CN109856344A - A kind of food safety sampling Detection equipment - Google Patents
A kind of food safety sampling Detection equipment Download PDFInfo
- Publication number
- CN109856344A CN109856344A CN201910114422.4A CN201910114422A CN109856344A CN 109856344 A CN109856344 A CN 109856344A CN 201910114422 A CN201910114422 A CN 201910114422A CN 109856344 A CN109856344 A CN 109856344A
- Authority
- CN
- China
- Prior art keywords
- food
- module
- detection
- sampling
- content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention belongs to technical field of food safety detection, a kind of food safety sampling Detection equipment is disclosed, the food safety sampling Detection equipment includes: Pesticides Testing module, brittleness detection module, nitrate detection module, formaldehyde examination module, main control module, extraction module, crushes module, safety evaluation module, detection data memory module, display module.The present invention can carry out multinomial detection by Pesticides Testing module to food residue pesticide, processing straightforward procedure, low in cost.Meanwhile ensure that the sample of extraction has the representativeness of science by safety evaluation module, fill up blank of the domestic food based on comprehensive assessment sampling by classification；Food implements sampling with unequal probability after being rationally layered, sample calculation according to more scientific, the method than existing food random sampling is more effective.
Description
Technical field
The invention belongs to technical field of food safety detection more particularly to a kind of food safety sampling Detection equipment.
Background technique
Food safety (foodsafety) refers to that food is nontoxic, harmless, meets the nutritional requirement that should have, not to human health
Cause any acute, subacute or chronic hazard.According to promise food safety again define, food safety be " it is toxic in food, have
Public health problem of the evil substance to Health Impact ".Food safety be also one it is special inquire into food processing, storage,
Ensure food hygiene and edible safety during sale etc., reduce disease risk, take precautions against an interdisciplinary fields of food poisoning,
So food safety is critically important.However, the method that existing food safety sampling Detection equipment is measured persticide residue, inspection
Survey process is cumbersome, spends human and material resources；Meanwhile the reliability of existing extraction mode is insufficient, testing result also will receive very big shadow
Ring, can not accurate response entirety food quality and safe condition.
In conclusion problem of the existing technology is:
(1) method that existing food safety sampling Detection equipment is measured persticide residue, detection process is cumbersome, consumption
Take manpower and material resources；Meanwhile the reliability of existing extraction mode is insufficient, testing result also will receive extreme influence, can not be accurately anti-
Answer the quality and safe condition of whole food.
(2) sound characteristic judges the mistake of food brittleness data when voice detector detection food samples crush in the prior art
Cheng Zhong, the discontinuity of hard threshold function will cause the oscillation of signal, and soft-threshold function is too smooth to will cause signal high frequency
Information is lost, and cannot be retained the characteristic spikes point information of original signal, be improved the distortion of signal.
(3) formaldehyde examination sensor detects in chilled class aquatic products and its dried product during content of formaldehyde data, adopts
Signal with the detection on traditional algorithm is undesirable, the weak phenomenon of anti-interference ability, at the same avoid to the signal of detection into
When row decomposes, using end effect present in empirical mode decomposition and modal overlap phenomenon.
(4) in the process of processing to different types of detection data, it using current data processing algorithm, reduces
The local search ability of K mean algorithm reduces convergence rate, not can effectively stop the generation of precocious phenomenon, can not achieve
Preferably convergence effect.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of food safety sampling Detection equipment.
The invention is realized in this way a kind of food safety sampling Detection method, comprising the following steps:
Extractor extracts food samples, carries out pulverization process to food samples using pulverizer；
The food samples of pulverization process are passed through into relevant operation, detect residue of pesticide concentration；Food samples are detected to crush
When sound characteristic judge food brittleness；Detect nitrate content in cure foods and meat products；Detect chilled class aquatic products and its
Content of formaldehyde in dried product；
Food is pacified according to the data of above-mentioned residue of pesticide concentration, food brittleness, nitrate content, content of formaldehyde detection
It is evaluated entirely；
Pass through pesticide residue, the brittleness, nitrate content, content of formaldehyde data information of memory storage detection；By aobvious
Show pesticide residue, brittleness, nitrate content, content of formaldehyde data information and the evaluation result of device display detection.
Further, the process that sound characteristic when food samples crush judges food brittleness data is detected by voice detector
In, using improved wavelet threshold speech de-noising algorithm, specifically include:
Step 1, selects db3 as wavelet basis, carries out 5 layers of wavelet decomposition to the voice signal f (k) that band is made an uproar, obtains one
The wavelet coefficient w of seriesJ, k；
Step 2, using formula
Wherein: y is improved threshold function table；T is to estimate threshold value by soft-threshold function；
Improved threshold function table is to wJ, kThreshold process is carried out, obtains estimation wavelet coefficientMakeTo the greatest extent
It may be small；
Step 3, the estimation wavelet coefficient obtained according to step 2Small echo is reconstructed, the signal after being denoised
Further, formaldehyde examination sensor detects the process of content of formaldehyde data in chilled class aquatic products and its dried product
In, joint Denoising Algorithm is decomposed using based on ICA algorithm and EEMD, is specifically included:
Step 1, is added white noise signal in original signal, the end effect that inhibits EMD decomposable process finally to occur and
Modal overlap phenomenon；
Step 2 carries out EMD decomposition to composite signal, obtains several IMF components, removes first containing main noise
IMF component；
Step 3 carries out FastICA algorithm to remaining IMFs, solves hybrid matrix A, separation matrix W and isolated component
The matrix S of composition；
Step 4, selects the independent source component of predominantly noise source, carries out EMD denoising to it, only after being denoised
Vertical source component S '；
Hybrid matrix A is multiplied with the independent source component S ' after denoising and reconstructs new IMF collection by step 5, the IMF collection phase
Add the signal after denoising can be obtained.
Further, the Pesticides Testing module detection method includes:
(1) training sample data are acquired, according to training sample data training sample, obtain remains of pesticide type and content
Curve；
(2) big data resource is concentrated to carry out processing point to the curve data of remains of pesticide type and content by Cloud Server
Analysis；
(3) food samples are obtained, food samples are pulverized；
(4) quantitative water is added, is diluted to food samples are extracted, and be sufficiently stirred, obtains food samples extracting solution；
(5) food samples extracting solution is obtained, farm chemical ingredients and content detection are carried out to food samples extracting solution；
(6) according to the farm chemical ingredients of detection and content, the severity of food residue pesticide is determined.
Further, the safety evaluation module evaluation method includes:
1) food samples are extracted by extractor；
2) food sampling is divided into two dimensions, is that ordinary trade food is sampled, cross-border electric business food is sampled respectively, simultaneously
To the two dimensions formulate respectively several first class index, two-level index and three-level index instant food type, weights of importance coefficient,
Risk rated ratio coefficient；
3) according to AHP analytic hierarchy process (AHP), several first class index, two-level index and three-level index of two dimensions are constructed
At the food sampling system model based on risk assessment；
4) according to the composition and Food Inspection regulatory status of China's food business, by convention will according to animal product and
Three kinds of its product, plant product and its product, food traditional food Main Foods types of service, inviting includes that food technology is special
Family, foods supervision expert and third party Quality Examination expert form expert group, and expert group's total number of persons is M, and asks expert
To several first class index of two dimensions, two-level index and three-level index instant food type, important coefficient, risk factor
The marking of solution degree is determining each index to determine several first class index of two dimensions, the weight of two-level index and three-level index
Before weight, the calculating of food expert domain coefficient, calculation formula are first carried out respectively are as follows:Its
In: N is expert's sum；I=expert's serial number, i=1,2,3...N；PiFor technical specialist authority's coefficient, whereinKi
It is technical specialist to the degree of understanding coefficient of index；SiFor the specialty background coefficient of technical specialist；
5) based food link and the characteristics of quality-monitoring, to the weights of importance coefficient Z of every kind of food typek, risk
Weight coefficient RkTest assessment is carried out, and generates the sampling comprehensive weight coefficient C of food sampling index system in turnk；
6) first class index instant food type is divided into K layers, sets total batch of food sampling as F, is taken out by introducing food
Sample comprehensive weight coefficient Ck, obtain 1-K layers of sampling batch FK, FK=Ck*F；
7) it according to the layering of first class index instant food type, carries out implementing sampling with unequal probability in K layers, enabling K layers is it
In a food type, this food type is divided into E kind unit again；The total scale measurement of K layers of food type is M, no
Size measurement with kind unit distinguishes M1、M2、M3...Me, whereinIntroduce K layers of sampling batch FK, then equidistantly
Sampling interval are as follows: D=M/FK；Then a number is taken at random between 1-D, it is assumed that be X, then the cell codes section where X
Corresponding unit is the unit drawn, later every D metric, i.e. X+D, X+2D, X+3D ... X+ (FK- 1) number such as D
The corresponding units in the cell codes section where word, the unit as drawn；When the Size measurement of all different cultivars units
When Me < D, determine that it is unduplicated sampling；As the Size measurement Me > D of some different cultivars unit, then determine e-th
Unit is possible to be repeated and draw；As the Size measurement Me < 2D of different cultivars unit, then determine e-th of unit be certain to by
Repetition is drawn.
Another object of the present invention is to provide a kind of computer programs for realizing the food safety sampling Detection method.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer
When upper operation, so that computer executes the food safety sampling Detection method.
Another object of the present invention is to provide a kind of food safety sampling Detection equipment, the food safety sampling Detection
Equipment includes:
Pesticides Testing module, connect with main control module, for by food Detecting Pesticide device detect vegetables, melon and fruit,
Residue of pesticide data in the food such as grain, tealeaves；
Brittleness detection module, connect with main control module, for detecting sound when food samples crush by voice detector
Feature judges food brittleness data；
Nitrate detection module, connect with main control module, for detecting cure foods and meat system by nitrate detector
Nitrate content data in product；
Formaldehyde examination module, connect with main control module, for detecting chilled class aquatic products by formaldehyde detector and its doing
Content of formaldehyde data in product；
Main control module, with Pesticides Testing module, brittleness detection module, nitrate detection module, formaldehyde examination module, extraction
Module crushes module, safety evaluation module, detection data memory module, display module connection, each for being controlled by single-chip microcontroller
A module works normally；
Extraction module is connect with main control module, for extracting food samples by extractor；
Module is crushed, is connect with main control module, for carrying out pulverization process to food samples by pulverizer；
Safety evaluation module, connect with main control module, for food-safe according to the data of detection by assessment process
It is evaluated；
Detection data memory module, connect with main control module, for the pesticide residue, crisp by memory storage detection
Degree, nitrate content, content of formaldehyde data information；
Display module is connect with main control module, for pesticide residue, the brittleness, nitrate by display display detection
Content, content of formaldehyde data information and evaluation result.
The food safety sampling Detection equipment Internet of Things food peace is carried another object of the present invention is to provide a kind of
Full detection platform.
Advantages of the present invention and good effect are as follows:
The present invention can carry out multinomial detection by Pesticides Testing module to food residue pesticide, processing straightforward procedure, at
This is cheap.The present invention can be realized qualitative and quantitative analysis, be suitable for the remaining measurement of Multiple Pesticides, high sensitivity, testing result
Accurately, the development and popularization of existing market are more advantageous to；Meanwhile passing through safety evaluation module country science measuring and calculating for the first time food quality
Important sexual factor is influenced safely and on the people's livelihood, thus one reasonable hierarchical structure sampling model of construction；First passage is comprehensive
It closes assessment to evaluate the element for influencing food safety quality, it is determined that regional large grain and oil and people's livelihood food, emphasis are quick
The weight coefficient for feeling overseas direct mail and cross-border electric business food variety is reasonably layered food, and determines sampling layer power, guarantees
The sample extracted has the representativeness of science, has filled up blank of the domestic food based on comprehensive assessment sampling by classification；Food closes
Implement sampling with unequal probability after reason layering, sample calculation according to more scientific, than existing food random sampling method more
Effectively.
Sound characteristic judges food when brittleness detection module detects food samples crushing by voice detector in the present invention
During brittleness data, in order to avoid the discontinuity of hard threshold function will cause the oscillation of signal, soft-threshold function is too
It is smooth to will cause signal high-frequency information loss；For the oscillation for reducing signal, retains the characteristic spikes point information of original signal, reduce
The distortion of signal, preferably estimation original signal, using a kind of improved wavelet threshold speech de-noising algorithm.
Formaldehyde examination module formaldehyde examination sensor detects formaldehyde in chilled class aquatic products and its dried product and contains in the present invention
During measuring data, in order to avoid the signal of the detection on traditional algorithm is undesirable, the weak phenomenon of anti-interference ability, simultaneously
In order to avoid being showed using end effect present in empirical mode decomposition and modal overlap when being decomposed to the signal of detection
As decomposing joint Denoising Algorithm based on ICA algorithm and EEMD using one kind.
In the present invention main control module to different types of detection data in the process of processing, in order to improve K mean value
The local search ability of algorithm accelerates convergence rate, effectively prevents the generation of precocious phenomenon, realizes preferably convergence effect
Fruit, using improved population K- means clustering algorithm.
Detailed description of the invention
Fig. 1 is food safety sampling Detection device structure block diagram provided in an embodiment of the present invention.
In figure: 1, Pesticides Testing module；2, brittleness detection module；3, nitrate detection module；4, formaldehyde examination module；5,
Main control module；6, extraction module；7, module is crushed；8, safety evaluation module；9, detection data memory module；10, mould is shown
Block.
Fig. 2 is food safety sampling Detection method flow diagram provided in an embodiment of the present invention.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing
Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, food safety sampling Detection equipment provided by the invention includes: Pesticides Testing module 1, brittleness detection
Module 2, formaldehyde examination module 4, main control module 5, extraction module 6, crushes module 7, safety evaluation mould at nitrate detection module 3
Block 8, detection data memory module 9, display module 10.
Pesticides Testing module 1 is connect with main control module 5, for detecting vegetables, melon by food Detecting Pesticide device
Residue of pesticide data in the food such as fruit, grain, tealeaves；
Brittleness detection module 2 is connect with main control module 5, for detecting sound when food samples crush by voice detector
Sound feature judges food brittleness data；(sound).
Nitrate detection module 3 is connect with main control module 5, for detecting cure foods and meat by nitrate detector
Nitrate content data in product；
Formaldehyde examination module 4 is connect with main control module 5, for by formaldehyde detector detect chilled class aquatic products and its
Content of formaldehyde data in dried product；
Main control module 5, with Pesticides Testing module 1, brittleness detection module 2, nitrate detection module 3, formaldehyde examination module
4, extraction module 6, crushing module 7, safety evaluation module 8, detection data memory module 9, display module 10 connect, for passing through
Single-chip microcontroller controls modules and works normally；
Extraction module 6 is connect with main control module 5, for extracting food samples by extractor；
Module 7 is crushed, is connect with main control module 5, for carrying out pulverization process to food samples by pulverizer；
Safety evaluation module 8 is connect with main control module 5, for being pacified according to the data of detection to food by assessment process
It is evaluated entirely；
Detection data memory module 9 is connect with main control module 5, for the pesticide residue, crisp by memory storage detection
Degree, nitrate content, content of formaldehyde data information；
Display module 10 is connect with main control module 5, for pesticide residue, the brittleness, nitric acid by display display detection
Salt content, content of formaldehyde data information and evaluation result.
As shown in Fig. 2, food safety sampling Detection method provided in an embodiment of the present invention, comprising the following steps:
S101: firstly, extractor extracts food samples, pulverization process is carried out to food samples using pulverizer；
S102: the food samples of pulverization process are passed through into relevant operation, detect residue of pesticide concentration；Detect food samples
Sound characteristic judges food brittleness when crushing；Detect nitrate content in cure foods and meat products；Detect chilled class aquatic products
And its content of formaldehyde in dried product；
S103: the data pair detected according to above-mentioned residue of pesticide concentration, food brittleness, nitrate content, content of formaldehyde
Food safety is evaluated；
S104: pass through pesticide residue, the brittleness, nitrate content, content of formaldehyde data information of memory storage detection；It is logical
Cross pesticide residue, brittleness, nitrate content, content of formaldehyde data information and the evaluation result of display display detection.
The brittleness detection module 2 detects sound characteristic when food samples crush by voice detector and judges food brittleness
During data, in order to avoid the discontinuity of hard threshold function will cause the oscillation of signal, soft-threshold function is too smooth
It will cause signal high-frequency information loss；For the oscillation for reducing signal, retains the characteristic spikes point information of original signal, reduce signal
Distortion, preferably estimation original signal specifically include and walk using a kind of improved wavelet threshold speech de-noising algorithm
It is rapid:
Step 1, selects db3 as wavelet basis, carries out 5 layers of wavelet decomposition to the voice signal f (k) that band is made an uproar, obtains one
The wavelet coefficient w of seriesJ, k；
Step 2, using formula
Wherein: y is improved threshold function table；T is to estimate threshold value by soft-threshold function；
Improved threshold function table is to wJ, kThreshold process is carried out, obtains estimation wavelet coefficientMakeTo the greatest extent
It may be small；
Step 3, the estimation wavelet coefficient obtained according to step 2Small echo is reconstructed, the signal after being denoised
The 4 formaldehyde examination sensor of formaldehyde examination module detects content of formaldehyde number in chilled class aquatic products and its dried product
During, in order to avoid the signal of the detection on traditional algorithm is undesirable, the weak phenomenon of anti-interference ability, while in order to
It avoids when being decomposed to the signal of detection, using end effect present in empirical mode decomposition and modal overlap phenomenon,
Joint Denoising Algorithm is decomposed based on ICA algorithm and EEMD using one kind, specifically includes step:
Step 1, is added white noise signal in original signal, the end effect that inhibits EMD decomposable process finally to occur and
Modal overlap phenomenon；
Step 2 carries out EMD decomposition to composite signal, obtains several IMF components, removes first containing main noise
IMF component；
Step 3 carries out FastICA algorithm to remaining IMFs, solves hybrid matrix A, separation matrix W and isolated component
The matrix S of composition；
Step 4, selects the independent source component of predominantly noise source, carries out EMD denoising to it, only after being denoised
Vertical source component S '；
Hybrid matrix A is multiplied with the independent source component S ' after denoising and reconstructs new IMF collection by step 5, the IMF collection phase
Add the signal after denoising can be obtained.
The main control module 5 to different types of detection data in the process of processing, in order to improve K
The local search ability of mean algorithm accelerates convergence rate, effectively prevents the generation of precocious phenomenon, realizes
Preferably convergence effect, using improved population K- means clustering algorithm, comprising the following specific steps
Step 1, setting population m provide maximum the number of iterations T, m initial solution X are randomly generated0；
Step 2, according to current location, with formula
Wherein: ωijIt is the element of S × K weighting matrix, value non-zero i.e. 1, is 1, when being not belonging to when sample i belongs to class j
It is 0；F is sum of the distance of all samples to corresponding cluster centre；CjPIt is poly-
Class center；
Calculate adaptive value f；Current adaptive value is set for individual extreme value Plid, current position is individual extreme value place Pxbest,
According to the individual extreme value P of each particlelid, find out global extremum GpdWith global extremum position Gxbest；
Step 3 starts to iterate to calculate, by formula
Vid(k+1)=ω Vid(k)+C1×φ1×
(Plid-Xid(k))+C2×φ2×(Pgd-Xid(k))；
The speed of more new particle, and it is limited in vmaxIt is interior；
By Formula Vid(k+1)=Xid(k)+Vid(k+1) the current position of more new particle；Wherein
K is the number of iteration；VidFor particle rapidity, XidFor particle position, ω is inertia weight coefficient；C1, C2Referred to as learn
It practises the factor or accelerates constant, φ1, φ2For 2 random normal numbers in (0,1) section, PlidFor individual extreme value, PgdFor global extremum；
The current position of more new particle；
Step 4, according to current location, each sample distributes to K cluster centre by minimal distance principle；
Step 5, by formula
Calculate fitness f；
Step 6 updates PlidAnd Pxbest, go to the individual extreme value P that step 2 calculates each particlelidAnd extreme value place
Pxbest；
Step 7, according to the individual extreme value P of each particlelidFind out global extremum GpdWith global extremum position Gxbest, sentence
It is disconnected whether to reach maximum the number of iterations T；
Step 8 finally exports global extremum GpdWith global extremum position Gxbest。
1 detection method of Pesticides Testing module provided by the invention is as follows:
(1) training sample data are acquired, according to training sample data training sample, obtain remains of pesticide type and content
Curve；
(2) big data resource is concentrated to carry out processing point to the curve data of remains of pesticide type and content by Cloud Server
Analysis；
(3) food samples are obtained, food samples are pulverized；
(4) quantitative water is added, is diluted to food samples are extracted, and be sufficiently stirred, obtains food samples extracting solution；
(5) food samples extracting solution is obtained, farm chemical ingredients and content detection are carried out to food samples extracting solution；
(6) according to the farm chemical ingredients of detection and content, the severity of food residue pesticide is determined.
Safety evaluation module 8 provided by the invention evaluation method is as follows:
1) food samples are extracted by extractor；
2) food sampling is divided into two dimensions, is that ordinary trade food is sampled, cross-border electric business food is sampled respectively, simultaneously
To the two dimensions formulate respectively several first class index, two-level index and three-level index instant food type, weights of importance coefficient,
Risk rated ratio coefficient；
3) according to AHP analytic hierarchy process (AHP), several first class index, two-level index and three-level index of two dimensions are constructed
At the food sampling system model based on risk assessment；
4) according to the composition and Food Inspection regulatory status of China's food business, by convention will according to animal product and
Three kinds of its product, plant product and its product, food traditional food Main Foods types of service, inviting includes that food technology is special
Family, foods supervision expert and third party Quality Examination expert form expert group, and expert group's total number of persons is M, and asks expert
To several first class index of two dimensions, two-level index and three-level index instant food type, important coefficient, risk factor
The marking of solution degree is determining each index to determine several first class index of two dimensions, the weight of two-level index and three-level index
Before weight, the calculating of food expert domain coefficient, calculation formula are first carried out respectively are as follows:Its
In: N is expert's sum；I=expert's serial number, i=1,2,3...N；PiFor technical specialist authority's coefficient, whereinKiFor
Degree of understanding coefficient of the technical specialist to index；SiFor the specialty background coefficient of technical specialist；
5) based food link and the characteristics of quality-monitoring, to the weights of importance coefficient Z of every kind of food typek, risk
Weight coefficient RkTest assessment is carried out, and generates the sampling comprehensive weight coefficient C of food sampling index system in turnk；
6) first class index instant food type is divided into K layers, sets total batch of food sampling as F, is taken out by introducing food
Sample comprehensive weight coefficient Ck, obtain 1-K layers of sampling batch FK, FK=Ck*F；
7) it according to the layering of first class index instant food type, carries out implementing sampling with unequal probability in K layers, enabling K layers is it
In a food type, this food type is divided into E kind unit again；The total scale measurement of K layers of food type is M, no
Size measurement with kind unit distinguishes M1、M2、M3...Me, whereinIntroduce K layers of sampling batch FK, then equidistant to take out
Sample interval are as follows: D=M/FK；Then a number is taken at random between 1-D, it is assumed that be X, then the cell codes section phase where X
The unit answered is the unit drawn, later every D metric, i.e. X+D, X+2D, X+3D ... X+ (FK- 1) numbers such as D
The corresponding units in the cell codes section at place, the unit as drawn；As the Size measurement Me of all different cultivars units
When < D, determine that it is unduplicated sampling；As the Size measurement Me > D of some different cultivars unit, then e-th of list is determined
Member is possible to be repeated and draw；As the Size measurement Me < 2D of different cultivars unit, then determine that e-th of unit is certain to be weighed
It draws again.The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form, all
It is any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belongs to this
In the range of inventive technique scheme.
Claims (9)
1. a kind of food safety sampling Detection method, which is characterized in that the food safety sampling Detection method includes following step
It is rapid:
Extractor extracts food samples, carries out pulverization process to food samples using pulverizer；
The food samples of pulverization process are passed through into relevant operation, detect residue of pesticide concentration；Detect sound when food samples crush
Sound feature judges food brittleness；Detect nitrate content in cure foods and meat products；Detect chilled class aquatic products and its drying
Content of formaldehyde in product；
According to above-mentioned residue of pesticide concentration, food brittleness, nitrate content, content of formaldehyde detection data it is food-safe into
Row evaluation；
Pass through pesticide residue, the brittleness, nitrate content, content of formaldehyde data information of memory storage detection；Pass through display
Show pesticide residue, brittleness, nitrate content, content of formaldehyde data information and the evaluation result of detection.
2. food safety sampling Detection method as described in claim 1, which is characterized in that detect food by voice detector
During sound characteristic judges food brittleness data when sample comminution, using improved wavelet threshold speech de-noising algorithm, tool
Body includes:
Step 1, selects db3 as wavelet basis, carries out 5 layers of wavelet decomposition to the voice signal f (k) that band is made an uproar, obtains a series of
Wavelet coefficient wJ, k '；
Step 2, using formula
Wherein: y is improved threshold function table；T is to estimate threshold value by soft-threshold function；
Improved threshold function table is to wJ, kThreshold process is carried out, obtains estimation wavelet coefficientMakeAs far as possible
It is small；
Step 3, the estimation wavelet coefficient obtained according to step 2Small echo is reconstructed, the signal after being denoised
3. food safety sampling Detection method as described in claim 1, which is characterized in that the detection of formaldehyde examination sensor is chilled
In class aquatic products and its dried product during content of formaldehyde data, calculated using joint denoising is decomposed based on ICA algorithm and EEMD
Method specifically includes:
White noise signal is added in step 1 in original signal, the end effect and mode for inhibiting EMD decomposable process finally to occur
Aliasing；
Step 2 carries out EMD decomposition to composite signal, obtains several IMF components, removes first IMF containing main noise
Component；
Step 3 carries out FastICA algorithm to remaining IMFs, solves hybrid matrix A, separation matrix W and isolated component composition
Matrix S；
Step 4, selects the independent source component of predominantly noise source, carries out EMD denoising to it, the independent source after being denoised
Component S '；
Hybrid matrix A is multiplied with the independent source component S ' after denoising and reconstructs new IMF collection by step 5, which is added
Signal after being denoised.
4. food safety sampling Detection method as described in claim 1, which is characterized in that Pesticides Testing module detection side
Method includes:
(1) acquisition training sample data obtain the song of remains of pesticide type and content according to training sample data training sample
Line；
(2) big data resource is concentrated to carry out processing analysis to the curve data of remains of pesticide type and content by Cloud Server；
(3) food samples are obtained, food samples are pulverized；
(4) quantitative water is added, is diluted to food samples are extracted, and be sufficiently stirred, obtains food samples extracting solution；
(5) food samples extracting solution is obtained, farm chemical ingredients and content detection are carried out to food samples extracting solution；
(6) according to the farm chemical ingredients of detection and content, the severity of food residue pesticide is determined.
5. food safety sampling Detection method as described in claim 1, which is characterized in that safety evaluation module evaluation side
Method includes:
1) food samples are extracted by extractor；
2) food sampling is divided into two dimensions, is the sampling of ordinary trade food, the sampling of cross-border electric business food respectively, while to this
Two dimensions formulate several first class index, two-level index and three-level index instant food type, weights of importance coefficient, risk respectively
Weight coefficient；
3) according to AHP analytic hierarchy process (AHP), several first class index, two-level index and three-level index of two dimensions are configured to base
In the food sampling system model of risk assessment；
It 4), by convention will be according to animal product and its system according to the composition and Food Inspection regulatory status of China's food business
Three kinds of product, plant product and its product, food traditional food Main Foods types of service, inviting includes food technology expert, food
Product administrative experts and third party Quality Examination expert form expert group, and expert group's total number of persons is M, and asks expert to two
Several first class index, two-level index and three-level index instant food type, the understanding journey of important coefficient, risk factor of a dimension
Degree marking is determining each index weights to determine several first class index of two dimensions, the weight of two-level index and three-level index
Before, the calculating of food expert domain coefficient, calculation formula are first carried out respectively are as follows:Wherein: N
For expert's sum；I=expert's serial number, i=1,2,3...N；PiFor technical specialist authority's coefficient, whereinKiFor skill
Degree of understanding coefficient of the art expert to index；SiFor the specialty background coefficient of technical specialist；
5) based food link and the characteristics of quality-monitoring, to the weights of importance coefficient Z of every kind of food typek, Risk rated ratio system
Number RkTest assessment is carried out, and generates the sampling comprehensive weight coefficient C of food sampling index system in turnk；
6) first class index instant food type is divided into K layers, sets total batch of food sampling as F, is sampled by introducing food comprehensive
Close weight coefficient Ck, obtain 1-K layers of sampling batch FK, FK=Ck*F；
7) it according to the layering of first class index instant food type, carries out implementing samplings with unequal probability in K layer, enabling K layers is wherein one
A food type, this food type are divided into E kind unit again；The total scale measurement of K layers of food type is M, different product
The Size measurement of kind unit distinguishes M1、M2、M3...Me, whereinIntroduce K layers of sampling batch FK, then equidistant sampling
Interval are as follows: D=M/FK；Then a number is taken at random between 1-D, it is assumed that be X, then the cell codes section where X is corresponding
Unit be the unit drawn, later every D metric, i.e. X+D, X+2D, X+3D ... X+ (FK- 1) the digital institute such as D
Cell codes section corresponding units, the unit as drawn；As the Size measurement Me < D of all different cultivars units
When, determine that it is unduplicated sampling；As the Size measurement Me > D of some different cultivars unit, then determine that e-th of unit has
It may be repeated and draw；As the Size measurement Me < 2D of different cultivars unit, then determine that e-th of unit is certain to be repeated pumping
In.
6. a kind of computer program for realizing food safety sampling Detection method described in Claims 1 to 5 any one.
7. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed
Benefit requires food safety sampling Detection method described in 1-5 any one.
8. a kind of food safety sampling Detection equipment, which is characterized in that the food safety sampling Detection equipment includes:
Pesticides Testing module, connect with main control module, for by food Detecting Pesticide device detect vegetables, melon and fruit, grain,
Residue of pesticide data in the food such as tealeaves；
Brittleness detection module, connect with main control module, for detecting sound characteristic when food samples crush by voice detector
Judge food brittleness data；
Nitrate detection module, connect with main control module, for being detected in cure foods and meat products by nitrate detector
Nitrate content data；
Formaldehyde examination module, connect with main control module, for detecting chilled class aquatic products and its dried product by formaldehyde detector
Middle content of formaldehyde data；
Main control module, with Pesticides Testing module, brittleness detection module, nitrate detection module, formaldehyde examination module, extraction mould
Block crushes module, safety evaluation module, detection data memory module, display module connection, each for being controlled by single-chip microcontroller
Module works normally；
Extraction module is connect with main control module, for extracting food samples by extractor；
Module is crushed, is connect with main control module, for carrying out pulverization process to food samples by pulverizer；
Safety evaluation module, connect with main control module, for the food-safe progress of data by assessment process according to detection
Evaluation；
Detection data memory module, connect with main control module, for pesticide residue, the brittleness, nitre by memory storage detection
Phosphate content, content of formaldehyde data information；
Display module is connect with main control module, for by display display detection pesticide residue, brittleness, nitrate content,
Content of formaldehyde data information and evaluation result.
9. food safety sampling Detection equipment Internet of Things food safety detection platform described in a kind of carrying claim 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910114422.4A CN109856344A (en) | 2019-02-14 | 2019-02-14 | A kind of food safety sampling Detection equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910114422.4A CN109856344A (en) | 2019-02-14 | 2019-02-14 | A kind of food safety sampling Detection equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109856344A true CN109856344A (en) | 2019-06-07 |
Family
ID=66897747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910114422.4A Pending CN109856344A (en) | 2019-02-14 | 2019-02-14 | A kind of food safety sampling Detection equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109856344A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113552298A (en) * | 2021-08-23 | 2021-10-26 | 湖南山水检测有限公司 | Method for detecting antibiotic residual quantity in animal-derived food |
| CN114778874A (en) * | 2022-06-16 | 2022-07-22 | 深圳市汇知科技有限公司 | Food safety detection analysis method and system based on Internet of things |
| CN115684507A (en) * | 2022-10-26 | 2023-02-03 | 深圳市赛泰诺生物技术有限公司 | Intelligent AI system and intelligent AI all-in-one machine for pesticide and heavy metal rapid detection |
| CN116067964A (en) * | 2023-03-06 | 2023-05-05 | 广东省农业科学院动物科学研究所 | Method and system for promoting fish muscle embrittlement by utilizing condensed tannin |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7807344B2 (en) * | 2006-10-26 | 2010-10-05 | Global Grain Security, Llc | Methods and systems for ensuring the security of grain stores |
| CN103902844A (en) * | 2014-04-24 | 2014-07-02 | 国家电网公司 | Transformer vibration signal de-noising method based on EEMD kurtosis threshold value |
| CN105301099A (en) * | 2015-10-15 | 2016-02-03 | 吉林大学 | Food crispness detection method |
| CN105450636A (en) * | 2015-11-06 | 2016-03-30 | 长春智信创联科技有限公司 | Cloud computing management system and management method of cloud computing management system |
| CN106094764A (en) * | 2016-08-01 | 2016-11-09 | 南京腾图节能科技有限公司 | A kind of industrial circulating cooling water system based on cloud computing monitoring system |
| CN106203827A (en) * | 2016-07-05 | 2016-12-07 | 杨林 | A kind of food safety risk early warning system |
| CN106485932A (en) * | 2016-12-28 | 2017-03-08 | 芜湖乐锐思信息咨询有限公司 | Highly effective path management system |
| CN106895906A (en) * | 2017-03-23 | 2017-06-27 | 西安理工大学 | A kind of feature extracting method of vibration of hydrogenerator set failure |
| CN107274908A (en) * | 2017-06-13 | 2017-10-20 | 南京邮电大学 | Small echo speech de-noising method based on new threshold function table |
| CN107679779A (en) * | 2017-11-21 | 2018-02-09 | 深圳市检验检疫科学研究院 | A kind of imported food sampling safe evaluation method |
| CN107860747A (en) * | 2017-12-19 | 2018-03-30 | 凌工二号（大连）科技有限公司 | A kind of detection method of food residue agricultural chemicals |
| CN108182555A (en) * | 2018-02-09 | 2018-06-19 | 重庆市农业科学院 | A kind of Safety of Food Quality risk evaluating system |
| CN108627621A (en) * | 2018-04-25 | 2018-10-09 | 遵义市产品质量检验检测院 | A kind of intelligence food quality rapid detection system |
-
2019
- 2019-02-14 CN CN201910114422.4A patent/CN109856344A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7807344B2 (en) * | 2006-10-26 | 2010-10-05 | Global Grain Security, Llc | Methods and systems for ensuring the security of grain stores |
| CN103902844A (en) * | 2014-04-24 | 2014-07-02 | 国家电网公司 | Transformer vibration signal de-noising method based on EEMD kurtosis threshold value |
| CN105301099A (en) * | 2015-10-15 | 2016-02-03 | 吉林大学 | Food crispness detection method |
| CN105450636A (en) * | 2015-11-06 | 2016-03-30 | 长春智信创联科技有限公司 | Cloud computing management system and management method of cloud computing management system |
| CN106203827A (en) * | 2016-07-05 | 2016-12-07 | 杨林 | A kind of food safety risk early warning system |
| CN106094764A (en) * | 2016-08-01 | 2016-11-09 | 南京腾图节能科技有限公司 | A kind of industrial circulating cooling water system based on cloud computing monitoring system |
| CN106485932A (en) * | 2016-12-28 | 2017-03-08 | 芜湖乐锐思信息咨询有限公司 | Highly effective path management system |
| CN106895906A (en) * | 2017-03-23 | 2017-06-27 | 西安理工大学 | A kind of feature extracting method of vibration of hydrogenerator set failure |
| CN107274908A (en) * | 2017-06-13 | 2017-10-20 | 南京邮电大学 | Small echo speech de-noising method based on new threshold function table |
| CN107679779A (en) * | 2017-11-21 | 2018-02-09 | 深圳市检验检疫科学研究院 | A kind of imported food sampling safe evaluation method |
| CN107860747A (en) * | 2017-12-19 | 2018-03-30 | 凌工二号（大连）科技有限公司 | A kind of detection method of food residue agricultural chemicals |
| CN108182555A (en) * | 2018-02-09 | 2018-06-19 | 重庆市农业科学院 | A kind of Safety of Food Quality risk evaluating system |
| CN108627621A (en) * | 2018-04-25 | 2018-10-09 | 遵义市产品质量检验检测院 | A kind of intelligence food quality rapid detection system |
Non-Patent Citations (1)
| Title |
|---|
| 姜邵飞: "基于EEMD与FastICA的损伤异常识别与定位", 《振动与冲击》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113552298A (en) * | 2021-08-23 | 2021-10-26 | 湖南山水检测有限公司 | Method for detecting antibiotic residual quantity in animal-derived food |
| CN114778874A (en) * | 2022-06-16 | 2022-07-22 | 深圳市汇知科技有限公司 | Food safety detection analysis method and system based on Internet of things |
| CN115684507A (en) * | 2022-10-26 | 2023-02-03 | 深圳市赛泰诺生物技术有限公司 | Intelligent AI system and intelligent AI all-in-one machine for pesticide and heavy metal rapid detection |
| CN116067964A (en) * | 2023-03-06 | 2023-05-05 | 广东省农业科学院动物科学研究所 | Method and system for promoting fish muscle embrittlement by utilizing condensed tannin |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109856344A (en) | A kind of food safety sampling Detection equipment | |
| Ye et al. | Increasing zooplankton size diversity enhances the strength of top‐down control on phytoplankton through diet niche partitioning | |
| Bartram | Experimental development of a model for the feeding of neritic copepods on phytoplankton | |
| Sliwinska et al. | Food analysis using artificial senses | |
| Masood et al. | Natural occurrence of aflatoxins in dry fruits and edible nuts | |
| CN107909109B (en) | SAR image classification method based on conspicuousness and multiple dimensioned depth network model | |
| Tuanmu et al. | Mapping understory vegetation using phenological characteristics derived from remotely sensed data | |
| CN106093321B (en) | A kind of market for farm products monitoring system and method | |
| CN106568907A (en) | Chinese mitten crab freshness damage-free detection method based on semi-supervised identification projection | |
| Davis et al. | A dietary-wide association study (DWAS) of environmental metal exposure in US children and adults | |
| CN102435713A (en) | Automatic detection system for quality of traditional Chinese medicine | |
| CN102819748B (en) | Classification and identification method and classification and identification device of sparse representations of destructive insects | |
| CN112147601B (en) | Sea surface small target detection method based on random forest | |
| CN112578725A (en) | Fishery water aquaculture facility safety early warning system and method based on Internet of things | |
| CN110501122A (en) | A kind of adaptive leakage detection method of pressure vessel based on cluster | |
| CN107330457A (en) | A kind of Classification of Polarimetric SAR Image method based on multi-feature fusion | |
| Ye et al. | NIR hyperspectral imaging technology combined with multivariate methods to identify shrimp freshness | |
| CN107679779A (en) | A kind of imported food sampling safe evaluation method | |
| Pipoyan et al. | Risk Characterization of the Armenian Population to Nickel: Application of Deterministic and Probabilistic Approaches to a Total Diet Study in Yerevan City | |
| Mittelbach et al. | What is the observed relationship between species richness and productivity? Reply | |
| Buer et al. | Nitrogen and phosphorous content in blue mussels (Mytilus spp.) across the Baltic Sea | |
| CN105738422A (en) | Method for quickly detecting storage time of walnuts based on electronic nose | |
| CN111982905B (en) | Wood quality intelligent detection system based on industrial big data image analysis | |
| CN106294838A (en) | The method of the big data structured of food safety poisonous and harmful substance inspection | |
| CN101814146A (en) | Method for extracting action potential feature based on multiple wavelet transformation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |