A food pesticide residue safety detection method, system and medium

Abstract

This invention relates to a method, system, and medium for detecting pesticide residues in food, belonging to the field of pesticide residue detection technology. The invention acquires crop category information of the current storage area and performs a search based on the crop category to obtain search results. It then matches the identification results with the search results to obtain matching results, and determines the abnormal location nodes of the current food based on the matching results, displaying them on a food detection terminal according to a preset method. This method can quickly detect crops with high pesticide residues in storage, thereby identifying the location nodes of crops with severe pesticide residues. This allows for the notification of relevant personnel to rectify the situation. This method can quickly detect pesticide residue locations, replacing manual methods with higher efficiency, and improves the safety of exported food, thus ensuring consumer safety to a certain extent.

Description

Technical Field

[0001] The present invention relates to the technical field of food pesticide residue detection, and in particular to a food pesticide residue safety detection method, system and medium. Background Art

[0002] my country is an agricultural country, and the safety of agricultural products has always been the focus and hotspot of public concern. Many agricultural product growers have a vague understanding of the boundaries of pesticide use and do not pay attention to the problem of pesticide residues. On the one hand, due to the lack of strict guidance of relevant professional and technical personnel in the use of pesticides, they are not clear about the basic characteristics and control functions of pesticides, and blindly mix and use them, resulting in reduced pesticide efficacy, and some pests quickly develop resistance. However, most agricultural product growers often cannot clearly define the control targets, cannot correctly select the pesticides, and cannot prescribe the right medicine. There are even a small number of growers who use pesticides at will regardless of the weather or the length of time, which not only leads to poor control effects, but sometimes even drug hazards or personnel poisoning. On the other hand, some agricultural product growers have low quality. Although they are aware of the harm caused by excessive pesticide residues, pesticides are a double-edged sword. While preventing and controlling pests and diseases, pesticide residues in agricultural products also have a direct impact on people's health. Therefore, people are very concerned about the pesticide residues left over from the cultivation of agricultural products, and the supply of agricultural products is gradually developing towards green ecology and pollution-free. The main reason for pesticide residues is that growers of agricultural products have weak safety awareness and do not have a deep understanding of the impact of excessive pesticide residues on people's health. At present, agricultural product sales are mainly divided into two ways: online and offline. Supervision also has loopholes in pesticide sales and agricultural product testing. The food safety of some agricultural products cannot be guaranteed. Food poisoning caused by excessive pesticide residues is emerging in an endless stream, which seriously affects public food safety. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method, system and medium for detecting safety of food pesticide residues.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention provides a method for detecting food pesticide residue safety, comprising the following steps:

[0006] Acquire gas electrical signal information in a current area within a preset time through an array-type odor sensor, and preprocess the gas electrical signal information to obtain a preprocessing result;

[0007] Establishing an odor recognition model, identifying the preprocessing result through the odor recognition model to obtain relevant odor types, and generating a recognition result according to the relevant types;

[0008] Acquire crop category information of the current storage area, and perform a search based on the crop category to obtain a search result;

[0009] A matching result is obtained by matching the recognition result and the search result, and the abnormal location node of the current food is determined based on the matching result, and is displayed on the food detection terminal in a preset manner.

[0010] Furthermore, in a preferred embodiment of the present invention, an odor recognition model is established, and the preprocessing result is recognized by the odor recognition model to obtain the relevant odor type, which specifically includes the following steps:

[0011] Establishing an odor recognition model based on a neural network, performing initialization parameter settings, obtaining odor sample data from a large database, training the odor recognition model using the odor sample data, adjusting model parameters and retaining optimal model parameters;

[0012] Performing feature extraction on the preprocessing result to obtain an odor feature matrix, and using the odor feature matrix as an input signal;

[0013] The input signal is input into a neural network, and after the convolution layer operation, a linear rectification function is used as an activation function, and the odor type is classified by a Softmax classifier to obtain odor types of related categories and odor types of unrelated categories;

[0014] If there is a related type of odor type, the related type of odor type is used as the related odor type, and a recognition result is generated according to the related odor type.

[0015] Furthermore, in a preferred embodiment of the present invention, the crop category information of the current storage area is obtained, and a search is performed according to the crop category to obtain a search result, which specifically includes the following steps:

[0016] Determining whether the recognition result is greater than a preset recognition result, and if the recognition result is greater than the preset recognition result, obtaining crop category information in the current storage area;

[0017] Obtain disease type information of various crops through a big data network, and obtain the type of pesticide to be used under the disease type information according to the disease type information;

[0018] Constructing a database, importing the pesticide usage types under the various disease type information into the database to obtain a trained database;

[0019] The crop category information in the current storage area is imported into the trained database to obtain the pesticide application type of the crops in the current storage area, and the pesticide application type is output as a search result.

[0020] Furthermore, in a preferred embodiment of the present invention, matching is performed according to the recognition result and the search result to obtain a matching result, which specifically includes the following steps:

[0021] Acquire pesticide character string information of a current recognition result, and generate first character string information according to the pesticide character string information of the current recognition result;

[0022] Acquire multiple string information of current search results, and generate multiple second string information according to the multiple string information of current search results;

[0023] Comparing the first character string information with the second character string information one by one to obtain multiple similarities, and determining whether the similarities are greater than a preset similarity;

[0024] If the similarity is greater than a preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node, and the abnormal pesticide residue node is output as a matching result.

[0025] Furthermore, in a preferred embodiment of the present invention, the abnormal location node of the current food is determined based on the matching result and displayed on the food detection terminal in a preset manner, which specifically includes the following steps:

[0026] Obtaining a planar structural diagram of the current storage area, and obtaining a storage area location node of each food according to the planar structural diagram;

[0027] Acquire the character string information of the storage area location node of each food, perform secondary matching with the matching result based on the character string information of the storage area location node of each food, and obtain the similarity of the secondary matching result;

[0028] Obtaining a location node where the character string information having a similarity of the secondary matching result greater than a preset similarity of the secondary matching result is located;

[0029] The position node where the character string information having the secondary matching result similarity greater than the preset secondary matching result similarity is located is taken as an abnormal position node, and the abnormal position node is displayed on the food detection terminal in a preset manner.

[0030] Furthermore, in a preferred embodiment of the present invention, the gas electrical signal information of the current area within a preset time is obtained by an array odor sensor, and the gas electrical signal information is preprocessed to obtain a preprocessing result, which specifically includes the following steps:

[0031] The gas electrical signal information of the current area at a preset time is obtained through an array odor sensor, and the relative deviation of N sample data within the preset time is calculated through the RSD method;

[0032] Determining whether the relative deviation is greater than a preset relative deviation, and if the relative deviation is greater than the preset relative deviation, establishing a gas sensor response data curve based on the gas electrical signal information of the current area at a preset time;

[0033] Performing smoothing processing on the gas sensor response data curve through a filter to obtain a processed response data curve;

[0034] The gas data information of the processed response data curve in the stable phase is obtained, and the gas data information is output as a preprocessing result.

[0035] A second aspect of the present invention provides a food pesticide residue safety detection system, the food pesticide residue safety detection system comprising a memory and a processor, the memory comprising a food pesticide residue safety detection method program, and when the food pesticide residue safety detection method program is executed by the processor, the following steps are implemented:

[0036] Acquire gas electrical signal information in a current area within a preset time through an array-type odor sensor, and preprocess the gas electrical signal information to obtain a preprocessing result;

[0037] Establishing an odor recognition model, identifying the preprocessing result through the odor recognition model to obtain relevant odor types, and generating a recognition result according to the relevant types;

[0038] Acquire crop category information of the current storage area, and perform a search based on the crop category to obtain a search result;

[0039] A matching result is obtained by matching the recognition result and the search result, and the abnormal location node of the current food is determined based on the matching result, and is displayed on the food detection terminal in a preset manner.

[0040] Furthermore, in a preferred embodiment of the present invention, an odor recognition model is established, and the preprocessing result is recognized by the odor recognition model to obtain the relevant odor type, which specifically includes the following steps:

[0041] Establishing an odor recognition model based on a neural network, performing initialization parameter settings, obtaining odor sample data from a large database, training the odor recognition model using the odor sample data, adjusting model parameters and retaining optimal model parameters;

[0042] Performing feature extraction on the preprocessing result to obtain an odor feature matrix, and using the odor feature matrix as an input signal;

[0043] The input signal is input into a neural network, and after the convolution layer operation, a linear rectification function is used as an activation function, and the odor type is classified by a Softmax classifier to obtain odor types of related categories and odor types of unrelated categories;

[0044] If there is a related type of odor type, the related type of odor type is used as the related odor type, and a recognition result is generated according to the related odor type.

[0045] Furthermore, in a preferred embodiment of the present invention, matching is performed according to the recognition result and the search result to obtain a matching result, which specifically includes the following steps:

[0046] Acquire pesticide character string information of a current recognition result, and generate first character string information according to the pesticide character string information of the current recognition result;

[0047] Acquire multiple string information of current search results, and generate multiple second string information according to the multiple string information of current search results;

[0048] Comparing the first character string information with the second character string information one by one to obtain multiple similarities, and determining whether the similarities are greater than a preset similarity;

[0049] If the similarity is greater than a preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node, and the abnormal pesticide residue node is output as a matching result.

[0050] The third aspect of the present invention provides a computer-readable storage medium, which includes a food pesticide residue safety detection method program. When the food pesticide residue safety detection method program is executed by a processor, the steps of any one of the food pesticide residue safety detection methods are implemented.

[0051] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0052] The present invention obtains the gas electric signal information of the current area within a preset time through an array odor sensor, and pre-processes the gas electric signal information to obtain a pre-processing result; establishes an odor recognition model, identifies the pre-processing result through the odor recognition model, obtains the relevant odor type, and generates an identification result according to the relevant type; obtains the crop category information of the current storage area, and searches according to the crop category to obtain a search result; matches according to the recognition result and the search result to obtain a matching result, and determines the abnormal position node of the current food based on the matching result, and displays it on the food detection terminal in a preset manner. Through this method, crops with more pesticide residues in crop storage can be quickly detected, so that the position nodes of the area where the crops with serious pesticide residues are located can be detected, and then the relevant personnel can be notified to rectify this type of crops. Through this method, the position nodes with pesticide residues can be quickly detected, which is more efficient than manual labor, and is conducive to improving the export safety of food, and to a certain extent, guarantees the safety of consumers. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0054] Figure 1 The overall method flow chart of a food pesticide residue safety detection method is shown;

[0055] Figure 2 A flow chart of a method for obtaining search results is shown;

[0056] Figure 3 A flow chart of a method for obtaining a matching result is shown;

[0057] Figure 4 A system block diagram of a food pesticide residue safety detection system is shown. DETAILED DESCRIPTION

[0058] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0060] Figure 1 The overall method flow chart of a food pesticide residue safety detection method is shown;

[0061] The first aspect of the present invention provides a method for detecting food pesticide residue safety, comprising the following steps:

[0062] S102: acquiring gas electrical signal information in a current area within a preset time through an array odor sensor, and preprocessing the gas electrical signal information to obtain a preprocessing result;

[0063] S104: establishing an odor recognition model, identifying the preprocessing result through the odor recognition model, obtaining a related odor type, and generating a recognition result according to the related type;

[0064] S106: Acquire crop category information of the current storage area, and perform a search based on the crop category to obtain a search result;

[0065] S108: Matching is performed according to the recognition result and the search result to obtain a matching result, and based on the matching result, an abnormal location node of the current food is determined, and the abnormal location node is displayed on the food detection terminal in a preset manner.

[0066] It should be noted that this method can quickly detect crops with more pesticide residues in crop storage, thereby detecting the location nodes of the areas where crops with serious pesticide residues are located, and then notifying relevant personnel to make rectifications on such crops; another method is that this method can quickly detect the location nodes with pesticide residues, which is more efficient than manual labor, and is conducive to improving the export safety of food, ensuring the safety of consumers to a certain extent.

[0067] Furthermore, in a preferred embodiment of the present invention, an odor recognition model is established, and the preprocessing result is recognized by the odor recognition model to obtain the relevant odor type, which specifically includes the following steps:

[0068] Establishing an odor recognition model based on a neural network, performing initialization parameter settings, obtaining odor sample data from a large database, training the odor recognition model using the odor sample data, adjusting model parameters and retaining optimal model parameters;

[0069] Performing feature extraction on the preprocessing result to obtain an odor feature matrix, and using the odor feature matrix as an input signal;

[0070] The input signal is input into a neural network, and after the convolution layer operation, a linear rectification function is used as an activation function, and the odor type is classified by a Softmax classifier to obtain odor types of related categories and odor types of unrelated categories;

[0071] If there is a related type of odor type, the related type of odor type is used as the related odor type, and a recognition result is generated according to the related odor type.

[0072] It should be noted that the identification result includes the type of residual pesticides, the concentration information of residual pesticides measured by the gas sensor, etc. The method can identify the pesticide odor information in the crop storage area. Since most of the crops are stored in refrigerated warehouses for refrigeration after picking, this can be beneficial to the preservation of crops. Refrigerated warehouses are equipped with storage areas for different crops. When there are a large amount of pesticide residues in the crops, the air in the cold storage room will also have a certain amount of pesticide residues. The air can be monitored by the gas sensor. When the pesticide residue content in the air is greater than the preset content (the identification result is greater than the preset identification result), unqualified crop storage nodes can be quickly detected.

[0073] Figure 2 A flow chart of a method for obtaining search results is shown;

[0074] Furthermore, in a preferred embodiment of the present invention, the crop category information of the current storage area is obtained, and a search is performed according to the crop category to obtain a search result, which specifically includes the following steps:

[0075] S202: determining whether the recognition result is greater than a preset recognition result, and if the recognition result is greater than the preset recognition result, obtaining crop category information in the current storage area;

[0076] S204: obtaining disease type information of various crops through a big data network, and obtaining the pesticide use type under the various disease type information according to the disease type information;

[0077] S206: constructing a database, importing the pesticide usage types under the various disease type information into the database, and obtaining a trained database;

[0078] S208: Importing the crop category information in the current storage area into the trained database to obtain the pesticide usage type of the crops in the current storage area, and outputting the pesticide usage type as a search result.

[0079] It should be noted that, since there are multiple types of crops stored in the current food storage area, each type of crop has different types of diseases or pests during the growth stage, so the types of pesticides used for various crops are roughly inconsistent or there may be some identical parts. Through this method, the types of pesticides that may have been used for various crops in the current storage area during the growth process due to diseases or pests can be quickly retrieved, so as to further quickly locate the crops with serious pesticide residues in the current storage area.

[0080] Figure 3 A flow chart of a method for obtaining a matching result is shown;

[0081] Furthermore, in a preferred embodiment of the present invention, matching is performed according to the recognition result and the search result to obtain a matching result, which specifically includes the following steps:

[0082] S302: Acquire pesticide character string information of a current recognition result, and generate first character string information according to the pesticide character string information of the current recognition result;

[0083] S304: Acquire multiple string information of the current search result, and generate multiple second string information according to the multiple string information of the current search result;

[0084] S306: Compare the first character string information with the second character string information one by one to obtain multiple similarities, and determine whether the similarities are greater than a preset similarity;

[0085] S308: If the similarity is greater than a preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node, and the abnormal pesticide residue node is output as a matching result.

[0086] It should be noted that this method can quickly detect crops or foods with serious pesticide residues. The preset similarity is set to 100%, and a similarity can be obtained by comparing the character strings. When the similarity is greater than the preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node.

[0087] Furthermore, in a preferred embodiment of the present invention, the abnormal location node of the current food is determined based on the matching result and displayed on the food detection terminal in a preset manner, which specifically includes the following steps:

[0088] Obtaining a planar structural diagram of the current storage area, and obtaining a storage area location node of each food according to the planar structural diagram;

[0089] Acquire the character string information of the storage area location node of each food, perform secondary matching with the matching result based on the character string information of the storage area location node of each food, and obtain the similarity of the secondary matching result;

[0090] Obtaining a location node where the character string information having a similarity of the secondary matching result greater than a preset similarity of the secondary matching result is located;

[0091] The position node where the character string information having the secondary matching result similarity greater than the preset secondary matching result similarity is located is taken as an abnormal position node, and the abnormal position node is displayed on the food detection terminal in a preset manner.

[0092] It should be noted that this method can quickly detect abnormal location nodes of crops with excessively high pesticide residues in the current storage area, thereby notifying relevant staff to further process the crops with excessively high pesticide residues. The preset secondary matching result similarity is 100%, and the location node where the string information with the secondary matching result similarity greater than the preset secondary matching result similarity is located is regarded as an abnormal location node. The plane structure schematic diagram is a plane structure schematic diagram of the placement area of ​​each crop set by the system at the beginning.

[0093] Furthermore, in a preferred embodiment of the present invention, the gas electrical signal information of the current area within a preset time is obtained by an array odor sensor, and the gas electrical signal information is preprocessed to obtain a preprocessing result, which specifically includes the following steps:

[0094] The gas electrical signal information of the current area at a preset time is obtained through an array odor sensor, and the relative deviation of N sample data within the preset time is calculated through the RSD method;

[0095] Determining whether the relative deviation is greater than a preset relative deviation, and if the relative deviation is greater than the preset relative deviation, establishing a gas sensor response data curve based on the gas electrical signal information of the current area at a preset time;

[0096] Performing smoothing processing on the gas sensor response data curve through a filter to obtain a processed response data curve;

[0097] The gas data information of the processed response data curve in the stable phase is obtained, and the gas data information is output as a preprocessing result.

[0098] It should be noted that in order for the program to intelligently determine whether the steady-state stage has been reached, the RSD (relative standard deviation) method is used to determine whether the response of the gas sensor has reached stability. Calculate the relative standard deviation of the data of N sample points within a continuous period of time (such as 30 seconds). If it is sufficiently smaller than the set threshold (such as 10%), the response curve of the sensor is considered to have reached stability. Every time a new sampling data arrives, it must be recalculated. When the response curves of n sensors are all stable at the same time, it is considered that the steady-state stage has been reached. Therefore, this method can make the gas signal obtained by the gas sensor more reliable.

[0099] In addition, the method may further comprise the following steps:

[0100] Obtaining the pesticide application type of the crops in the current storage area, and determining whether there are crops using the same pesticide as the pesticide application type of the crops in the current storage area;

[0101] If there are crops using the same pesticide as the pesticide use type of the crops in the current storage area, then the pesticide use growth stage corresponding to the crop is obtained through the big data network;

[0102] Determining whether the used growth stage is a preset growth stage, and if the growth stage is a preset growth stage, taking the current crop as an abnormal pesticide residue node;

[0103] If the growth stage is not the preset growth stage, the current crop is regarded as a normal node.

[0104] It should be noted that during the detection process, two different crops may have the same pesticide usage in each growth stage, and the pesticide is not suitable for early use for some crops, but is suitable for use in the later stage of the crop growth stage. When this happens, by judging whether the growth stage is the preset growth stage, if the growth stage is the preset growth stage, it means that the crop is an abnormal pesticide residue node. For example, some pesticides can only be used in the early stage of crops and are not suitable for later use; it means that the growth stage is not the preset growth stage, and the current crop is regarded as a normal node. This method can make the detection more reasonable.

[0105] In addition, the method may further comprise the following steps:

[0106] Obtaining the process information of the current abnormal pesticide residue node, and retrieving the work information in the process information;

[0107] Determine whether there is a vacancy in the current work information, and if there is a vacancy in the work information, obtain the process node where the work information is located;

[0108] Obtaining working information of the current process node according to the process node, and obtaining the location node of the current abnormal pesticide residue node;

[0109] The working information of the current process node and the location node of the current abnormal pesticide residue node are displayed on the food detection terminal in a preset manner.

[0110] It should be noted that the process information can be understood as crops need to undergo a series of processing before being transported to a designated area for storage, such as needing to undergo cleaning, disinfection, etc. The work information can be whether they have undergone cleaning or disinfection. When there are gaps in the work information, the method can quickly detect whether there are abnormalities in the process information of the abnormal pesticide residue node. When there are abnormalities, the missing work information is supplemented in time to prompt the staff to process the food in time according to the work information.

[0111] Figure 4 A system block diagram of a food pesticide residue safety detection system is shown.

[0112] A second aspect of the present invention provides a food pesticide residue safety detection system, the food pesticide residue safety detection system comprises a memory 41 and a processor 62, the memory 41 comprises a food pesticide residue safety detection method program, and when the food pesticide residue safety detection method program is executed by the processor 62, the following steps are implemented:

[0113] Acquire gas electrical signal information in a current area within a preset time through an array-type odor sensor, and preprocess the gas electrical signal information to obtain a preprocessing result;

[0114] Establishing an odor recognition model, identifying the preprocessing result through the odor recognition model to obtain relevant odor types, and generating a recognition result according to the relevant types;

[0115] Acquire crop category information of the current storage area, and perform a search based on the crop category to obtain a search result;

[0116] A matching result is obtained by matching the recognition result and the search result, and the abnormal location node of the current food is determined based on the matching result, and is displayed on the food detection terminal in a preset manner.

[0117] It should be noted that this method can quickly detect crops with more pesticide residues in crop storage, thereby detecting the location nodes of the areas where crops with serious pesticide residues are located, and then notifying relevant personnel to make rectifications on such crops; another method is that this method can quickly detect the location nodes with pesticide residues, which is more efficient than manual labor, and is conducive to improving the export safety of food, ensuring the safety of consumers to a certain extent.

[0118] Furthermore, in a preferred embodiment of the present invention, an odor recognition model is established, and the preprocessing result is recognized by the odor recognition model to obtain the relevant odor type, which specifically includes the following steps:

[0119] Establishing an odor recognition model based on a neural network, performing initialization parameter settings, obtaining odor sample data from a large database, training the odor recognition model using the odor sample data, adjusting model parameters and retaining optimal model parameters;

[0120] Performing feature extraction on the preprocessing result to obtain an odor feature matrix, and using the odor feature matrix as an input signal;

[0121] The input signal is input into a neural network, and after the convolution layer operation, a linear rectification function is used as an activation function, and the odor type is classified by a Softmax classifier to obtain odor types of related categories and odor types of unrelated categories;

[0122] If there is a related type of odor type, the related type of odor type is used as the related odor type, and a recognition result is generated according to the related odor type.

[0123] Furthermore, in a preferred embodiment of the present invention, matching is performed according to the recognition result and the search result to obtain a matching result, which specifically includes the following steps:

[0124] Acquire pesticide character string information of a current recognition result, and generate first character string information according to the pesticide character string information of the current recognition result;

[0125] Acquire multiple string information of current search results, and generate multiple second string information according to the multiple string information of current search results;

[0126] Comparing the first character string information with the second character string information one by one to obtain multiple similarities, and determining whether the similarities are greater than a preset similarity;

[0127] If the similarity is greater than a preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node, and the abnormal pesticide residue node is output as a matching result.

[0128] Furthermore, in a preferred embodiment of the present invention, the abnormal location node of the current food is determined based on the matching result and displayed on the food detection terminal in a preset manner, which specifically includes the following steps:

[0129] Obtaining a planar structural diagram of the current storage area, and obtaining a storage area location node of each food according to the planar structural diagram;

[0130] Acquire the character string information of the storage area location node of each food, perform secondary matching with the matching result based on the character string information of the storage area location node of each food, and obtain the similarity of the secondary matching result;

[0131] Obtaining a location node where the character string information having a similarity of the secondary matching result greater than a preset similarity of the secondary matching result is located;

[0132] The position node where the character string information having the secondary matching result similarity greater than the preset secondary matching result similarity is located is taken as an abnormal position node, and the abnormal position node is displayed on the food detection terminal in a preset manner.

[0133] It should be noted that this method can quickly detect abnormal location nodes of crops with excessively high pesticide residues in the current storage area, thereby notifying relevant staff to carry out further processing work on the crops with excessively high pesticide residues.

[0134] Furthermore, in a preferred embodiment of the present invention, the gas electrical signal information of the current area within a preset time is obtained by an array odor sensor, and the gas electrical signal information is preprocessed to obtain a preprocessing result, which specifically includes the following steps:

[0135] The gas electrical signal information of the current area at a preset time is obtained through an array odor sensor, and the relative deviation of N sample data within the preset time is calculated through the RSD method;

[0136] Determining whether the relative deviation is greater than a preset relative deviation, and if the relative deviation is greater than the preset relative deviation, establishing a gas sensor response data curve based on the gas electrical signal information of the current area at a preset time;

[0137] Performing smoothing processing on the gas sensor response data curve through a filter to obtain a processed response data curve;

[0138] The gas data information of the processed response data curve in the stable phase is obtained, and the gas data information is output as a preprocessing result.

[0139] It should be noted that in order to allow the program to intelligently determine whether the steady-state stage has been reached, we use the RSD method to determine whether the response of the gas sensor has reached stability. Calculate the relative standard deviation of the data of N sample points within a continuous period of time (such as 30 seconds). If it is sufficiently smaller than the set threshold (such as 10%), the response curve of the sensor is considered to have reached stability. Every time a new sampling data arrives, it must be recalculated. When the response curves of n sensors are all stable at the same time, it is considered that the steady-state stage has been reached. Therefore, this method can make the gas signal obtained by the gas sensor more reliable.

[0140] A third aspect of the present invention provides a computer-readable storage medium, which includes a food pesticide residue safety detection method program. When the food pesticide residue safety detection method program is executed by the processor 62, the steps of any one of the food pesticide residue safety detection methods are implemented.

[0141] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0142] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0143] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0144] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0145] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0146] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for detecting safety of pesticide residues in food, characterized in that: The following steps are involved: Acquire gas electrical signal information in a current area within a preset time through an array-type odor sensor, and preprocess the gas electrical signal information to obtain a preprocessing result; Establishing an odor recognition model, identifying the preprocessing result through the odor recognition model to obtain relevant odor types, and generating a recognition result according to the relevant odor types; Acquire crop category information of the current storage area, and perform a search based on the crop category to obtain a search result; Matching is performed according to the recognition result and the search result to obtain a matching result, and based on the matching result, an abnormal location node of the current food is determined, and displayed on the food detection terminal in a preset manner; The following steps are also included: Obtaining the pesticide application type of the crops in the current storage area, and determining whether there are crops using the same pesticide as the pesticide application type of the crops in the current storage area; If there are crops using the same pesticide as the pesticide use type of the crops in the current storage area, then the pesticide use growth stage corresponding to the crop is obtained through the big data network; Determining whether the used growth stage is a preset growth stage, and if the growth stage is a preset growth stage, taking the current crop as an abnormal pesticide residue node; If the growth stage is not the preset growth stage, the current crop is regarded as a normal node; The preprocessing result is identified by the odor recognition model to obtain the relevant odor type, which specifically includes the following steps: establishing an odor recognition model based on a neural network, performing initialization parameter setting, obtaining odor sample data from a large database, training the odor recognition model with the odor sample data, adjusting the model parameters and retaining the optimal model parameters; Performing feature extraction on the preprocessing result to obtain an odor feature matrix, and using the odor feature matrix as an input signal; The input signal is input into a neural network, and after the convolution layer operation, a linear rectification function is used as an activation function, and the odor type is classified by a Softmax classifier to obtain odor types of related categories and odor types of unrelated categories; If there is a related type of odor type, use the related type of odor type as the related odor type, and generate a recognition result according to the related odor type; The gas electrical signal information of the current area within a preset time is obtained by using an array type odor sensor, and the gas electrical signal information is preprocessed to obtain a preprocessing result, which specifically includes the following steps: The gas electrical signal information of the current area at a preset time is obtained through an array odor sensor, and the relative deviation of N sample data within the preset time is calculated through the RSD method; Determining whether the relative deviation is greater than a preset relative deviation, and if the relative deviation is greater than the preset relative deviation, establishing a gas sensor response data curve based on the gas electrical signal information of the current area at a preset time; Performing smoothing processing on the gas sensor response data curve through a filter to obtain a processed response data curve; The gas data information of the processed response data curve in the stable phase is obtained, and the gas data information is output as a preprocessing result.

2. A food pesticide residue safety detection method according to claim 1, characterized in that: Acquiring crop category information of the current storage area, and searching according to the crop category to obtain a search result, specifically comprising the following steps: determining whether the recognition result is greater than a preset recognition result, and if the recognition result is greater than the preset recognition result, acquiring crop category information in the current storage area; Obtain disease type information of various crops through a big data network, and obtain the type of pesticide to be used under the disease type information according to the disease type information; Constructing a database, importing the pesticide usage types under the various disease type information into the database to obtain a trained database; The crop category information in the current storage area is imported into the trained database to obtain the pesticide application type of the crops in the current storage area, and the pesticide application type is output as a search result.

3. A food pesticide residue safety detection method according to claim 1, characterized in that: Matching is performed according to the recognition result and the search result to obtain a matching result, specifically comprising the following steps: obtaining pesticide character string information of a current recognition result, and generating first character string information according to the pesticide character string information of the current recognition result; Acquire multiple string information of current search results, and generate multiple second string information according to the multiple string information of current search results; Comparing the first character string information with the second character string information one by one to obtain multiple similarities, and determining whether the similarities are greater than a preset similarity; If the similarity is greater than a preset similarity, the crop corresponding to the pesticide type of the second character string information is used as an abnormal pesticide residue node, and the abnormal pesticide residue node is output as a matching result.

4. A food pesticide residue safety detection method according to claim 1, characterized in that: Determine the abnormal location node of the current food based on the matching result, and display it on the food detection terminal in a preset manner, specifically including the following steps: obtain a plane structure diagram of the current storage area, and obtain the storage area location node of each food according to the plane structure diagram; Acquire the character string information of the storage area location node of each food, perform secondary matching with the matching result based on the character string information of the storage area location node of each food, and obtain the similarity of the secondary matching result; Obtaining a location node where the character string information having a similarity of the secondary matching result greater than a preset similarity of the secondary matching result is located; The position node where the character string information having the secondary matching result similarity greater than the preset secondary matching result similarity is located is taken as an abnormal position node, and the abnormal position node is displayed on the food detection terminal in a preset manner.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a food pesticide residue safety detection method program, and when the food pesticide residue safety detection method program is executed by a processor, the steps of the food pesticide residue safety detection method as described in any one of claims 1 to 4 are implemented.

Images