CN118463864A - Fabric flatness detection control method, equipment and application thereof - Google Patents
Fabric flatness detection control method, equipment and application thereof Download PDFInfo
- Publication number
- CN118463864A CN118463864A CN202410918848.6A CN202410918848A CN118463864A CN 118463864 A CN118463864 A CN 118463864A CN 202410918848 A CN202410918848 A CN 202410918848A CN 118463864 A CN118463864 A CN 118463864A
- Authority
- CN
- China
- Prior art keywords
- fabric
- flatness
- data
- receiver
- detection
- 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.)
- Granted
Links
- 239000004744 fabric Substances 0.000 title claims abstract description 168
- 238000001514 detection method Methods 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000011156 evaluation Methods 0.000 claims abstract description 75
- 238000005457 optimization Methods 0.000 claims abstract description 32
- 230000035945 sensitivity Effects 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 230000008713 feedback mechanism Effects 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 33
- 238000012360 testing method Methods 0.000 claims description 24
- 238000010801 machine learning Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 22
- 230000005856 abnormality Effects 0.000 claims description 21
- 238000009826 distribution Methods 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 20
- 230000002159 abnormal effect Effects 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 12
- 239000003086 colorant Substances 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 230000009471 action Effects 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000007637 random forest analysis Methods 0.000 description 21
- 238000012545 processing Methods 0.000 description 10
- 230000008676 import Effects 0.000 description 9
- 239000000463 material Substances 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 238000003066 decision tree Methods 0.000 description 5
- 230000001965 increasing effect Effects 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000012706 support-vector machine Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000009499 grossing Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 208000025174 PANDAS Diseases 0.000 description 3
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013210 evaluation model Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 239000004753 textile Substances 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 240000004718 Panda Species 0.000 description 2
- 235000016496 Panda oleosa Nutrition 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 239000010755 BS 2869 Class G Substances 0.000 description 1
- 229920000742 Cotton Polymers 0.000 description 1
- 229920000297 Rayon Polymers 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000000149 argon plasma sintering Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 229920000728 polyester Polymers 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000002964 rayon Substances 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- 210000002268 wool Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06H—MARKING, INSPECTING, SEAMING OR SEVERING TEXTILE MATERIALS
- D06H3/00—Inspecting textile materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Materials Engineering (AREA)
- Computing Systems (AREA)
- Textile Engineering (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Chemical & Material Sciences (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application provides a fabric flatness detection control method, equipment and application thereof, and aims to solve the problems of insufficient detection precision, low intelligent level, low automation degree and poor adaptability to complex fabrics in the prior art. According to the method, the emitter and the receiver are matched, the light intensity and the sensitivity are adaptively adjusted according to the reflection characteristics of the fabric, the primary flatness assessment is carried out by utilizing a multi-threshold judgment and flattening weighting algorithm, and the detection precision is remarkably improved by combining an optimization model and a secondary detection mechanism. The method specifically comprises the following steps: signal acquisition and preprocessing, primary flatness evaluation, optimization model evaluation, secondary detection, comprehensive evaluation, feedback mechanism and the like. According to the application, by introducing key technologies such as self-adaptive signal adjustment, optimization model evaluation, comprehensive evaluation feedback and the like, the accuracy, efficiency and flexibility of fabric flatness detection are remarkably improved.
Description
Technical Field
The application relates to the technical field of detection, in particular to a fabric flatness detection control method, equipment and application thereof.
Background
With the rapid development of textile industry, the requirements of consumers on the quality of the fabric are increasingly increased, and the flatness of the fabric becomes one of important indexes for measuring the quality and the comfort of the fabric. The traditional fabric flatness detection methods are mostly dependent on manual visual inspection or simple mechanical touch modes, and are low in efficiency, high in subjectivity and difficult to meet the requirements of modern industrial mass production.
In recent years, optical detection techniques have been widely used in the detection of textile flatness. The optical detection principle is to irradiate the surface of the fabric by using a light source and evaluate the flatness of the fabric by analyzing the distribution condition of reflected light. However, the prior art (such as CN 220552428U) still has some drawbacks, mainly including:
1. Limitations of data processing algorithms: the existing algorithm usually adopts only single threshold value judgment, cannot accurately distinguish fabrics with different flatness levels, and lacks consideration on the importance of a detection area, so that the comprehensiveness and accuracy of a detection result are affected.
2. Defects of the signal processing algorithm: in the actual detection process, the received signals are easy to be interfered by environmental noise, and the influence of different fabric materials and colors on the fabric flatness detection result is not fully considered, so that the detection result is unstable.
3. Lack of intelligent detection model: most of the current detection systems are based on fixed rules and parameters, and a large number of fabric samples cannot be analyzed by fully utilizing a machine learning technology, so that more accurate flatness grade judgment is realized.
4. Absence of the result calibration mechanism: in the prior art, when the complex conditions such as tiny pits on the surface of the fabric are treated, misjudgment is easy to generate, and an effective secondary detection and result correction mechanism is lacked.
5. The degree of automation is not high: although partial detection equipment has realized preliminary automation, manual intervention is still required in the aspects of fabric positioning, detection flow control, result analysis and the like, and the detection efficiency and accuracy are reduced.
Aiming at the problems, the invention provides a brand new fabric flatness detection technical scheme, and aims to greatly improve the accuracy and efficiency of fabric flatness detection and meet the requirements of the modern textile industry on high-quality and high-efficiency production by optimizing a data processing algorithm, enhancing signal processing capacity, applying an optimization model, establishing a result calibration algorithm and realizing high automation of a system. The technical scheme of the invention is not only suitable for various types of fabrics, but also suitable for complex and changeable production environments, and has wide application prospect.
Disclosure of Invention
The embodiment of the application provides a fabric flatness detection control method, equipment and application thereof, aiming at the problems of insufficient detection precision, low intelligent level, low automation degree, poor adaptability to complex fabrics and the like in the prior art.
The core technology of the invention mainly relates to a method and a system for realizing high-precision fabric flatness automatic detection and evaluation by utilizing light transmission and reflection characteristics and combining with a self-adaptive parameter adjustment, multi-threshold judgment, machine learning optimization and comprehensive evaluation model.
In a first aspect, the present application provides a fabric flatness detection method, the method comprising the steps of:
S00, emitting light through the emitter, receiving the light passing through the fabric through the receiver, noise filtering the received signal, and adjusting the light intensity of the emitter and/or the sensitivity of the receiver according to the reflection characteristics of different fabrics;
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
S10, calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation to consider flatness differences of all parts of the fabric and mark out areas possibly having abnormality;
S20, inputting a result of the preliminary flatness grade and receiving proportion data of a receiver into a pre-trained optimization model to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
S30, if an area with the possibility of abnormality exists, performing secondary detection to determine and calibrate a result and taking the result as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
s40, analyzing final result data, giving comprehensive evaluation of fabric flatness according to a preset rule and/or an analysis model, and displaying detection results in real time.
Further, in the step S00, the specific steps of adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the reflection characteristics of different fabrics are as follows:
Before detection, initializing a transmitter and a receiver, and setting default light intensity and sensitivity;
in the detection process, the signal intensity received by the receiver is monitored in real time, and the signal intensity can reflect the reflection characteristic of the fabric on light;
analyzing the received signal strength and judging whether the received signal strength exceeds a preset dynamic range;
If not, the light intensity of the emitter is insufficient or the fabric absorbs excessive light, so that the light intensity of the emitter and/or the sensitivity of the receiver need to be improved; if yes, the light intensity of the emitter is too high or the fabric emits too much light, and the light intensity of the emitter and/or the sensitivity of the receiver need to be reduced;
Adjusting the light intensity of the transmitter and/or the sensitivity of the receiver based on the analysis result; wherein the magnitude of the adjustment is based on a predefined scaling factor to prevent over-adjustment of the parameter;
the adjusted parameters are applied to the transmitter and/or receiver and then the signal strength is re-monitored to form a closed loop control until the signal strength stabilizes within a preset dynamic range.
Further, in the step S00, the specific step of adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the reflection characteristics of different fabrics further includes:
The optimal parameter settings under different fabric types and colors are stored for direct recall at the next detection.
Further, in step S10, a plurality of thresholds are set according to the light intensity distribution data to distinguish different flatness states, as a preset flatness level standard;
for each set of measurement data, classifying it into different flatness classes according to its relationship with a set threshold;
carrying out weighted average on the flatness grade of each position of the detection area of the fabric;
mapping the flatness grade obtained by weighted average to a preset flatness grade standard;
and outputting the overall flatness grade of the fabric to reflect the average flatness level of the fabric.
Further, in step S20, the training step of the optimization model is as follows:
acquiring data containing flatness information of the fabric, wherein the data at least comprises an image of the fabric, receiving proportion data of a receiver and the type of the fabric;
Performing data cleaning to remove invalid or wrong data, and distributing a flatness grade label for each piece of cleaned data;
extracting features from the cleaned data, wherein the features at least comprise texture features, color features and statistical features of receiving proportion data of the fabric;
Dividing the data set to obtain a training set, a verification set and a test set;
Establishing a machine learning model, training the machine learning model by using a training set, and performing model selection and super-parameter tuning by using verification set data in the training process so as to prevent over-fitting;
and evaluating the generalization capability of the machine learning model by the test set until the machine learning model is trained, so as to obtain an optimized model.
Further, in step S40, key features are extracted from the final result data, where the key features include texture features of the fabric, reflection features of color light, statistical features of the received ratio data, and distribution and severity of abnormal regions;
Inputting the extracted key features into a preset rule and/or analysis model for evaluation to obtain final flatness evaluation;
the final flatness assessment is converted to a flatness level and a feedback mechanism is established and a report is generated, including charts and graphs, lists of key indicators, and suggested actions.
Further, the method also comprises a step S50, wherein the analysis model is continuously optimized according to the detection result and feedback so as to improve the accuracy and reliability of the evaluation.
In a second aspect, the present application provides a fabric flatness detecting device, including:
The device comprises a transmitter and a receiver, wherein the transmitter emits light, the receiver receives the light passing through the fabric, noise filtering is carried out on the received signal, and meanwhile, the light intensity of the transmitter and/or the sensitivity of the receiver are adjusted according to the reflection characteristics of different fabrics;
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
the signal preprocessing module is used for calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation so as to consider the flatness difference of each part of the fabric and mark out the area possibly having abnormality;
The preliminary evaluation module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The optimization module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The secondary detection module is used for executing secondary detection if an area with the possibility of abnormality exists, so as to determine and calibrate a result and serve as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
the comprehensive evaluation module is used for analyzing the final result data, giving out comprehensive evaluation of the fabric flatness according to a preset rule and/or an analysis model, and displaying the detection result in real time;
and the output module is used for outputting the comprehensive evaluation and detection result.
In a third aspect, the present application provides an electronic device comprising a memory in which a computer program is stored, and a processor arranged to run the computer program to perform the fabric flatness detection method described above.
In a fourth aspect, the present application provides a readable storage medium having stored therein a computer program comprising program code for controlling a process to perform a process comprising a fabric flatness detection method according to the above.
The main contributions and innovation points of the invention are as follows:
1. adaptive signal conditioning mechanism: the invention introduces the function of adaptively adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the type and the color of the fabric, ensures the accurate detection of different fabric characteristics, and solves the problem of detection errors caused by fixed setting of the light intensity or the sensitivity in the prior art.
2. Multi-threshold and smoothing weighting algorithm: the preliminary flatness evaluation stage adopts a multi-threshold judgment and smoothing weighting algorithm, so that flatness differences of all parts of the fabric can be identified more carefully, abnormal areas are marked effectively, and the comprehensiveness and accuracy of preliminary evaluation are improved.
3. Optimizing a model and carrying out secondary detection: by inputting the primary evaluation result and the receiving proportion data into a pre-trained optimization model and assisting with a secondary detection mechanism, the flatness evaluation accuracy is remarkably improved, and a more accurate solution is provided particularly for complex or difficult-to-judge situations.
4. Comprehensive evaluation and feedback mechanism: the comprehensive evaluation module not only evaluates the key characteristics of the final result data, but also establishes a feedback mechanism and a report generating function, thereby being beneficial to real-time monitoring and continuous optimization of the detection process and improving the practicability and efficiency of the whole detection system.
5. Intelligent and automatic integration: the whole detection method and system design is highly intelligent and automatic, and a set of closed-loop control system is formed from signal acquisition, preprocessing and evaluation to final result analysis, so that manual intervention is greatly reduced, and detection speed and consistency are improved.
6. Continuous optimization and iteration capabilities: by establishing a model continuous optimization mechanism, the invention can continuously adjust and improve the evaluation model according to the detection result and feedback, ensure the advancement and adaptability of the technology and meet the requirements of different application scenes.
7. Modification with low cost: the method can be directly used on the basis of the existing equipment, the hardware structure of the existing equipment is not required to be modified, and the upgrading modification can be completed only by introducing the detection method of the application.
In conclusion, the accuracy, efficiency and flexibility of fabric flatness detection are remarkably improved by introducing key technologies such as self-adaptive signal adjustment, multi-threshold judgment, optimal model evaluation, comprehensive evaluation feedback and the like.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the other features, objects, and advantages of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a flowchart of a fabric flatness detection method according to an embodiment of the present application;
fig. 2 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with aspects of one or more embodiments of the present description as detailed in the accompanying claims.
It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than described in this specification. Furthermore, individual steps described in this specification, in other embodiments, may be described as being split into multiple steps; while various steps described in this specification may be combined into a single step in other embodiments.
Example 1
The application aims to provide a fabric flatness detection method, and particularly relates to a method for detecting fabric flatness, which comprises the following steps of:
S00, emitting light through the emitter, receiving the light passing through the fabric through the receiver, performing noise filtration (such as a filter) on the received signal, and adjusting the light intensity of the emitter and/or the sensitivity of the receiver according to the reflection characteristics of different fabrics;
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
In particular, the reception ratio data of the receiver refers to the ratio of the intensity or energy of the light received by the receiver with respect to the intensity or energy of the light emitted by the emitter. Specifically, when the emitter emits light toward the surface of the fabric, a portion of the light is reflected or transmitted by the fabric, and the receiver is responsible for capturing the portion of the light. The ratio between the amount of light received by the receiver and the amount of light emitted by the emitter, namely the received ratio data, is one of the key indexes for evaluating the flatness of the fabric. The reception ratio data is important because:
Reflecting the characteristics of the fabric: the flatness of the fabric affects the way it reflects and transmits light. The smooth surface of the fabric can reflect light more uniformly, and the uneven or wrinkled fabric can cause light scattering or blocking, so that the amount of light received by the receiver is changed.
Quantifying the flatness difference: by comparing the receiving proportion data of different areas, the flatness difference of different parts of the fabric can be quantified. If the proportion of received area is significantly lower than the surrounding area, this may indicate that the area has poor flatness.
Adapt to different fabric materials and colors: the receiving proportion data can adapt to different reflection characteristics of different materials and colors on light rays. Even under the condition of material and color change of the fabric, the flatness of the fabric can be effectively reflected by the receiving proportion data by adjusting parameters of the transmitter and the receiver.
Inputting a machine learning model: receiving scale data as features into a machine learning model may help the model learn and identify patterns related to flatness. The trained model is able to predict the flatness level of the fabric based on the received ratio data, and to make accurate evaluations even in the face of new fabric types.
In this embodiment, the specific steps of adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the reflection characteristics of different fabrics are as follows:
S01, initializing a transmitter and a receiver before detection, and setting default light intensity and sensitivity;
s02, monitoring the signal intensity received by a receiver in real time in the detection process, wherein the signal intensity can reflect the reflection characteristic of the fabric on light;
S03, analyzing the received signal strength and judging whether the received signal strength exceeds a preset dynamic range; if the signal is too weak, it may be due to insufficient light intensity from the emitter or excessive light absorption by the fabric; if the signal is too strong, it may be due to too high a light intensity from the emitter or too much light reflected from the fabric.
S04, if not, the light intensity of the emitter is insufficient or the fabric absorbs excessive light, so that the light intensity of the emitter and/or the sensitivity of the receiver need to be improved; if yes, the light intensity of the emitter is too high or the fabric emits too much light, and the light intensity of the emitter and/or the sensitivity of the receiver need to be reduced;
s05, adjusting the light intensity of the transmitter and/or the sensitivity of the receiver based on the analysis result; wherein the magnitude of the adjustment is based on a predefined scaling factor to prevent over-adjustment of the parameter;
The light intensity of the transmitter and the sensitivity of the receiver are adjusted, as based on the results of the signal analysis. For example, if the signal is too weak, the light intensity of the emitter may be increased; if the signal is too strong, the light intensity of the transmitter may be reduced or the gain of the receiver may be increased (i.e., the sensitivity may be increased). The magnitude of the adjustment should be based on a predefined scale factor, such as 1%, to prevent over-adjustment of the parameters.
S06, applying the adjusted parameters to the transmitter and/or the receiver, and then re-monitoring the signal intensity to form closed-loop control until the signal intensity is stabilized within a preset dynamic range;
s07, storing optimal parameter settings under different fabric types and colors, and directly calling the optimal parameter settings when detecting next time.
In order to improve efficiency, the algorithm can have a memory function, namely, optimal parameter settings under different fabric types and colors are stored, and when similar fabrics are encountered next time, the parameters can be directly applied, so that the adjustment process is reduced.
Preferably, the steps S01-S07 may be implemented based on PID controllers (proportional-integral-derivative controllers) or more advanced Adaptive Dynamic Programming (ADP) algorithms. PID controllers are a common feedback control mechanism that adjusts control parameters through proportional, integral, and derivative terms to achieve the desired dynamic response and stability. ADP is a reinforcement learning-based method capable of learning and optimizing control strategies online. ADP algorithm can learn the influence of fabric material and color on detection result, and automatically adjust parameters of the transmitter and the receiver to achieve optimal detection performance.
Dynamic range adjustment as with PID controllers:
Proportional term (P): and directly adjusting according to the deviation between the current signal intensity and the target intensity.
Integral term (I): and accumulating historical deviation to eliminate static error and avoid deviation accumulation in long time.
Differential term (D): the adjustment is performed according to the change rate of the deviation, so that the system is helped to respond to the change more quickly, and overshoot is reduced.
Through continuous iteration and optimization of PID parameters, the system can achieve ideal detection effect under different materials and colors of the fabrics.
S10, calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation to consider flatness differences of all parts of the fabric and mark out areas possibly having abnormality;
In this embodiment, the specific steps for calculating the preliminary flatness level of the fabric using the multi-threshold judgment and flattening weighting algorithm are:
S11, setting a plurality of thresholds to distinguish different flatness states according to the light intensity distribution data, wherein the different flatness states are used as preset flatness grade standards;
Such as: the thresholds corresponding to the three flatness levels of high, medium and low are defined, for example, the high flatness threshold is set to 90%, the medium flatness is 70% -90%, and the low flatness is lower than 70%. The flatness level of the fabric is automatically determined by comparing the percentage of light received by the receiver with these thresholds. Of course, a decision tree or rule engine may be introduced to automatically match the closest flatness levels based on the received ratio of the receiver.
S12, classifying each group of measurement data into different flatness grades according to the relation between the measurement data and a set threshold value; for example, if the measured value is lower than 70%, it is determined that the flatness is poor; if the measured value is between 70% and 90%, the flatness is judged to be general; and so on.
S13, carrying out weighted average on the flatness grade of each position of the detection area of the fabric; in the detection process, a plurality of areas of the detected fabric are sampled, the flatness measurement result of each area gives different weights according to the importance or the area size of the flatness measurement result in the fabric, and finally the weighted average flatness is calculated.
Of course, in order to comprehensively consider the flatness of each position on the fabric, the flatness level of each position is weighted and averaged. The weight may be set according to the importance level or the detection accuracy of each measurement point on the fabric, for example, the weight of the central area of the fabric may be higher than that of the edge area. The weighted average flatness class G is calculated as follows:
Where ω i is the weight of the i-th measurement point, g i is the flatness ratio of the area, and N is the total number of measurement points.
S14, mapping the flatness grade obtained by weighted average to a preset flatness grade standard;
for example, the flatness level G obtained by weighted averaging is mapped to a preset flatness level standard. This typically means that the continuous flatness values are converted into discrete levels, e.g. level a, level B, level C, etc.
S15, outputting the overall flatness grade of the fabric to reflect the average flatness level of the fabric. This rating should reflect the average flatness level of the fabric while also taking into account some local anomalies that may be present on the fabric.
In this way, the multi-threshold determination provides sensitivity to different levels of flatness, while the weighted average ensures the objectivity and accuracy of the overall evaluation, avoiding excessive impact of local anomalies on the overall evaluation.
S20, inputting a result of the preliminary flatness grade and receiving proportion data of a receiver into a pre-trained optimization model to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
in this embodiment, the training steps of the optimization model are as follows:
S21, acquiring data containing flatness information of the fabric, wherein the data at least comprise images of the fabric, receiving proportion data of a receiver, environmental conditions, fabric types and the like;
s22, cleaning data to remove invalid or wrong data, such as missing value processing, abnormal value detection and correction, and distributing flatness grade labels to each piece of cleaned data, wherein the labels can be manually given by an expert or are estimated based on a certain standard;
S23, extracting features from the cleaned data, such as texture features, color features, statistical features of the receiving proportion data and the like of the fabric;
S24, dividing the data set into a training set, a verification set and a test set, wherein the typical proportion can be 70% of the training set, 15% of the verification set and 15% of the test set;
S25, establishing a machine learning model, training the machine learning model by using a training set, and performing model selection and super-parameter tuning by using verification set data in the training process so as to prevent over-fitting;
For example, an appropriate machine learning model, such as a random forest, support vector machine, neural network, etc., is selected based on the nature of the problem. For classification problems, a multi-classification model may be selected because there may be multiple classes of flatness levels.
The model is trained using the training set data, and parameters of the model are adjusted to minimize training errors. During the training process, the verification set data is used for model selection and super-parameter tuning to prevent overfitting. Multiple iterative training may be required to optimize model performance using cross-validation techniques and the like.
S26, evaluating the generalization capability of the machine learning model by the test set until the machine learning model is trained, and obtaining an optimized model.
The evaluation index may include accuracy, recall, F1 score, etc., depending on the characteristics of the flatness level prediction task.
Thus, when prediction is performed using a model, a preliminary evaluation result (such as a result of a multi-threshold judgment and smoothing weighting algorithm) and reception ratio data of the receiver are input as features into the model. The model is able to learn the relationship between these features and the flatness levels, identifying subtle differences, and thus providing more accurate predictions.
S30, if an area with the possibility of abnormality exists, performing secondary detection to determine and calibrate a result and taking the result as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
s40, analyzing final result data, giving comprehensive evaluation of fabric flatness according to a preset rule and/or an analysis model, and displaying detection results in real time.
In this embodiment, the specific steps are:
S41, extracting key features from final result data, wherein the key features comprise statistics such as average value, standard deviation and the like of a receiving proportion, distribution and severity of an abnormal area, fabric materials, reflection characteristics of colors to light and other factors affecting flatness;
wherein, the final result data can be integrated with all data obtained from different detection methods, including: reception ratio data of the receiver; multi-threshold judgment and output of a smoothing weighting algorithm; the result of the secondary detection (if any); any other relevant measured data, such as environmental factors like temperature, humidity etc.
S42, inputting the extracted key features into a preset rule and/or analysis model for evaluation, and obtaining final flatness evaluation;
Such as: rule-based system: a series of rules is predefined, such as "if the standard deviation of the reception ratio is greater than x, the flatness level is y".
Machine learning model: the flatness level is predicted from the features using a previously trained analytical model, such as a random forest, support vector machine, or neural network.
Comprehensive evaluation model: and combining the outputs of the various rules and analysis models to give a final flatness evaluation.
S43, converting the final flatness evaluation into flatness levels, establishing a feedback mechanism and generating a report, wherein the report comprises a chart and a graph, a list of key indexes and suggested actions.
Wherein the output of the model is converted to an understandable level of flatness. Such as a numerical score, rating (e.g., A, B, C levels), or descriptive rating (e.g., "very flat," "slightly uneven," "severely uneven").
Wherein a feedback mechanism is established to feed back the analysis results to the control system or operator for: adjusting subsequent production processes or parameters according to the flatness level; providing operators with real-time flatness status so that they take necessary corrective action; long-term trends are recorded and analyzed to continually improve process and product quality.
Wherein, automatically generating reports, summarizing detection results and comprehensive evaluation may include: a chart and a graph which intuitively show the flatness distribution; a list of key indicators, such as average flatness, standard deviation, number of abnormal areas; suggested actions are based on the detection results and the evaluation.
And S50, continuously optimizing the analysis model according to the detection result and feedback so as to improve the accuracy and reliability of the evaluation.
Through the steps, the detection result can be automatically analyzed, and comprehensive evaluation of the fabric flatness is given out according to a preset rule or model, so that the production process is optimized, and the product quality is ensured.
Example two
Based on the first embodiment, this embodiment provides a specific example, and a machine learning model is constructed using Scikit-learn library of Python for optimizing evaluation of fabric flatness. In this example, a random forest classifier will be used as the model, but the same procedure may be applied to other types of machine learning models. The following is a simplified code example of building and training a model:
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, recall_score, f1_score
from sklearn.model_selection import GridSearchCV
# suppose df is dataset DATAFRAME, which contains all necessary features and labels
#Df=pd.read_csv ('fabric_data.csv') # reads CSV file
The following is a hypothetical dataset for a presentation
data = {
'texture': [0.1, 0.2, 0.3, 0.4, 0.5],
'color': ['red', 'blue', 'green', 'yellow', 'purple'],
'reception_ratio': [0.9, 0.8, 0.7, 0.6, 0.5],
'environment': ['bright', 'dim', 'bright', 'dim', 'bright'],
'fabric_type': ['cotton', 'silk', 'wool', 'polyester', 'rayon'],
'flatness_level': [1, 2, 3, 2, 1]
}
df = pd.DataFrame(data)
# Data preprocessing to convert non-numeric features into numeric values
df['color'] = df['color'].astype('category').cat.codes
df['environment'] = df['environment'].astype('category').cat.codes
df['fabric_type'] = df['fabric_type'].astype('category').cat.codes
# Definition of features and target variables
X = df.drop('flatness_level', axis=1)
y = df['flatness_level']
# Dividing data set
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
Model training
rf_model = RandomForestClassifier(n_estimators=100, random_state=42)
Super parameter tuning
param_grid = {'n_estimators': [100, 200, 300], 'max_depth': [None, 10, 20]}
grid_search = GridSearchCV(rf_model, param_grid, cv=5, scoring='accuracy')
grid_search.fit(X_train, y_train)
# Best model
best_rf_model = grid_search.best_estimator_
# Prediction
y_pred = best_rf_model.predict(X_test)
Performance evaluation #
print("Accuracy:", accuracy_score(y_test, y_pred))
print("Recall:", recall_score(y_test, y_pred, average='weighted'))
print("F1 Score:", f1_score(y_test, y_pred, average='weighted'))
Of these, numpy, pandas and scikit-learn are libraries of Python, which can be installed by pips, such as PIP INSTALL numpy PANDAS SCIKIT-learn. In addition GRIDSEARCHCV is used for hyper-parametric tuning to find the best random forest model configuration.
Preferably, other machine learning models are also possible, such as support vector machines or neural networks.
Example III
Based on the first embodiment, the present embodiment shows the training steps of the analysis model:
step 1, data preparation
Characteristic engineering: key features are extracted from the detected data including, but not limited to, texture features, color features, statistics of the received scale data, distribution and severity of abnormal regions, etc.
Data cleaning: and cleaning the data, removing abnormal values and missing values, and ensuring the quality of the data.
Labeling: each record in the dataset is assigned a label for the smoothness evaluation, which may be a professional evaluation by an expert or an evaluation based on some criteria.
Step2 data set partitioning
The data set is divided into a training set, a validation set and a test set. Typical partitioning ratios may be 70% training set, 15% validation set, 15% test set.
Step 3, model selection
An appropriate model is selected based on the nature of the problem. For the flatness evaluation, classification models such as random forests, support Vector Machines (SVMs), gradient-lifted trees (GBDT), or deep learning models such as Convolutional Neural Networks (CNNs) are suitable. The present embodiment is preferably a random forest.
Step 4, model training
The selected model is trained using the training set data. During training, model parameters are adjusted to minimize training errors.
Model selection and super-parameter tuning are performed using the validation set data to prevent overfitting. This may include using a grid search (GRID SEARCH) or a random search (Randomized Search) to find the best combination of hyper-parameters.
Step 5, model evaluation
The generalization ability of the model was evaluated on the test set. Model performance is measured using evaluation metrics such as accuracy, recall, F1 score, etc.
Step 6, model optimization
Adjusting the model according to the test results may require repeating steps 4 and 5 until satisfactory model performance is obtained.
Step 7, model deployment
And after the model reaches the expected performance, the model is deployed into practical application and used for evaluating the flatness of the fabric in real time or in batches.
Step8, model maintenance and updating
The model is retrained periodically using new data to accommodate possibly varying fabric characteristics and detection environments, maintaining the effectiveness of the model.
Simplified code examples for training random forest classifiers using the Python and scikit-learn libraries are as follows:
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
let X be the feature matrix and y be the label vector
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
Creating random forest classifier instances
clf = RandomForestClassifier(n_estimators=100, random_state=42)
Training model #
clf.fit(X_train, y_train)
# Predictive test set
y_pred = clf.predict(X_test)
Calculation accuracy rate #
accuracy = accuracy_score(y_test, y_pred)
print(f"Model Accuracy: {accuracy}")
The above code demonstrates that the dataset is first segmented, then a random forest classifier is created, then a model is trained and its performance on the test set is evaluated.
The Random Forest is selected in the second and third embodiments because Random Forest (Random Forest) is a powerful machine learning algorithm and is particularly excellent in dealing with classification and regression problems. In the scene of fabric flatness evaluation, the reasons for adopting random forests mainly include the following points:
1. robustness and accuracy
The random forest can reduce the overfitting risk of a single decision tree by constructing a plurality of decision trees and summarizing the prediction results of the decision trees, thereby improving the accuracy and the stability of the model. Random forests can provide relatively stable predictions even if certain features or sample data are noisy or biased.
2. Feature importance assessment
Random forests can naturally evaluate the importance of features, which is important for understanding and optimizing feature engineering. In the fabric flatness evaluation, knowing which features (such as texture, color, receiving ratio, etc.) have the greatest effect on the final evaluation can help us optimize the data collection and preprocessing process.
3. Processing high-dimensional data
Random forests can effectively handle high-dimensional datasets without concern for dimension disasters. In fabric detection, a large number of features (e.g., multiple physical properties of the fabric, environmental conditions, etc.) may be involved, and random forests can cope well with this situation.
4. Parallel processing capability
The construction process of random forests can be parallelized, which means that model training can be accelerated in multi-core processors or distributed computing environments, which is particularly useful for processing large-scale data sets.
5. Easy to understand and explain
Although random forests are made up of multiple trees, each tree is based on decision tree principles, which makes the predictive logic of the model relatively easy to understand and interpret.
6. Strong generalization ability
Because random sampling and feature selection are adopted in the training process of the random forest, the method can effectively reduce overfitting and improve the generalization capability of the model. This means that the model can also maintain good predictive performance on new data that has not been seen.
Example IV
Based on the same conception, the application also provides fabric flatness detection equipment, which comprises:
The device comprises a transmitter and a receiver, wherein the transmitter emits light, the receiver receives the light passing through the fabric, noise filtering is carried out on the received signal, and meanwhile, the light intensity of the transmitter and/or the sensitivity of the receiver are adjusted according to the reflection characteristics of different fabrics;
Wherein both the transmitter and the receiver are prior art, such as the scheme mentioned in CN220552428U, the present application can be used or modified appropriately based on the hardware structure of the patent to use the scheme of the present application, such as increasing the number of transmitters and receivers.
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
the signal preprocessing module is used for calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation so as to consider the flatness difference of each part of the fabric and mark out the area possibly having abnormality;
The preliminary evaluation module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The optimization module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The secondary detection module is used for executing secondary detection if an area with the possibility of abnormality exists, so as to determine and calibrate a result and serve as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
the comprehensive evaluation module is used for analyzing the final result data, giving out comprehensive evaluation of the fabric flatness according to a preset rule and/or an analysis model, and displaying the detection result in real time;
and the output module is used for outputting the comprehensive evaluation and detection result.
Example five
This embodiment also provides an electronic device, referring to fig. 2, comprising a memory 404 and a processor 402, the memory 404 having stored therein a computer program, the processor 402 being arranged to run the computer program to perform the steps of any of the method embodiments described above.
In particular, the processor 402 may include a Central Processing Unit (CPU), or an application specific integrated circuit (ApplicationSpecificIntegratedCircuit, abbreviated as ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present application.
The memory 404 may include, among other things, mass storage 404 for data or instructions. By way of example, and not limitation, memory 404 may comprise a hard disk drive (HARDDISKDRIVE, abbreviated HDD), a floppy disk drive, a solid state drive (SolidStateDrive, abbreviated SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Memory 404 may include removable or non-removable (or fixed) media, where appropriate. Memory 404 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 404 is a Non-Volatile (Non-Volatile) memory. In particular embodiments, memory 404 includes Read-only memory (ROM) and Random Access Memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (ProgrammableRead-only memory, abbreviated PROM), an erasable PROM (ErasableProgrammableRead-only memory, abbreviated EPROM), an electrically erasable PROM (ElectricallyErasableProgrammableRead-only memory, abbreviated EEPROM), an electrically rewritable ROM (ElectricallyAlterableRead-only memory, abbreviated EAROM) or a FLASH memory (FLASH), or a combination of two or more of these. The RAM may be a static random access memory (StaticRandom-access memory, abbreviated SRAM) or a dynamic random access memory (DynamicRandomAccessMemory, abbreviated DRAM) where the DRAM may be a fast page mode dynamic random access memory 404 (FastPageModeDynamicRandomAccessMemory, abbreviated FPMDRAM), an extended data output dynamic random access memory (ExtendedDateOutDynamicRandomAccessMemory, abbreviated EDODRAM), a synchronous dynamic random access memory (SynchronousDynamicRandom-access memory, abbreviated SDRAM), or the like, where appropriate.
Memory 404 may be used to store or cache various data files that need to be processed and/or used for communication, as well as possible computer program instructions for execution by processor 402.
The processor 402 reads and executes the computer program instructions stored in the memory 404 to implement any of the fabric flatness detection methods of the above embodiments.
Optionally, the electronic apparatus may further include a transmission device 406 and an input/output device 408, where the transmission device 406 is connected to the processor 402 and the input/output device 408 is connected to the processor 402.
The transmission device 406 may be used to receive or transmit data via a network. Specific examples of the network described above may include a wired or wireless network provided by a communication provider of the electronic device. In one example, the transmission device includes a network adapter (Network Interface Controller, simply referred to as a NIC) that can connect to other network devices through the base station to communicate with the internet. In one example, the transmission device 406 may be a Radio Frequency (RF) module, which is configured to communicate with the internet wirelessly.
The input-output device 408 is used to input or output information.
Example six
The present embodiment also provides a readable storage medium having stored therein a computer program including program code for controlling a process to execute the process including the fabric flatness detection method according to the first embodiment.
It should be noted that, specific examples in this embodiment may refer to examples described in the foregoing embodiments and alternative implementations, and this embodiment is not repeated herein.
In general, the various embodiments may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects of the invention may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the invention is not limited thereto. While various aspects of the invention may be illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.
Embodiments of the invention may be implemented by computer software executable by a data processor of a mobile device, such as in a processor entity, or by hardware, or by a combination of software and hardware. Computer software or programs (also referred to as program products) including software routines, applets, and/or macros can be stored in any apparatus-readable data storage medium and they include program instructions for performing particular tasks. The computer program product may include one or more computer-executable components configured to perform embodiments when the program is run. The one or more computer-executable components may be at least one software code or a portion thereof. In addition, in this regard, it should be noted that any blocks of the logic flow may represent program steps, or interconnected logic circuits, blocks and functions, or a combination of program steps and logic circuits, blocks and functions. The software may be stored on physical media such as memory chips or memory blocks implemented within the processor, magnetic media such as hard or floppy disks, and optical media such as, for example, DVDs and data variants thereof, CDs, etc. The physical medium is a non-transitory medium.
It should be understood by those skilled in the art that the technical features of the above embodiments may be combined in any manner, and for brevity, all of the possible combinations of the technical features of the above embodiments are not described, however, they should be considered as being within the scope of the description provided herein, as long as there is no contradiction between the combinations of the technical features.
The foregoing examples illustrate only a few embodiments of the application, which are described in greater detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit of the application, which are within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (10)
1. The fabric flatness detection method is characterized by comprising the following steps of:
S00, emitting light through the emitter, receiving the light passing through the fabric through the receiver, noise filtering the received signal, and adjusting the light intensity of the emitter and/or the sensitivity of the receiver according to the reflection characteristics of different fabrics;
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
s10, calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation to consider flatness differences of all parts of the fabric and mark out areas possibly having abnormality;
s20, inputting the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
S30, if an area with the possibility of abnormality exists, performing secondary detection to determine and calibrate a result and taking the result as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
s40, analyzing final result data, giving comprehensive evaluation of fabric flatness according to a preset rule and/or an analysis model, and displaying detection results in real time.
2. The fabric flatness detection method of claim 1, wherein in the S00 step, the specific steps of adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the reflection characteristics of different fabrics are:
Before detection, initializing a transmitter and a receiver, and setting default light intensity and sensitivity;
in the detection process, the signal intensity received by the receiver is monitored in real time, and the signal intensity can reflect the reflection characteristic of the fabric on light;
analyzing the received signal strength and judging whether the received signal strength exceeds a preset dynamic range;
If not, the light intensity of the emitter is insufficient or the fabric absorbs excessive light, so that the light intensity of the emitter and/or the sensitivity of the receiver need to be improved; if yes, the light intensity of the emitter is too high or the fabric emits too much light, and the light intensity of the emitter and/or the sensitivity of the receiver need to be reduced;
Adjusting the light intensity of the transmitter and/or the sensitivity of the receiver based on the analysis result; wherein the magnitude of the adjustment is based on a predefined scaling factor to prevent over-adjustment of the parameter;
the adjusted parameters are applied to the transmitter and/or receiver and then the signal strength is re-monitored to form a closed loop control until the signal strength stabilizes within a preset dynamic range.
3. The fabric flatness detection method of claim 2, wherein in the S00 step, the specific step of adjusting the light intensity of the transmitter and the sensitivity of the receiver according to the reflection characteristics of different fabrics further comprises:
The optimal parameter settings under different fabric types and colors are stored for direct recall at the next detection.
4. The fabric flatness detection method of claim 1, wherein in step S10, a plurality of thresholds are set according to the light intensity distribution data to distinguish different flatness states as a preset flatness level standard;
for each set of measurement data, classifying it into different flatness classes according to its relationship with a set threshold;
carrying out weighted average on the flatness grade of each position of the detection area of the fabric;
mapping the flatness grade obtained by weighted average to a preset flatness grade standard;
and outputting the overall flatness grade of the fabric to reflect the average flatness level of the fabric.
5. The fabric flatness detection method of claim 1, wherein in step S20, the training step of the optimization model is as follows:
acquiring data containing flatness information of the fabric, wherein the data at least comprises an image of the fabric, receiving proportion data of a receiver and the type of the fabric;
Performing data cleaning to remove invalid or wrong data, and distributing a flatness grade label for each piece of cleaned data;
extracting features from the cleaned data, wherein the features at least comprise texture features, color features and statistical features of receiving proportion data of the fabric;
Dividing the data set to obtain a training set, a verification set and a test set;
Establishing a machine learning model, training the machine learning model by using a training set, and performing model selection and super-parameter tuning by using verification set data in the training process so as to prevent over-fitting;
and evaluating the generalization capability of the machine learning model by the test set until the machine learning model is trained, so as to obtain an optimized model.
6. The fabric flatness detection method of any one of claims 1-5, wherein in step S40, key features are extracted from the final result data, the key features including texture features of the fabric, reflection features of color light, statistical features of the reception ratio data, distribution and severity of abnormal areas;
Inputting the extracted key features into a preset rule and/or analysis model for evaluation to obtain final flatness evaluation;
the final flatness assessment is converted to a flatness level and a feedback mechanism is established and a report is generated, including charts and graphs, lists of key indicators, and suggested actions.
7. The fabric flatness detection method of claim 6, further comprising S50 of continuously optimizing the analytical model based on the detection result and feedback to improve accuracy and reliability of the evaluation.
8. The utility model provides a surface fabric roughness check out test set which characterized in that includes:
The device comprises a transmitter and a receiver, wherein the transmitter emits light, the receiver receives the light passing through the fabric, noise filtering is carried out on the received signal, and meanwhile, the light intensity of the transmitter and/or the sensitivity of the receiver are adjusted according to the reflection characteristics of different fabrics;
The emitter and the receiver cover the whole detection area of the fabric to completely acquire the light intensity distribution data of the surface of the fabric; the signal received by the receiver is receiving proportion data, and the receiving proportion data refers to the proportion between the light quantity received by the receiver and the light quantity sent by the transmitter;
the signal preprocessing module is used for calculating the preliminary flatness grade of the fabric by using a multi-threshold judgment and flattening weighting algorithm based on the light intensity distribution data, and taking the preliminary flatness grade as preliminary evaluation so as to consider the flatness difference of each part of the fabric and mark out the area possibly having abnormality;
The preliminary evaluation module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The optimization module inputs the result of the preliminary flatness grade and the receiving proportion data of the receiver into a pre-trained optimization model so as to optimize the accuracy of flatness evaluation and mark a region possibly having abnormality;
The secondary detection module is used for executing secondary detection if an area with the possibility of abnormality exists, so as to determine and calibrate a result and serve as final result data; if no abnormal area exists, the data output by the optimization model is used as final result data;
the comprehensive evaluation module is used for analyzing the final result data, giving out comprehensive evaluation of the fabric flatness according to a preset rule and/or an analysis model, and displaying the detection result in real time;
and the output module is used for outputting the comprehensive evaluation and detection result.
9. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the fabric flatness detection method of any one of claims 1 to 7.
10. A readable storage medium, characterized in that the readable storage medium has stored therein a computer program comprising program code for controlling a process to execute the process, the process comprising the fabric flatness detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410918848.6A CN118463864B (en) | 2024-07-10 | 2024-07-10 | Fabric flatness detection control method, equipment and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410918848.6A CN118463864B (en) | 2024-07-10 | 2024-07-10 | Fabric flatness detection control method, equipment and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118463864A true CN118463864A (en) | 2024-08-09 |
| CN118463864B CN118463864B (en) | 2024-10-01 |
Family
ID=92161002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410918848.6A Active CN118463864B (en) | 2024-07-10 | 2024-07-10 | Fabric flatness detection control method, equipment and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118463864B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118980339A (en) * | 2024-09-13 | 2024-11-19 | 常州市维意美嘉新材料有限公司 | Paper flatness detection system, method and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1040895A1 (en) * | 1982-01-27 | 1984-02-29 | Московский авиационный технологический институт им.К.Э.Циолковского | Method for measuring roughness of products |
| FR2725512A1 (en) * | 1994-10-07 | 1996-04-12 | Lorraine Laminage | Moving surface of device to be laminated flatness evaluating device esp. metal bands |
| JPH10170247A (en) * | 1996-12-11 | 1998-06-26 | Fukushima Pref Gov | Method and device for non-contact measurement of surface roughness |
| JP2002192213A (en) * | 2000-12-22 | 2002-07-10 | Abb Ab | Flatness control method and apparatus |
| CN110849845A (en) * | 2019-11-19 | 2020-02-28 | 佛山科学技术学院 | Device and method for on-line detection and control of micro-forming of composite microstructure array |
| CN210893027U (en) * | 2019-12-02 | 2020-06-30 | 联芯集成电路制造(厦门)有限公司 | System for detecting back flatness of wafer |
| DE102020112496A1 (en) * | 2020-05-08 | 2021-11-11 | Carl Zeiss Smt Gmbh | Method and device for inspecting the surface of an optical element |
| CN113932758A (en) * | 2021-09-15 | 2022-01-14 | 同济大学 | A kind of road surface smoothness prediction method and device |
| CN114210741A (en) * | 2021-11-23 | 2022-03-22 | 北京科技大学 | Device and method for on-line detection of surface roughness of strip steel |
| CN116086390A (en) * | 2023-03-20 | 2023-05-09 | 浙江星视界科技有限公司 | Flatness detection device and system for LED display screen |
| CN117470151A (en) * | 2023-11-09 | 2024-01-30 | 中国联合网络通信集团有限公司 | Paint film flatness detection methods, devices, equipment and storage media |
| CN220552428U (en) * | 2023-08-23 | 2024-03-01 | 浙江永顺窗饰材料有限公司 | Fabric flatness detection device |
| CN118241529A (en) * | 2024-03-20 | 2024-06-25 | 河北习畅检测技术有限公司 | A method for testing the smoothness of an intelligent road |
-
2024
- 2024-07-10 CN CN202410918848.6A patent/CN118463864B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1040895A1 (en) * | 1982-01-27 | 1984-02-29 | Московский авиационный технологический институт им.К.Э.Циолковского | Method for measuring roughness of products |
| FR2725512A1 (en) * | 1994-10-07 | 1996-04-12 | Lorraine Laminage | Moving surface of device to be laminated flatness evaluating device esp. metal bands |
| JPH10170247A (en) * | 1996-12-11 | 1998-06-26 | Fukushima Pref Gov | Method and device for non-contact measurement of surface roughness |
| JP2002192213A (en) * | 2000-12-22 | 2002-07-10 | Abb Ab | Flatness control method and apparatus |
| CN110849845A (en) * | 2019-11-19 | 2020-02-28 | 佛山科学技术学院 | Device and method for on-line detection and control of micro-forming of composite microstructure array |
| CN210893027U (en) * | 2019-12-02 | 2020-06-30 | 联芯集成电路制造(厦门)有限公司 | System for detecting back flatness of wafer |
| DE102020112496A1 (en) * | 2020-05-08 | 2021-11-11 | Carl Zeiss Smt Gmbh | Method and device for inspecting the surface of an optical element |
| CN113932758A (en) * | 2021-09-15 | 2022-01-14 | 同济大学 | A kind of road surface smoothness prediction method and device |
| CN114210741A (en) * | 2021-11-23 | 2022-03-22 | 北京科技大学 | Device and method for on-line detection of surface roughness of strip steel |
| CN116086390A (en) * | 2023-03-20 | 2023-05-09 | 浙江星视界科技有限公司 | Flatness detection device and system for LED display screen |
| CN220552428U (en) * | 2023-08-23 | 2024-03-01 | 浙江永顺窗饰材料有限公司 | Fabric flatness detection device |
| CN117470151A (en) * | 2023-11-09 | 2024-01-30 | 中国联合网络通信集团有限公司 | Paint film flatness detection methods, devices, equipment and storage media |
| CN118241529A (en) * | 2024-03-20 | 2024-06-25 | 河北习畅检测技术有限公司 | A method for testing the smoothness of an intelligent road |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118980339A (en) * | 2024-09-13 | 2024-11-19 | 常州市维意美嘉新材料有限公司 | Paper flatness detection system, method and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118463864B (en) | 2024-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117786373B (en) | Equipment operation diagnosis system based on big data corrugated paper processing | |
| CN118463864B (en) | Fabric flatness detection control method, equipment and application thereof | |
| EP3296822B1 (en) | Model-plant mismatch detection using model parameter data clustering for paper machines or other systems | |
| CN118396252B (en) | MES client data analysis optimization method based on cloud computing | |
| US11275362B2 (en) | Test time reduction for manufacturing processes by substituting a test parameter | |
| US12007745B2 (en) | Apparatus for predicting equipment damage | |
| US20180157225A1 (en) | Apparatus and method for automatic model identification from historical data for industrial process control and automation systems | |
| CN118014165B (en) | Traceability management method and traceability management system for lithium ion battery production | |
| CN118779643B (en) | Big data-based ink coating distribution monitoring system and method | |
| CN118482817B (en) | Digital textile equipment management method and system based on data analysis | |
| CN119322485A (en) | Manufacturing equipment monitoring method based on Internet of things | |
| CN118761689A (en) | Construction site quality management evaluation system based on fuzzy C-means clustering | |
| CN119168833A (en) | An intelligent supervision system and method for environmental monitoring quality control | |
| CN117977536A (en) | Smart power grid load identification method | |
| CN119090095B (en) | Loom output prediction method based on big data | |
| CN118469268B (en) | Digital factory consumable management system and method | |
| WO2014117139A1 (en) | Reducing product variations via variance partition analysis | |
| CN118714597B (en) | Wireless ad hoc network signal optimization method and system | |
| US20230216726A1 (en) | Monitoring of target system, such as communication network or industrial process | |
| CN117708622B (en) | Abnormal index analysis method and system of operation and maintenance system and electronic device | |
| US20240370012A1 (en) | Quality cause analysis system | |
| CN119578248A (en) | Method, device, equipment and medium for optimizing and recommending parameters of fabric shaping equipment traveling machine | |
| CN119830609A (en) | Rare earth extraction production flow simulation system and method | |
| Basit et al. | Predictive Maintenance Using Deep Learning: Enhancing Reliability and Reducing Electrical System Downtime | |
| CN119292209A (en) | Intelligent control method and system for vegetarian meat powder production line |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |