, the production environment monitoring data in the production environment monitoring data queue can be ensured not to be missed or repeated, and therefore the sequencing reliability of the production environment monitoring data queue is ensured.

By the design, the production environment monitoring data subsets are sorted, and the production environment monitoring data in the production environment monitoring data subsets are sorted respectively, so that the integrity of a production environment monitoring data queue can be ensured.

And 400, generating a target food safety identification information tag queue aiming at the environmental state of the target food production line by the food safety identification system based on the production environment monitoring data queue.

In an embodiment of the present application, the target food safety identification information tag queue includes at least two target food safety identification information tag contents. Different target food safety identification information label contents can refer to different food safety identification results, such as a mildew identification result, a cooking process missing identification result or other identification results, and the embodiments of the application are not listed.

It can be understood that the contents of the target food safety identification information tags in the target food safety identification information tag queue are sorted from high to low according to the importance degree, so that the sorted position of the more important target food safety identification information tag contents in the queue is ensured to be ahead, and the more important target food safety identification information tag contents are conveniently and quickly positioned and analyzed.

In still other embodiments, the above steps may be implemented in conjunction with artificial intelligence techniques. In other words, the above steps can be implemented by the relevant machine learning network model.

Based on this, in some possible embodiments, the obtaining of the correlation coefficient between each piece of the production environment monitoring data in the production environment monitoring data set and the target food production line environment state described in step S200 may include the following: and respectively inputting the production environment monitoring data into the trained information tag identification network, extracting key description features of the production environment monitoring data based on a key description extraction network layer of the data block dimension in the trained information tag identification network, and obtaining a correlation coefficient corresponding to the production environment monitoring data output by the key description extraction network layer.

In a related embodiment, the above-described steps of inputting the production environment monitoring data into the trained information tag identification network, performing key description feature extraction on the production environment monitoring data based on the key description extraction network layer of the data block dimension in the trained information tag identification network, and obtaining a correlation coefficient corresponding to the production environment monitoring data output by the key description extraction network layer may include the following technical solutions described in steps 210 to 240.

Step 210, inputting the production environment monitoring data into the key description extraction network layer, and mapping the production environment monitoring data to a preset vector space based on a data block transformation network layer in the key description extraction network layer to obtain the description content of the production environment monitoring data.

And step 220, respectively converting the description contents of the monitoring data of each production environment into corresponding food safety index vectors through a global feature processing strategy.

And 230, extracting the dynamic characteristics between the food safety index vector of each piece of production environment monitoring data and the food safety index vectors of other pieces of production environment monitoring data except the piece of production environment monitoring data on the basis of the key description extraction network layer.

And 240, obtaining a correlation coefficient between each piece of production environment monitoring data and the target food production line environment state based on the dynamic characteristics corresponding to each piece of production environment monitoring data.

In this way, by implementing the steps S210 to S240, the description content of the production environment monitoring data can be determined by combining the preset vector space, so that the correlation coefficient between each piece of production environment monitoring data and the environmental state of the target food production line can be accurately calculated by the global feature processing strategy and the key description extraction network layer.

Further, the sorting the production environment monitoring data according to the correlation coefficient corresponding to each production environment monitoring data and the food safety index characteristic of each production environment monitoring data described in step 300 to obtain the corresponding production environment monitoring data queue may include the following contents: and respectively inputting the production environment monitoring data and the correlation coefficient corresponding to the production environment monitoring data into a feature processing cascade network layer in the trained information label identification network, carrying out multi-dimensional feature identification processing and sorting on the production environment monitoring data based on the feature processing cascade network layer, and obtaining first partial description contents of segment dimensions output by the feature processing cascade network layer, wherein the production environment monitoring data segments in the first partial description contents are combined to form the production environment monitoring data queue.

In related embodiments, the processing and sorting of the monitoring data of each production environment based on the feature processing cascade network layer in the foregoing steps to obtain the first partial description content of the segment dimension output by the feature processing cascade network layer may specifically include the technical solutions described in the following steps (1) to (5).

(1) And based on the trained information label identification network, processing a cascade network layer, and mapping each piece of production environment monitoring data to a preset vector space to obtain a fragment vector queue corresponding to each piece of production environment monitoring data.

(2) And extracting the description content of the segment vector queue corresponding to each piece of production environment monitoring data through continuous pooling treatment to obtain the food safety index description content of each piece of production environment monitoring data.

(3) And globally processing the food safety index description content of each production environment monitoring data according to the correlation coefficient corresponding to each production environment monitoring data to obtain the key food safety index description content of each production environment monitoring data.

(4) And carrying out multi-dimensional characteristic identification processing on the key food safety index description content based on each piece of production environment monitoring data to obtain at least two production environment monitoring data subsets.

(5) And sorting all the production environment monitoring data subsets, sorting all the production environment monitoring data in each production environment monitoring data subset, splicing the key food safety index description contents of all the production environment monitoring data, and performing segment dimension conversion to obtain the first local description content.

In this way, by implementing the steps (1) to (5), the data fragment analysis can be performed through the feature processing cascade network (the feature recognition network and the sorting network), and the key food safety index description content of each piece of production environment monitoring data can be obtained through global processing, so that the production environment monitoring data subset can be accurately obtained, and the complete splicing and fragment dimension conversion of the key food safety index description content can be performed after the data set sorting is performed, so as to ensure the integrity and the availability of the first local description content.

Further, the generating of the target food safety identification information tag queue for the environmental status of the target food production line based on the production environment monitoring data queue described in step 400 specifically includes: inputting the local description content into a description feature extraction layer in the trained information label identification network, and extracting dynamic key description features based on the description feature extraction layer to obtain the target food safety identification information label queue output by the description feature extraction layer.

In a related embodiment, the inputting the local description content into the description feature extraction layer in the trained information tag identification network described in the above step, and performing dynamic key description feature extraction based on the description feature extraction layer to obtain the target food safety identification information tag queue output by the description feature extraction layer may specifically include the following: and sequentially generating the contents of each food safety identification information tag in the target food safety identification information tag queue by adopting a set processing mode, wherein one food safety identification information tag in the target food safety identification information queue at least comprises one food safety identification information tag content.

In some embodiments, the setting process may be understood as a loop iteration process, and accordingly, during one setting process, the following operations may be performed: inputting the label content of the target food safety identification information output in the previous round into the description feature extraction layer, wherein the input in the first round into the description feature extraction layer is a preset initial reference segment; analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data fragment in the sample queue through a dynamic identification strategy, wherein the matching probability represents the matching degree between the production environment monitoring data fragment and the food safety identification information label content output in the previous round; globally processing the matching probability and the food safety index vector queue of the production environment monitoring data fragments in the production environment monitoring data queue, and inputting the processed data into a long-term and short-term neural network to obtain the target food safety index description content of the production environment monitoring data queue output in the current round; and generating the target food safety identification information label content output in the current round based on the target food safety identification information label content output in the previous round and the target food safety index description content.

By the design, the relevance and the matching among the contents of the food safety identification information labels can be ensured.

In some possible embodiments, before the analyzing, by a dynamic identification strategy, matching probabilities of the target food safety identification information tag content output from the previous round and each production environment monitoring data segment in the sample queue, the method further includes: taking the target production environment monitoring data subsets selected in the current round and the associated production environment monitoring data subsets of the target production environment monitoring data subsets as significant production environment monitoring data subsets, and taking other production environment monitoring data subsets as potential production environment monitoring data subsets, wherein the target production environment monitoring data subsets selected each time are determined based on the sequence among the production environment monitoring data subsets; adding a first matching feature to a production environment monitoring data segment in a significant production environment monitoring data subset in the production environment monitoring data queue, and adding a second matching feature to a production environment monitoring data segment in a potential production environment monitoring data subset in the production environment monitoring data queue to obtain a first matching food safety index vector corresponding to each production environment monitoring data segment in the sample queue; and adding the first matching features to the target food safety identification information label content output in the previous round to obtain a corresponding second matching food safety index vector.

On the basis of the above, analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data segment in the sample queue by a dynamic identification strategy, specifically comprising: and analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data fragment in the sample queue based on a dynamic identification strategy by combining the first matching vector corresponding to each production environment monitoring data fragment in the sample queue and the second matching food safety index vector corresponding to the target food safety identification information label content output in the previous round.

It can be understood that, through the above, the matching probability between the target food safety identification information tag content output in the previous round and each production environment monitoring data segment in the sample queue can be analyzed based on a dynamic identification strategy, so that the accuracy of the matching probability of each production environment monitoring data segment can be ensured in a global aspect, and the error of the matching probability of each production environment monitoring data segment can be reduced as much as possible.

In the above embodiment, the trained information tag identification network is obtained by training according to a sample training set, sample data in the sample training set includes sample production environment monitoring data with a calibrated correlation coefficient identifier, and the correlation coefficient identifier indicates whether the sample production environment monitoring data is related to the environmental state of the sample food production line.

In some alternative embodiments, the information label recognition network is trained as follows.

Firstly, acquiring the sample training set aiming at least one sample food production line environment state; according to the sample data in the sample training set, performing setting processing training on the untrained information tag identification network to obtain the trained information tag identification network; wherein, each setting process training process comprises the following operations: selecting a group of sample data aiming at the environmental state of the same sample food production line from the sample training set, respectively inputting sample production environment monitoring data contained in each selected sample data into a key description extraction network layer of the dimensionality of a data block in the untrained information tag identification network, and obtaining a correlation coefficient corresponding to each piece of sample production environment monitoring data output by the key description extraction network layer; and constructing a first model evaluation function based on the difference between the correlation coefficient corresponding to the sample production environment monitoring data and the correlation coefficient before identification.

Secondly, respectively inputting the sample production environment monitoring data in each selected sample data and the correlation coefficient corresponding to the sample production environment monitoring data into a feature processing cascade network layer in the untrained information tag identification network, and performing multi-dimensional feature identification processing on the sample production environment monitoring data based on the feature processing cascade network layer to obtain at least two production environment monitoring data subsets; sorting each production environment monitoring data subset based on the characteristic processing cascade network layer to obtain second local description content of segment dimensionality output by the characteristic processing cascade network layer; inputting the second local description content into a description feature extraction layer in the untrained information tag identification network, and performing dynamic key description feature extraction based on the description feature extraction layer to obtain a group of test food safety identification information tag queues output by the description feature extraction layer, wherein the test food safety identification information tag queues comprise at least two test food safety identification information tag contents; constructing a second model evaluation function based on the distribution difference of the content of the test food safety identification information tag in the test food safety identification information tag queue and the content of the actual food safety identification information tag in the actual food safety identification information tag queue;

and then, constructing a third model evaluation function based on the matching degree of the production environment monitoring data segments in each production environment monitoring data subset.

And finally, performing model parameter adjustment on the untrained information label identification network according to the first model evaluation function, the second model evaluation function and the third model evaluation function.

It can be understood that the first model evaluation function, the second model evaluation function and the third model evaluation function respectively correspond to different types of loss functions, and thus, by analyzing the first model evaluation function, the second model evaluation function and the third model evaluation function, accurate parameter adjustment of the information tag identification network can be realized, so that the model stability and the prediction accuracy of the information tag identification network are ensured, and the model performance of the information tag identification network is improved.

In some optional embodiments, the constructing a second model evaluation function based on the distribution difference between the content of the test food safety identification information tag in the test food safety identification information tag queue and the content of the actual food safety identification information tag in the actual food safety identification information tag queue, which is described in the above step, specifically includes: for any one test food safety identification information label content, determining the distribution difference between the test food safety identification information label content in the test food safety identification information label queue and the actual food safety identification information label content in the actual food safety identification information label queue based on the distribution possibility of the test food safety identification information label content in a preset food safety identification information label content set and the distribution possibility of the test food safety identification information label content in the production environment monitoring data set; constructing the second model merit function based on the determined distribution diversity. In this way, the second model merit function may be accurately determined based on the distribution variability.

In conclusion, when the technical scheme is implemented, the correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the target food production line environment state can be obtained, and the corresponding production environment monitoring data queue is obtained by combining the food safety index characteristics of each piece of production environment monitoring data, so that the target food safety identification information tag queue for the target food production line environment state can be generated based on the production environment monitoring data queue. The production environment monitoring data queue is obtained through sequencing and sorting, the importance degrees of the target food safety identification information tag contents in the target food safety identification information tag queue can be correspondingly arranged in a descending order, the target food safety identification information tag contents in the target food safety identification information tag queue at the front of the sequence can be accurately positioned, and the food safety identification can meet the actual requirement through the target food safety identification information tag contents in the front of the sequence. This allows further optimization of the relevant big data food safety identification technology.

It is to be understood that the above description of some alternative embodiments is to be construed as exemplary and not as an essential feature of the implementation of the present solution.

Next, in view of the above-mentioned food safety identification method based on big data, an exemplary food safety identification device based on big data is further provided in the embodiment of the present invention, as shown in fig. 2, the food safety identification device based on big data 200 may include the following functional modules.

A data obtaining module 210, configured to obtain a production environment monitoring data set for a target food production line environment state, where the production environment monitoring data set includes at least two pieces of production environment monitoring data.

A correlation determination module 220, configured to obtain a correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the target food production line environment state.

And the data sorting module 230 is configured to sort the production environment monitoring data according to the correlation coefficient corresponding to each production environment monitoring data and the food safety index characteristic of each production environment monitoring data, so as to obtain a corresponding production environment monitoring data queue.

A tag generating module 240, configured to generate a target food safety identification information tag queue for the target food production line environmental status based on the production environment monitoring data queue, where the target food safety identification information tag queue includes at least two target food safety identification information tag contents.

Further, referring to fig. 3 in combination, the food safety recognition system 10 may include a processing engine 110, a network module 120, and a memory 130, wherein the processing engine 110 and the memory 130 communicate through the network module 120.

Processing engine 110 may process the relevant information and/or data to perform one or more of the functions described herein. For example, in some embodiments, processing engine 110 may include at least one processing engine (e.g., a single core processing engine or a multi-core processor). By way of example only, the Processing engine 110 may include a Central Processing Unit (CPU), an Application-Specific Integrated Circuit (ASIC), an Application-Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

Network module 120 may facilitate the exchange of information and/or data. In some embodiments, the network module 120 may be any type of wired or wireless network or combination thereof. Merely by way of example, the Network module 120 may include a cable Network, a wired Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a Wireless personal Area Network, a Near Field Communication (NFC) Network, and the like, or any combination thereof. In some embodiments, the network module 120 may include at least one network access point. For example, the network module 120 may include wired or wireless network access points, such as base stations and/or network access points.

The Memory 130 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 130 is used for storing a program, and the processing engine 110 executes the program after receiving the execution instruction.

It will be appreciated that the configuration shown in fig. 3 is merely illustrative and that the food safety identification system 10 may include more or fewer components than shown in fig. 3 or may have a different configuration than shown in fig. 3. The components shown in fig. 3 may be implemented in hardware, software, or a combination thereof.

It should be understood that, for the above, a person skilled in the art can deduce from the above disclosure to determine the meaning of the related technical term without doubt, for example, for some values, coefficients, weights, indexes, factors, and other terms, a person skilled in the art can deduce and determine from the logical relationship between the above and the following, and the value range of these values can be selected according to the actual situation, for example, 0 to 1, for example, 1 to 10, and for example, 50 to 100, which are not limited herein.

The skilled person can unambiguously determine some preset, reference, predetermined, set and target technical features/terms, such as threshold values, threshold intervals, threshold ranges, etc., from the above disclosure. For some technical characteristic terms which are not explained, the technical solution can be clearly and completely implemented by those skilled in the art by reasonably and unambiguously deriving the technical solution based on the logical relations in the previous and following paragraphs. Prefixes of unexplained technical feature terms, such as "first", "second", "previous", "next", "current", "history", "latest", "best", "target", "specified", and "real-time", etc., can be unambiguously derived and determined from the context. Suffixes of technical feature terms not to be explained, such as "list", "feature", "sequence", "set", "matrix", "unit", "element", "track", and "list", etc., can also be derived and determined unambiguously from the foregoing and the following.

The foregoing disclosure of embodiments of the present invention will be apparent to those skilled in the art. It should be understood that the process of deriving and analyzing technical terms, which are not explained, by those skilled in the art based on the above disclosure is based on the contents described in the present application, and thus the above contents are not an inventive judgment of the overall scheme.

It should be appreciated that the system and its modules shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory for execution by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided, for example, on a carrier medium such as a diskette, CD-or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application may be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, for example, or by a combination of the above hardware circuits and software (e.g., firmware).

It is to be noted that different embodiments may produce different advantages, and in different embodiments, any one or combination of the above advantages may be produced, or any other advantages may be obtained.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered merely illustrative and not restrictive of the broad application. Various modifications, improvements and adaptations to the present application may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the present application is included in at least one embodiment of the present application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the present application may be combined as appropriate.

Moreover, those skilled in the art will appreciate that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereon. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, a conventional programming language such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, a dynamic programming language such as Python, Ruby, and Groovy, or other programming languages, and the like. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Numerals describing the number of components, attributes, etc. are used in some embodiments, it being understood that such numerals used in the description of the embodiments are modified in some instances by the use of the modifier "about", "approximately" or "substantially". Unless otherwise indicated, "about", "approximately" or "substantially" indicates that the numbers allow for adaptive variation. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that may vary depending upon the desired properties of the individual embodiments. In some embodiments, the numerical parameter should take into account the specified significant digits and employ a general digit preserving approach. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the range are approximations, in the specific examples, such numerical values are set forth as precisely as possible within the scope of the application.

The entire contents of each patent, patent application publication, and other material cited in this application, such as articles, books, specifications, publications, documents, and the like, are hereby incorporated by reference into this application. Except where the application is filed in a manner inconsistent or contrary to the present disclosure, and except where the claim is filed in its broadest scope (whether present or later appended to the application) as well. It is noted that the descriptions, definitions and/or use of terms in this application shall control if they are inconsistent or contrary to the statements and/or uses of the present application in the material attached to this application.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present application. Other variations are also possible within the scope of the present application. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the present application can be viewed as being consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. A food safety identification method based on big data is applied to a food safety identification system, and the method at least comprises the following steps:

generating a target food safety identification information tag queue aiming at the environmental state of the target food production line based on the production environment monitoring data queue, wherein the target food safety identification information tag queue comprises at least two target food safety identification information tag contents;

wherein, the obtaining of the correlation coefficient between each piece of production environment monitoring data in the production environment monitoring data set and the environmental state of the target food production line specifically includes: respectively inputting the production environment monitoring data into a trained information tag identification network, extracting key description features of the production environment monitoring data based on a key description extraction network layer of data block dimensions in the trained information tag identification network, and obtaining correlation coefficients corresponding to the production environment monitoring data output by the key description extraction network layer;

the method comprises the following steps of sorting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristics of the production environment monitoring data to obtain a corresponding production environment monitoring data queue, and specifically comprises the following steps: respectively inputting the production environment monitoring data and the correlation coefficient corresponding to the production environment monitoring data into a feature processing cascade network layer in the trained information label identification network, carrying out multi-dimensional feature identification processing and sorting on the production environment monitoring data based on the feature processing cascade network layer, and obtaining first local description contents of segment dimensions output by the feature processing cascade network layer, wherein the production environment monitoring data segments in the first local description contents are combined to form the production environment monitoring data queue;

the generating of the target food safety identification information tag queue for the target food production line environment state based on the production environment monitoring data queue specifically includes: inputting the local description content into a description feature extraction layer in the trained information label identification network, and extracting dynamic key description features based on the description feature extraction layer to obtain the target food safety identification information label queue output by the description feature extraction layer; the trained information label recognition network is obtained by training according to a sample training set, sample data in the sample training set comprises sample production environment monitoring data with calibrated correlation coefficient identifications, and the correlation coefficient identifications show whether the sample production environment monitoring data are related to the environmental state of a sample food production line.

2. The method according to claim 1, wherein the sorting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristic of the production environment monitoring data to obtain the corresponding production environment monitoring data queue specifically comprises:

3. The method of claim 2, wherein the splitting the production environment monitoring data according to the correlation coefficient corresponding to the production environment monitoring data and the food safety index characteristic of the production environment monitoring data to obtain at least two production environment monitoring data subsets comprises:

4. The method of claim 2, wherein the sorting the subsets of the production environment monitoring data and the sorting the production environment monitoring data in the subsets of the production environment monitoring data to obtain the production environment monitoring data queue comprises:

5. The method of claim 1, wherein the respectively inputting the pieces of production environment monitoring data into a trained information tag recognition network, and performing key description feature extraction on the pieces of production environment monitoring data based on a key description extraction network layer of data block dimensions in the trained information tag recognition network to obtain correlation coefficients corresponding to the pieces of production environment monitoring data output by the key description extraction network layer specifically include:

6. The method according to claim 1, wherein the performing multidimensional feature identification processing and sorting on the pieces of production environment monitoring data based on the feature processing cascade network layer to obtain the first partial description content of the segment dimension output by the feature processing cascade network layer specifically includes:

7. The method according to claim 1, wherein the inputting the local description content into a descriptive feature extraction layer in the trained information tag identification network, performing dynamic key descriptive feature extraction based on the descriptive feature extraction layer, and obtaining the target food safety identification information tag queue output by the descriptive feature extraction layer specifically comprises:

wherein, in one setting process, the following operations are executed:

analyzing the matching probability of the target food safety identification information label content output in the previous round and each production environment monitoring data fragment in a sample queue through a dynamic identification strategy, wherein the matching probability represents the matching degree between the production environment monitoring data fragment and the food safety identification information label content output in the previous round;

8. The method of claim 7, wherein before analyzing the matching probability of the target food safety identification information tag content output from the previous round and each production environment monitoring data segment in the sample queue by the dynamic identification strategy, the method further comprises:

9. A food safety identification system is characterized by comprising a processing engine, a network module and a memory; the processing engine and the memory communicate through the network module, the processing engine reading a computer program from the memory and operating to perform the method of any of claims 1-8.