CN110378393A - A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM - Google Patents
A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM Download PDFInfo
- Publication number
- CN110378393A CN110378393A CN201910558320.1A CN201910558320A CN110378393A CN 110378393 A CN110378393 A CN 110378393A CN 201910558320 A CN201910558320 A CN 201910558320A CN 110378393 A CN110378393 A CN 110378393A
- Authority
- CN
- China
- Prior art keywords
- value
- particle
- gsa
- pso
- svm
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Quality & Reliability (AREA)
- Automation & Control Theory (AREA)
- Computational Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The mixing printing product acetes chinensis method based on PSO-GSA-SVM that the invention discloses a kind of, belongs to machine learning techniques field.In industrial production, acetes chinensis is a critically important technology, and application field is also than wide.The present invention first wants the image of collection site determinand;Field samples and determinand acquisition image are pre-processed;Extract the color characteristic Value Data rgb value of live determinand acquisition image RGB color;The L*a*b* value of CIELAB color space is calculated using PSO-GSA-SVM optimization algorithm model;The value for importing the L*a*b* of sample image carries out Colorimetry using CMC (l, c) colour difference formula;Finally export acetes chinensis result.The present invention can realize accurately acetes chinensis effectively by depth learning technology independently training study.
Description
Technical field
The present invention relates to a kind of based on depth learning technology to the field of acetes chinensis, and in particular to one kind is based on PSO-
The mixing printing product acetes chinensis method of GSA-SVM.
Technical background
In the industrial production, an important feature of the color as product itself, can enrich the appearance of product itself.Print
The quality of brush product dyeing quality is also to evaluate a key factor of textile product quality.Therefore, accurate intelligence easily printing
The acetes chinensis method of product has great importance in entire dyeing.
Scientific and technological level and people's substance at present increasingly pursue intelligent, the mixing printing product color based on PSO-GSA-SVM
Difference detection is exactly to use depth learning technology, instead of traditional artificial treatment, more intelligently, is eliminated by observer's color vision
The color evaluation that view condition variation generates is inconsistent with seeing, and overcomes the work difficulty under the severe production environment of factory, realizes life
People from provinceization's operation in producing line, reduces the human cost of enterprise.The present invention is of great significance to and practical value.
Summary of the invention
1, the purpose invented
The present invention is existing to plain color cloth acetes chinensis low efficiency in current industry in order to solve, and the degree of automation is not high
Situation, a kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM proposed by the present invention incorporated machine vision
And depth learning technology, the hybrid optimization algorithm of PSO-GSA-SVM combine the fuzzy rule inference ability and mind of fuzzy system
Self study and adaptive ability through network can be retouched preferably compared to individual nerve network system or individual fuzzy system
System object is stated, can be more intelligent, greatly improve detection efficiency
2, technical solution used by inventing
The mixing printing product acetes chinensis method based on PSO-GSA-SVM that the invention discloses a kind of, comprising the following steps:
The image of step 1, collecting sample and determinand;
Step 2 pre-processes sample and determinand acquisition image;
Step 3 extracts the RGB color characteristic value data for acquiring the RGB color of object image to be measured;
Step 4, the L*a*b* value that CIELAB color space is calculated using PSO-GSA-SVM optimization algorithm model;
Step 5 carries out Colorimetry using CMC (l, c) colour difference formula;
Step 6, output acetes chinensis result.
Further, sample and determinand acquisition image are pre-processed described in step 2, specific steps method is such as
Under:
Step 2.1, the conversion that sample image is carried out to color space
The space CIELAB being transformed into from RGB to treated image, steps are as follows:
The wherein tristimulus values X in this system methodn、Yn、Zn, working flare and the elevation angle can set selection, if X/Xn、
Y/Yn、Z/ZnValue be greater thanThen
Step 2.2, by behind converting colors space sample and determinand acquisition image carry out gaussian filtering noise reduction process;It is logical
Cross formula (1)Wherein σ indicates Gaussian Distribution Parameters, and σ value eventually determines the width of filter, σ value with
Being positively correlated property of smooth Chengdu.
Further, acquisition described in step 3 calculates the RGB color of sample image L*a*b* value and object image to be measured
RGB color characteristic value data.
Further, the L* of CIELAB color space is calculated described in step 4 using PSO-GSA-SVM optimization algorithm model
A*b* value:
Data set is divided into training set and test set by ten folding cross validations by step 4.1, i.e., by 90% data set
As training data and 10% data set as test data;
Step 4.2, the input parameter for initializing PSO-GSA particle swarm algorithm, including population scale, weight factor are maximum
The number of iterations, the upper bound of search space, the lower bound of search space and primary coding;
Step 4.3 calculates fitness value, sets current location for the individual wisdom of each particle, is adapted to using utilizing
Degree function calculates the fitness value of each particle, and taking fitness, good particle does corresponding individual extreme value as initial global pole
Value;
Step 4.4 is iterated calculating, the position of more new particle and speed according to the position and speed more new formula of particle
Degree;
Step 4.5, the fitness value that each particle after each iteration is calculated according to the fitness function of particle;
Step 4.6 makes comparisons the fitness value of each particle with the fitness value of its individual extreme value, if more preferably
Words, then more new individual extreme value, otherwise retains initial value;
Step 4.7, the global optimum that particle position and fitness are calculated using GSA optimization algorithm
Step 4.8 judges whether to meet termination condition, if reaching maximum number of iterations or the convergence of gained solution or gained
Solution has had reached optimal SVR parameter combination (C, γ), just terminates iteration, otherwise return step 4.3;
Step 4.9, the optimal SVR parameter combination (C, γ) of output, import SVR model and complete model construction;
Step 4.10 imports test set parameter
Step 4.11 is predicted using the SVR model for the optimized parameter model built;
Step 4.12, performance evaluation.
Further, color difference is calculated using colour difference formula, selects CMC(l:c)Colour difference formula carries out calculating value of chromatism.
Further, the step 4.4, according to the position and speed more new formula of particle be iterated calculating, update
The position and speed of particle;It is that individual increases memory and society's letter in gravitation search algorithm in conjunction with the characteristics of particle swarm algorithm
Exchange capacity is ceased, proposes to improve gravitation search and Particle Swarm Optimization-based Hybrid Optimization Algorithm, discrete particles' motion equation are as follows:
Vij(t+1)=WVij(t)+c1r1(p-Xij(t))+c2r2(g-Xij(t))+aij(t)
In formula, r1, r2It is the random number of distribution for [0,1], by adjusting c1, c2Value, adjustable Particles Moving mistake
The influence degree exchanged in journey by gravitation rule, memory and social information.
Further, the fitness value of the fitness value of each particle and its individual extreme value the step 4.6: is made into ratio
Compared with if more preferably talked about, otherwise more new individual extreme value retains initial value;Notice that w is the inertia weight factor, what is embodied is
Particle inherits the ability of previous speed, improves inertia weight calculating formula:
In formula, WsFor initial inertia weight;WeInertia weight when for iteration to maximum times;K is current iteration algebra;
TmaxFor greatest iteration algebra, w is slower in kinematic nonlinearity variation w early period variation, and value is larger, and the overall situation for maintaining algorithm is searched
Suo Nengli.
3, the invention has the advantages that:
The present invention has merged machine vision and depth learning technology to a variety of acetes chinensis, and the scope of application is wider, automatically
Change degree is relatively high, can satisfy the testing requirements of current most printed matters.The hybrid optimization algorithm of PSO-GSA-SVM
The self study and adaptive ability of the fuzzy rule inference ability and neural network of fuzzy system are combined, compared to individually nerve
Network system or individual fuzzy system can better describe system object.Algorithm of the invention and current traditional color difference are examined
Survey method is fast with more speed, and precision is good, greatly improves efficiency.
Detailed description of the invention:
Fig. 1 is flow chart of the invention;
Fig. 2 is the hybrid optimization algorithm flow chart of PSO-GSA-SVM;
Fig. 3 is the L*a*b* value flow chart of the reckoning CIELAB color space of PSO-GSA-SVM;
Fig. 4 is Support vector regression data prediction schematic diagram.
Fig. 5 is 80 groups of test data set distributions to be entered.
Fig. 6 is 80 groups of prediction result Distribution values.
The value and actual comparison relationship that Fig. 7 is the color component L* after prediction.
The value and actual comparison relationship that Fig. 8 is the color component a* after prediction.
The value and actual comparison relationship that Fig. 9 is the color component b* after prediction.
Specific embodiment
In order to better understand the present invention, in conjunction with the embodiments, the present invention is further described for attached drawing.
Embodiment:
As shown in Fig. 1 flow chart of the invention, this method are realized by the following method:
A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM comprising following steps:
1, the image of collection site sample and determinand;
Image Acquisition: shooting, collecting is carried out to dyeing under standard sources by array CCD camera, and is divided into training set
And test set;
2, field samples and determinand acquisition image are pre-processed;
Color space conversion: sample collection image is carried out to the conversion of color space.
By the space CIELAB is transformed into from RGB to the image of acquisition, steps are as follows:
The wherein tristimulus values X in this system methodn、Yn、ZnHave six kinds of modes according to working flare and the elevation angle it can be selected that
Parameter list is as follows
If X/Xn、Y/Yn、Z/ZnValue be greater than
Noise reduction process: sample and determinand acquisition image are subjected to gaussian filtering noise reduction process.
Pass through formula (1)Wherein σ indicates Gaussian Distribution Parameters, and σ value eventually determines filter
Width, σ value and smooth being positively correlated property of Chengdu.
3, the color characteristic Value Data of field samples and determinand acquisition image is extracted;
The Δ H, Δ S, Δ V of the color difference Δ E and characterization color difference that are calculated, Δ L Δ a, Δ b eigenvalue cluster is at one group of color difference
Data set;
4, the L*a*b* value of CIELAB color space is calculated using PSO-GSA-SVM optimization algorithm model;
Data set is divided into training set and test set by ten folding cross validations by step 4.1, i.e., by 90% data set
As training data and 10% data set as test data;
Step 4.2, the input parameter for initializing PSO-GSA particle swarm algorithm, including population scale, weight factor are maximum
The number of iterations, the upper bound of search space, the lower bound of search space and primary coding;
Step 4.3 calculates fitness value, sets current location for the individual wisdom of each particle, is adapted to using utilizing
Degree function calculates the fitness value of each particle, and taking fitness, good particle does corresponding individual extreme value as initial global pole
Value;
Step 4.4 is iterated calculating, the position of more new particle and speed according to the position and speed more new formula of particle
Degree;In conjunction with the characteristics of particle swarm algorithm, it is that individual increases memory and social information's exchange capacity in gravitation search algorithm, proposes
Improve gravitation search and Particle Swarm Optimization-based Hybrid Optimization Algorithm, discrete particles' motion equation are as follows:
Vij(t+1)=WVij(t)+c1r1(p-Xij(t))+c2r2(g-Xij(t))+aij(t)
In formula, r1, r2It is the random number of distribution for [0,1], by adjusting c1, c2Value, adjustable Particles Moving mistake
The influence degree exchanged in journey by gravitation rule, memory and social information.
Step 4.5, the fitness value that each particle after each iteration is calculated according to the fitness function of particle;
Step 4.6 makes comparisons the fitness value of each particle with the fitness value of its individual extreme value, if more preferably
Words, then more new individual extreme value, otherwise retains initial value;Notice that w is the inertia weight factor, what is embodied is that particle is inherited previously
The ability of speed, in order to improve the search capability of algorithm, set forth herein a kind of improved inertia weight calculating formulas:
In formula, WsFor initial inertia weight;WeInertia weight when for iteration to maximum times;K is current iteration algebra;
TmaxFor greatest iteration algebra.W is slower in kinematic nonlinearity variation w early period variation, and value is larger, and the overall situation for maintaining algorithm is searched
Suo Nengli;The variation of later period w is very fast, greatly improves the local optimal searching ability of algorithm, solves effect well to obtain.
Step 4.7, the global optimum that particle position and fitness are calculated using GSA optimization algorithm
Step 4.8 judges whether to meet termination condition, if reaching maximum number of iterations or the convergence of gained solution or gained
Solution has had reached optimal SVR parameter combination (C, γ), just terminates iteration, otherwise return step 4.3;
Step 4.9, the optimal SVR parameter combination (C, γ) of output, import SVR model and complete model construction;
Support vector regression (SVR) can be described as the problem of data are predicted: system generates a hyperplane not
It is disconnected to adjust this plane, until the point that data concentrate on being belonging respectively to two classes is located at the two sides of hyperplane, and make sample set
In point to classification plane distance it is big as far as possible, as shown in Figure 4
Point on H1, H2 is supporting vector, and H is optimal hyperlane, and edge margin is the vertical range of H1 and H2.
Open circles and square respectively represent a kind of sample, and hyperplane H is the plane that two class samples are correctly separated.
H1, H2 are respectively the straight line for passing through point nearest from Optimal Separating Hyperplane in Different categories of samples and being parallel to classifying face.Optimal Separating Hyperplane H
Not only two class samples are properly separated, and will most the distance between ambassador H1 and H2.Here is for support vector machines
Description:
In sample space, dividing hyperplane can be used following classification linear equation to describe: ωTX+b=0, wherein
ω=(ω1;ω2;...;ωd) it is normal vector, it decides the direction of hyperplane;B is displacement item, decides hyperplane and original
The distance between point.
We are denoted as (ω, b) below, and the distance of arbitrary point x to hyperplane (ω, b) can be written as in sample space
Assuming that hyperplane can correctly classify training sample, i.e., for (xi, yi) ∈ D, if yi=+1, then there is ωTx+b
> 0;If yi, then there is ω in=- 1TX+b < 0;It enables
These training samples nearest apart from hyperplane can make formulaEqual sign at
Vertical, they are referred to as supporting vector, and the sum of the distance of two foreign peoples's supporting vectors to hyperplane can indicate are as follows:
Referred to as it is spaced.
In order to find the division hyperplane with largest interval, that is, to find and be able to satisfy formulaIn the parameter ω and b of constraint meet so that M is maximumSo that
yi(ωTX+b) >=1, i=1 ..., m, in order to maximize interval, it is only necessary to maximize | | ω | |-1, this is equivalent to minimize | | ω |
|2Then formulaIt can be re-written as
For the support vector machines under linearly inseparable situation, soft edges conceptual description can use are as follows:
Wherein C is the preset balance parameters for being used to balance empiric risk and structure risk.
5, color difference is calculated using colour difference formula select CMC(l:c), colour difference formula carries out calculating value of chromatism.
Export the result of Colorimetry.
The present invention can also be there are many case study on implementation, under the premise of without departing substantially from design spirit of the present invention, the common skill in this field
The various changes and improvements that art personnel make technical solution of the present invention should belong to claim belonging to the present invention and guarantor
Protect range.Such as the emulation schematic diagram that Fig. 5-9 is actual value and predicted value, it can be seen that coincide substantially.
The mixing printing product acetes chinensis method based on PSO-GSA-SVM that the present invention provides a kind of, has merged machine view
Feel and depth learning technology is to a variety of acetes chinensis, the scope of application is wider, and the degree of automation is relatively high, can satisfy current
The testing requirements of most printed matters.The fuzzy rule that the hybrid optimization algorithm of PSO-GSA-SVM combines fuzzy system pushes away
Individual nerve network system or individual fuzzy system energy are compared in the self study and adaptive ability of reason ability and neural network
Enough better describe system object.Algorithm of the invention and current traditional acetes chinensis method fast, the precision with more speed
It is good, greatly improve efficiency.
Claims (7)
1. a kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM, it is characterised in that the following steps are included:
The image of step 1, collecting sample and determinand;
Step 2 pre-processes sample and determinand acquisition image;
Step 3 extracts the RGB color characteristic value data for acquiring the RGB color of object image to be measured;
Step 4, the L*a*b* value that CIELAB color space is calculated using PSO-GSA-SVM optimization algorithm model;
Step 5 carries out Colorimetry using CMC (l, c) colour difference formula;
Step 6, output acetes chinensis result.
2. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that step
Sample and determinand acquisition image are pre-processed described in rapid 2, specific steps method is as follows:
Step 2.1, the conversion that sample image is carried out to color space
The space CIELAB being transformed into from RGB to treated image, steps are as follows:
The wherein tristimulus values X in this system methodn、Yn、Zn, working flare and the elevation angle can set selection, if X/Xn、Y/Yn、
Z/ZnValue be greater thanThen
Step 2.2, by behind converting colors space sample and determinand acquisition image carry out gaussian filtering noise reduction process;Pass through public affairs
Formula (1)Wherein σ indicates Gaussian Distribution Parameters, and σ value eventually determines the width of filter, σ value and smooth
Being positively correlated property of Chengdu.
3. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that step
Acquisition described in rapid 3 calculates the RGB color characteristic value number of the RGB color of sample image L*a*b* value and object image to be measured
According to.
4. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that step
The L*a*b* value of CIELAB color space is calculated described in rapid 4 using PSO-GSA-SVM optimization algorithm model:
Data set is divided into training set and test set by ten folding cross validations by step 4.1, i.e., using 90% data set as
Training data and 10% data set as test data;
Step 4.2, the input parameter for initializing PSO-GSA particle swarm algorithm, including population scale, weight factor, greatest iteration
Number, the upper bound of search space, the lower bound of search space and primary coding;
Step 4.3 calculates fitness value, sets current location for the individual wisdom of each particle, utilizes utilization fitness letter
Number calculates the fitness value of each particle, and taking fitness, good particle does corresponding individual extreme value as initial global extremum;
Step 4.4 is iterated calculating, the position and speed of more new particle according to the position and speed more new formula of particle;
Step 4.5, the fitness value that each particle after each iteration is calculated according to the fitness function of particle;
Step 4.6 makes comparisons the fitness value of each particle with the fitness value of its individual extreme value, if more preferably talked about,
More new individual extreme value, otherwise retains initial value;
Step 4.7, the global optimum that particle position and fitness are calculated using GSA optimization algorithm
Step 4.8 judges whether to meet termination condition, if reaching maximum number of iterations or the convergence of gained solution or gained solution
Through having reached optimal SVR parameter combination (C, γ), iteration is just terminated, otherwise return step 4.3;
Step 4.9, the optimal SVR parameter combination (C, γ) of output, import SVR model and complete model construction;
Step 4.10 imports test set parameter
Step 4.11 is predicted using the SVR model for the optimized parameter model built;
Step 4.12, performance evaluation.
5. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that:
Color difference is calculated using colour difference formula, selects CMC(l:c)Colour difference formula carries out calculating value of chromatism.
6. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that institute
The step 4.4 stated is iterated calculating, the position and speed of more new particle according to the position and speed more new formula of particle;Knot
The characteristics of closing particle swarm algorithm is that individual increases memory and social information's exchange capacity in gravitation search algorithm, proposes to improve
Gravitation search and Particle Swarm Optimization-based Hybrid Optimization Algorithm, discrete particles' motion equation are as follows:
In formula, r1, r2It is the random number of distribution for [0,1], by adjusting c1, c2Value, during adjustable Particles Moving
The influence degree exchanged by gravitation rule, memory and social information.
7. the mixing printing product acetes chinensis method according to claim 1 based on PSO-GSA-SVM, it is characterised in that institute
The step 4.6 stated: the fitness value of each particle is made comparisons with the fitness value of its individual extreme value, if more preferably talked about,
More new individual extreme value, otherwise retains initial value;Notice that w is the inertia weight factor, what is embodied is that particle inherits previous speed
Ability, improve inertia weight calculating formula:
In formula, WsFor initial inertia weight;WeInertia weight when for iteration to maximum times;K is current iteration algebra;TmaxFor
Greatest iteration algebra, w is slower in kinematic nonlinearity variation w early period variation, and value is larger, maintains the global search energy of algorithm
Power.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910558320.1A CN110378393A (en) | 2019-06-26 | 2019-06-26 | A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910558320.1A CN110378393A (en) | 2019-06-26 | 2019-06-26 | A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110378393A true CN110378393A (en) | 2019-10-25 |
Family
ID=68249375
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910558320.1A Pending CN110378393A (en) | 2019-06-26 | 2019-06-26 | A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110378393A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111751294A (en) * | 2020-06-30 | 2020-10-09 | 深兰科技(达州)有限公司 | Cloth defect detection method based on learning and memory |
CN113469285A (en) * | 2021-07-26 | 2021-10-01 | 上海大学 | Dye double-spelling formula prediction method based on PSO-LSSVM |
CN113536687A (en) * | 2021-07-26 | 2021-10-22 | 上海大学 | Construction method of monochrome cotton fabric color and dye concentration prediction model based on PSO-LSSVM |
CN113670443A (en) * | 2021-07-09 | 2021-11-19 | 北京中科慧眼科技有限公司 | Color difference measuring method and system based on device-independent color space and intelligent terminal |
CN113781476A (en) * | 2021-10-27 | 2021-12-10 | 南通博纳纺织品有限公司 | Textile dyeing quality evaluation method and system based on image processing |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102170516A (en) * | 2011-04-07 | 2011-08-31 | 陕西科技大学 | Color space transition method based on fuzzy logic and neural network |
CN109141640A (en) * | 2018-08-03 | 2019-01-04 | 深圳市销邦科技股份有限公司 | Acetes chinensis method, system, equipment and storage medium based on machine vision |
-
2019
- 2019-06-26 CN CN201910558320.1A patent/CN110378393A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102170516A (en) * | 2011-04-07 | 2011-08-31 | 陕西科技大学 | Color space transition method based on fuzzy logic and neural network |
CN109141640A (en) * | 2018-08-03 | 2019-01-04 | 深圳市销邦科技股份有限公司 | Acetes chinensis method, system, equipment and storage medium based on machine vision |
Non-Patent Citations (1)
Title |
---|
HONGXIN XUE等: "A Novel Hybrid Model Based on TVIW-PSO-GSA Algorithm and Support Vector Machine for Classification Problems", 《IEEE ACCESS》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111751294A (en) * | 2020-06-30 | 2020-10-09 | 深兰科技(达州)有限公司 | Cloth defect detection method based on learning and memory |
CN113670443A (en) * | 2021-07-09 | 2021-11-19 | 北京中科慧眼科技有限公司 | Color difference measuring method and system based on device-independent color space and intelligent terminal |
CN113670443B (en) * | 2021-07-09 | 2024-01-12 | 北京中科慧眼科技有限公司 | Color difference measurement method and system based on device-independent color space and intelligent terminal |
CN113469285A (en) * | 2021-07-26 | 2021-10-01 | 上海大学 | Dye double-spelling formula prediction method based on PSO-LSSVM |
CN113536687A (en) * | 2021-07-26 | 2021-10-22 | 上海大学 | Construction method of monochrome cotton fabric color and dye concentration prediction model based on PSO-LSSVM |
CN113781476A (en) * | 2021-10-27 | 2021-12-10 | 南通博纳纺织品有限公司 | Textile dyeing quality evaluation method and system based on image processing |
CN113781476B (en) * | 2021-10-27 | 2024-05-28 | 汕头市鼎泰丰实业有限公司 | Textile dyeing quality assessment method and system based on image processing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110378393A (en) | A kind of mixing printing product acetes chinensis method based on PSO-GSA-SVM | |
CN110111335B (en) | Urban traffic scene semantic segmentation method and system for adaptive countermeasure learning | |
CN112766199B (en) | Hyperspectral image classification method based on self-adaptive multi-scale feature extraction model | |
CN106778604B (en) | Pedestrian re-identification method based on matching convolutional neural network | |
CN112507793B (en) | Ultra-short term photovoltaic power prediction method | |
CN105574505B (en) | The method and system that human body target identifies again between a kind of multiple-camera | |
CN110414368A (en) | A kind of unsupervised pedestrian recognition methods again of knowledge based distillation | |
CN104657718B (en) | A kind of face identification method based on facial image feature extreme learning machine | |
CN110457982A (en) | A kind of crop disease image-recognizing method based on feature transfer learning | |
CN108764298B (en) | Electric power image environment influence identification method based on single classifier | |
CN109559310A (en) | Power transmission and transformation inspection image quality evaluating method and system based on conspicuousness detection | |
CN104598924A (en) | Target matching detection method | |
CN107133575A (en) | A kind of monitor video pedestrian recognition methods again based on space-time characteristic | |
CN110781829A (en) | Light-weight deep learning intelligent business hall face recognition method | |
CN108520114A (en) | A kind of textile cloth defect detection model and its training method and application | |
CN108876797A (en) | A kind of image segmentation system and method based on Spiking-SOM neural network clustering | |
CN108647736A (en) | A kind of image classification method based on perception loss and matching attention mechanism | |
CN107341432A (en) | A kind of method and apparatus of micro- Expression Recognition | |
CN109376787A (en) | Manifold learning network and computer visual image collection classification method based on it | |
CN107680116A (en) | A kind of method for monitoring moving object in video sequences | |
CN108647595A (en) | Vehicle recognition methods again based on more attribute depth characteristics | |
CN108960404A (en) | A kind of people counting method and equipment based on image | |
CN106874879A (en) | Handwritten Digit Recognition method based on multiple features fusion and deep learning network extraction | |
CN110321862A (en) | A kind of pedestrian's recognition methods again based on the loss of compact ternary | |
CN110503078A (en) | A kind of remote face identification method and system based on deep learning |
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 |