CN102509008A - Method for evaluating scratchiness of ramie fabrics objectively - Google Patents
Method for evaluating scratchiness of ramie fabrics objectively Download PDFInfo
- Publication number
- CN102509008A CN102509008A CN201110340422XA CN201110340422A CN102509008A CN 102509008 A CN102509008 A CN 102509008A CN 201110340422X A CN201110340422X A CN 201110340422XA CN 201110340422 A CN201110340422 A CN 201110340422A CN 102509008 A CN102509008 A CN 102509008A
- Authority
- CN
- China
- Prior art keywords
- fabric
- prodding
- network
- scratchiness
- neural network
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for evaluating scratchiness of ramie fabrics objectively. The method comprises the following steps of: selecting ramie fabrics with different specification as main test samples, selecting healthy and sensitive female university students as testees, and performing a forearm test for evaluating and testing the scratchiness subjectively on the selected test samples; testing the selected test samples by using a KES-FB fabric style instrument to obtain 16 performance indexes about friction, compression, bending and tensile and shearing performances; and by applying a programming method of a Matlab neural network tool kit, taking a scratchiness level obtained in the subjective evaluation as output, taking the 16 performance indexes tested by the KES-FB fabric style instrument as input, and performing training and studying by a BP neural network so as to fulfill a preset network error target and realize the objective evaluation on the scratchiness of the ramie fabrics. The method for evaluating the scratchiness of the ramie fabrics objectively has the advantages that: various performances of the fabrics can be tested by a KES-FB fabric style system; and through establishment and simulation of the BP neural network, the scratchiness level of the ramie fabrics can be pre-tested relatively accurately.
Description
Technical field
The present invention relates to the ramie fabric field, mainly is a kind of method for objectively evaluating of The Study Of Ramee Fabric.
Background technology
Prodding and itching feeling is meant that can produce puncture when human body wears next to the skin wool or bast fiber fabrics usually causes the discomfort sensation of itching, and its generation is the comprehensive of complex processes such as physics, physiology, psychology.The crystallinity of ramee Yin Qigao and the degree of orientation, it is thick and stiff that the fiber appearance is straight, poor flexibility, elongation at break is low, causes ramie yarn and thread and cloth cover filoplume thereof many and not flexible when firm and stressed, when this fiber contacts with skin, makes the people feel strong scratchy.
At present, the The Study Of Ramee Fabric evaluation mainly is divided into subjective assessment and objective evaluation two big classes.Subjective assessment is mainly subjective touch feeling evaluation, and is similar with traditional hand valuation.Comparatively ripe method mainly contains two types: forearm experiment and wearing feeling are estimated.Subjective assessment can intuitively reflect the power of fabric prodding and itching feeling, still receives main body factor and Effect of Environmental bigger, and evaluation procedure is implemented comparatively complicated.The objective appraisal method is mainly measured through content and fabric tubbiness filoplume amount three to the bendind rigidity of fiber, thick and stiff fiber, like the fiber needle or the fiber brush Ci Zhafa of fiber bendind rigidity mensuration, crude fiber content evaluation assessment, filoplume counting method, apparent thickness mensuration and simulation fiber puncture effect and the scratchy effect essence of process expression fabric.Because it is comparatively complicated to influence the factor of fabric prodding and itching feeling, these objective evaluation indexs are comparatively single, can't reflect the situation that prodding and itching feeling changes comprehensively.
KES-FB Fabric Style system is to reflect that as far as possible comprehensively fabric property is a starting point; Test has characterized the overall process of the stretching of fabric under little stress, the small deformation condition, shearing, bending, compression; Comprise 16 physical quantitys that are used to evaluate the fabric feeling style, have complicated nonlinear relationship between these physical quantitys and the fabric prodding and itching feeling.
The BP neural network is a kind of multilayer feedforward network, generally has the neural network more than three layers or three layers, comprises input layer, latent layer and output layer, and it can be realized from being input to any Nonlinear Mapping of output.The learning process of BP algorithm is made up of forward-propagating and backpropagation.In the forward-propagating process, input information is successively handled through hidden layer from input layer, and passes to output layer.The neuronic state of each layer only influences the neuronic state of one deck down.If the output that output layer can not get expecting then changes backpropagation over to, this moment, error signal was propagated to input layer from output layer, and adjusted neuronic connection weight coefficient of each layer and threshold value on the way, and error signal is constantly reduced.The BP neural network has characteristics such as self study, self-organization, self-adaptation and nonlinear dynamic processing, is particularly suitable for nonlinear relationship complicated between deal with data.
The present invention is output with the prodding and itching feeling grade that the test of prodding and itching feeling subjective assessment forearm obtains; 16 physical performance indexs that KES-FB records are input; Set up the BP neural network model of a fabric mechanics characteristic and prodding and itching feeling rank correlation, be used to estimate The Study Of Ramee Fabric.
Summary of the invention
The not direct and specific aim that The present invention be directed to existing fabric prodding and itching feeling objective evaluation is poor; Rely on the problem that subjective sensation is judged; Proposing a kind of method for objectively evaluating of The Study Of Ramee Fabric, is a kind of assessment method based on BP neural network prediction The Study Of Ramee Fabric, particularly can be through each item performance of KES-FB Fabric Style system measurement fabric; Through the structure and the emulation of BP neural network, dope the prodding and itching feeling grade of ramie fabric comparatively exactly.
Technical scheme below the present invention has adopted: the method for objectively evaluating of this The Study Of Ramee Fabric, this method comprises the steps:
(1), the ramie fabric of choosing different size is as sample; Under standard damp-warm syndrome degree; Utilize KES-FB Fabric Style appearance that selected sample is tested, obtain 16 physical performance indexs of relevant ramie fabric mantle friction performance, compression performance, bending property and tensile and shear property;
(2), selected sample is carried out fabric prodding and itching feeling subjective assessment-forearm test, the fabric prodding and itching feeling subjective assessment grade of records appraisal gained under standard atmosphere;
(3), the programmed method of Application of Matlab Neural Network Toolbox, the input layer of design BP neural network, output layer, latent layer and network training parameter are carried out normalization with raw data and are handled the back fan-in network; As input, promptly the input neuron number is 16 with 16 measured physical performance indexs, with subjective assessment prodding and itching feeling grade as output; Then the output neuron number is 1, and the hidden neuron number is 5-15, and the transport function of the hidden layer neuron of network is tansig; The neuronic transport function of output layer is purelin, adopts elasticity gradient descent method, and its training function is trainrp; Designing maximum frequency of training is 2000, and the training requirement precision is 1 * 10
-5, all the other parameter values are default;
(4), utilize the newff order of Matlab to set up aforesaid three layers of BP neural network; The BP network model of setting up is carried out weights, the threshold value output of simulation training and map network; For setting up good network model; Only need under the corresponding weights of network and threshold value, to import 16 performance datas that the ramie fabric of other specification records through KES-FB Fabric Style appearance, i.e. the prodding and itching feeling grade of measurable this fabric.
As preferably, raw data to be carried out normalization handle, it is be the number between interval [0,1] in order to eliminate the influence that numerical value differs greatly neural network is caused, to be about to data processing that the normalization of raw data is handled; Be in the foundation of BP network model data to be carried out pretreated conventional method.Input layer data by formula (1) are handled, and the output layer data are handled according to formula (2).
Wherein, X representes the raw data of selected each performance index of sample; Xmin representes numerical value minimal data in a certain performance index of selected sample; Xmax representes the maximum data of numerical value in a certain performance index of selected sample.
Wherein, X representes selected sample prodding and itching feeling subjective assessment raw data.
The effect that the present invention is useful is: the BP neural network model can be associated the subjective assessment of fabric prodding and itching feeling with each item performance index in the KES-FB Fabric Style system.This network model can be accomplished training study rapidly, and the output result has very high precision and accuracy, only needs can obtain through 16 objective physical performance indexs of test fabric the grade of prodding and itching feeling subjective assessment.Therefore, it is comparatively quick, easy to substitute the prodding and itching feeling subjective assessment with the BP neural network model.
Description of drawings
Fig. 1 is three layers of BP neural network structure figure;
Fig. 2 is the figure as a result of BP neural metwork training among the present invention.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is described further:
The thinking of this patent is to utilize the performance index of 16 fabrics that the BP neural network model records prodding and itching feeling subjective assessment (forearm test) result and KES Fabric Style appearance to associate.The ramie fabric of choosing different size carries out prodding and itching feeling subjective assessment test forearm test as the experimenter to selected sample with healthy sensitive collegegirl as main sample.Utilize KES-FB Fabric Style appearance that selected sample is tested simultaneously, obtain 16 performance index of relevant fabric friction, compression, bending and tensile and shear property.The programmed method of Application of Matlab Neural Network Toolbox then; With subjective assessment prodding and itching feeling grade as output; With measured 16 physical indexs of KES-FB Fabric Style appearance as input; Through the training study of BP neural network, reach preset network error target, realize the objective evaluation of The Study Of Ramee Fabric.After modelling is got up, only need just can obtain fabric prodding and itching feeling grade through the KES test, the method for subjective assessment is not the summary of the invention of our patent.
Choose fabric sample damping under standard state and carry out the test of subjective assessment of fabric prodding and itching feeling and KES Fabric Style then more than 24 hours, forearm test, 16 objective performance index of record prodding and itching feeling subjective assessment grade and fabric are taked in subjective assessment.The input layer of design BP neural network, output layer, latent layer and network training parameter; Raw data is carried out normalization handle the back fan-in network; The programmed method of Application of Matlab Neural Network Toolbox is realized; At last the network model of setting up is carried out simulation training, be used for the prediction of fabric prodding and itching feeling grade.
Illustrate:
Choosing with crudefiber crop and blend fabric thereof is 10 fabric sample of master, and specimen size is 20cm * 20cm, and every sample is numbered 1～10, and wherein 1#～9# is the sample group of network training study, and 10# is the network verification sample.
The forearm test is adopted in the prodding and itching feeling subjective assessment.Opinion rating is divided into Pyatyi, is respectively no scratchy, slightly scratchy, more scratchy, very scratchy and extremely scratchy.With 30 healthy sensitive collegegirls as the experimenter.In temperature is (20 ± 1) ℃, and relative humidity is under the environment of (65 ± 5) %, and the experimenter is sitting in and takes a seat on the chair and be familiar with beginning experiment after prodding and itching feeling is estimated scale.The eyes of muffling the experimenter with soft fabric lightly impact prodding and itching feeling to avoid appearance of fabrics.Experimenter's right hand is put on rubber gloves, stretches out left forearm square on desktop, and facies medialis brachii upwards before making.The main examiner takes out 1 sample at random it is put in experimenter's forearm from 10 samples; Please the experimenter use the right hand applying light fabric of band gloves and promote fabric back and forth then; Make fabric and forearm inside skin that extruding and friction take place, experience the prodding and itching feeling degree of this fabric and judge its prodding and itching feeling grade.The main examiner inquires that the experimenter is to the grading of this fabric prodding and itching feeling and keep a record.30 people's opinion rating is averaged, draw this fabric prodding and itching feeling subjective assessment grade.
Under standard damp-warm syndrome degree; 16 physical quantitys that comprise fabric tension cutting performance, bending property, compression performance, mantle friction performance that record with KES-FB Fabric Style appearance are as the objective evaluation index, and each item performance is all carried out under the standard test condition of instrument.
As input, promptly the input neuron number is 16 with measured 16 physical indexs of KES-FB Fabric Style appearance.As output, then the output neuron number is 1 with subjective assessment prodding and itching feeling grade.The transport function of the hidden layer neuron of network is tansig, and the neuronic transport function of output layer is purelin, adopts elasticity gradient descent method, and its training function is trainrp.Designing maximum frequency of training is 2000, and the training requirement precision is 1 * 10
-5, all the other parameter values are default.Because the hidden neuron number can change in a big way, therefore design the hidden neuron number could vary, the error size that is produced when training according to network model is confirmed.Training result shows, the hidden neuron number is between 5～15 the time, and network error is all less and be more or less the same, and can choose intermediate value 10 and be the hidden neuron number.
In order to eliminate the influence that numerical value differs greatly neural network is caused, need carry out normalization to raw data and handle.Input layer data by formula (1) are handled, and the output layer data are handled according to formula (2).
Utilize the newff order of Matlab to set up three layers of BP neural network of aforesaid 16-10-1.Program code can be with reference to as follows:
net＝newffminmax(P)，[10，1]，{′tansig′，′purelin′}，′trainrp′)；
net＝init(net)；
net.trainParam.epochs＝2000；
net.trainParam.goal＝0.00001；
net＝train(net，P，T)；
The network training conditional curve is as shown in Figure 1.
The BP network model of setting up is carried out weights, the threshold value output of simulation training and map network, and program code can be with reference to as follows:
y＝sim(net，P)；
error＝y-T；
res＝norm(error)；
w＝net.IW{1，1}；
b＝net.b{1}；
The program output of training back:
Error amount:
error＝-0.0020，-0.0019，-0.0063，0.0039，0.0034，0.0013，0.0009，-0.0031，-0.0003；
res＝0.0093；
Weights:
Threshold value:
b＝-1.0627
-1.4265
-2.6028
-0.8442
-0.0870
-0.4772
0.1124
0.5605
0.2478
-2.8643
Training output valve and actual comparison that above-mentioned BP neural network model simulation training obtains are as shown in table 1.
Table 1 network simulation training result
Can know that by last table the training output valve of BP neural network model and relative error＜3% of subjective assessment are used for the subjective assessment of alternative fabrics prodding and itching feeling and have suitable accuracy.
When the BP neural network adapts to setting up the used data of model fully, may bring the problem that adapts to, promptly to complete qualified data, the conclusion that draws is necessarily correct, but there are the data than mistake in some, and adaptability just descends.Therefore, the sample of selection will possess representativeness, and sample size is big as far as possible.Sample size is less in the instance explanation of the present invention, and the possibility that adaptive faculty in forecasting process, occurred is bigger.The design of BP neural network and train does not subjectively have regular rule to follow, and learning and memory has instability, and weights and the threshold value of each training all are at random, can take the method for some optimization initial weights to improve stability of network.
Except that the foregoing description, the present invention can also have other embodiments.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop on the protection domain of requirement of the present invention.
Claims (2)
1. the method for objectively evaluating of a The Study Of Ramee Fabric, it is characterized in that: this method comprises the steps:
(1), the ramie fabric of choosing different size is as sample; Under standard damp-warm syndrome degree; Utilize KES-FB Fabric Style appearance that selected sample is tested, obtain 16 physical performance indexs of relevant ramie fabric mantle friction performance, compression performance, bending property and tensile and shear property;
(2), selected sample is carried out fabric prodding and itching feeling subjective assessment-forearm test, the fabric prodding and itching feeling subjective assessment grade of records appraisal gained under standard atmosphere;
(3), the programmed method of Application of Matlab Neural Network Toolbox, the input layer of design BP neural network, output layer, latent layer and network training parameter are carried out normalization with raw data and are handled the back fan-in network; As input, promptly the input neuron number is 16 with 16 measured physical performance indexs, with subjective assessment prodding and itching feeling grade as output; Then the output neuron number is 1, and the hidden neuron number is 5-15, and the transport function of the hidden layer neuron of network is tansig; The neuronic transport function of output layer is purelin, adopts elasticity gradient descent method, and its training function is trainrp; Designing maximum frequency of training is 2000, and the training requirement precision is 1 * 10
-5, all the other parameter values are default;
(4), utilize the newff order of Matlab to set up aforesaid three layers of BP neural network; The BP network model of setting up is carried out weights, the threshold value output of simulation training and map network; For setting up good network model; Only need under the corresponding weights of network and threshold value, to import 16 performance datas that the ramie fabric of other specification records through KES-FB Fabric Style appearance, i.e. the prodding and itching feeling grade of measurable this fabric.
2. the method for objectively evaluating of The Study Of Ramee Fabric according to claim 1 is characterized in that: raw data is carried out normalization handle, input layer data by formula (1) are handled, and the output layer data are handled according to formula (2);
Wherein, X representes the raw data of selected each performance index of sample; Xmin representes numerical value minimal data in a certain performance index of selected sample; Xmax representes the maximum data of numerical value in a certain performance index of selected sample;
Wherein, X representes selected sample prodding and itching feeling subjective assessment raw data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110340422XA CN102509008A (en) | 2011-11-01 | 2011-11-01 | Method for evaluating scratchiness of ramie fabrics objectively |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110340422XA CN102509008A (en) | 2011-11-01 | 2011-11-01 | Method for evaluating scratchiness of ramie fabrics objectively |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102509008A true CN102509008A (en) | 2012-06-20 |
Family
ID=46221093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110340422XA Pending CN102509008A (en) | 2011-11-01 | 2011-11-01 | Method for evaluating scratchiness of ramie fabrics objectively |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102509008A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304351A (en) * | 2017-12-27 | 2018-07-20 | 广州唯品会研究院有限公司 | A kind of fabric touch information transmitting methods |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1220877C (en) * | 2003-09-12 | 2005-09-28 | 东华大学 | Fabric planeness gradation objective evaluation method |
| CN1948603A (en) * | 2006-11-10 | 2007-04-18 | 苏州大学 | Method of identifying woven fabric defect |
| CN101419706A (en) * | 2008-12-11 | 2009-04-29 | 天津工业大学 | Jersey wear flokkit and balling up grading method based on image analysis |
-
2011
- 2011-11-01 CN CN201110340422XA patent/CN102509008A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1220877C (en) * | 2003-09-12 | 2005-09-28 | 东华大学 | Fabric planeness gradation objective evaluation method |
| CN1948603A (en) * | 2006-11-10 | 2007-04-18 | 苏州大学 | Method of identifying woven fabric defect |
| CN101419706A (en) * | 2008-12-11 | 2009-04-29 | 天津工业大学 | Jersey wear flokkit and balling up grading method based on image analysis |
Non-Patent Citations (6)
| Title |
|---|
| 刘宇清等: "苎麻织物毛羽刺痒感的力学评价", 《中国麻业》 * |
| 曹建达等: "用BP神经网络预测棉织物的手感", 《纺织学报》 * |
| 杨秀波: "基于BP神经网络的织物折皱等级评定", 《东华大学学报（自然科学版）》 * |
| 梅兴波 等: "BP神经网络预测织物拉伸性能", 《东华大学学报(自然科学版)》 * |
| 欧建文等: "织物柔软度等级评价的BP网络方法", 《山东纺织科技 》 * |
| 王革辉等: "织物刺痒感的主观评价研究", 《毛纺科技》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304351A (en) * | 2017-12-27 | 2018-07-20 | 广州唯品会研究院有限公司 | A kind of fabric touch information transmitting methods |
| CN108304351B (en) * | 2017-12-27 | 2022-01-07 | 广州品唯软件有限公司 | Fabric touch information transmission method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pacewicz et al. | A meta-analytic review of the relationship between social constructs and athlete burnout | |
| Killham et al. | Women athletes’ self-compassion, self-criticism, and perceived sport performance. | |
| CN104856704B (en) | The subjective and objective Psychological Evaluation method and system being combined | |
| CN104997523A (en) | Upper limb rehabilitation robot rehabilitation training motion function assessment method | |
| CN101916334B (en) | A kind of skin Forecasting Methodology and prognoses system thereof | |
| CN114822763A (en) | Personalized exercise prescription recommendation method driven by exercise data | |
| Raiola et al. | Statistical study on bodily communication skills in volleyball to improve teaching methods | |
| CN111150410B (en) | Psychological pressure evaluation method based on fusion of electrocardiosignals and electromyographic signals | |
| Rodrigues et al. | Assessing Need Satisfaction and Frustration in Portuguese Exercise Instructors: scale validity, reliabity and invariance between gender | |
| CN110717524A (en) | Method for predicting thermal comfort of old people | |
| Mohamad et al. | Measuring positive youth development: Confirmatory factor analysis | |
| Lee et al. | The effects of health perception on living health belief, living satisfaction and wellbeing-oriented activities according to swimming participation with middle-aged women | |
| Tang et al. | Prediction of college students’ physical fitness based on K-means clustering and SVR | |
| Sülar et al. | Objective evaluation of fabric handle by simple measurement methods | |
| CN102509008A (en) | Method for evaluating scratchiness of ramie fabrics objectively | |
| Zhang et al. | Neural network-based thermal comfort prediction for the elderly | |
| Jain et al. | The Coaching Behavior Scale for Sport (CBS-S): Factor structure examination for elite Indian sportsperson. | |
| Apay et al. | Validity and Reliability Study of Modified Weight Bias Internalization Scale in Turkish. | |
| Nuruhidin et al. | An evaluation of powerlifting and weightlifting development program | |
| CN106295202B (en) | Juvenile healthy situation dynamic analysing method based on Hale indexes | |
| CN116739037A (en) | Personality model construction method and device with personality characteristics | |
| Wu et al. | Assessment of upper limb tremors in patients with Parkinson’s disease based on displacement and acceleration information | |
| Sun et al. | Measuring and multilevel fuzzy comprehensive predicting comfort parameters of soft materials by a new handle evaluation system | |
| CN108304351A (en) | A kind of fabric touch information transmitting methods | |
| Wang et al. | Model of normal university students’ education quality based on artificial neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120620 |