CN113418919A - Textile fiber component qualitative and quantitative online analysis system and method - Google Patents
Textile fiber component qualitative and quantitative online analysis system and method Download PDFInfo
- Publication number
- CN113418919A CN113418919A CN202110480355.5A CN202110480355A CN113418919A CN 113418919 A CN113418919 A CN 113418919A CN 202110480355 A CN202110480355 A CN 202110480355A CN 113418919 A CN113418919 A CN 113418919A
- Authority
- CN
- China
- Prior art keywords
- analysis
- unit
- qualitative
- fiber
- quantitative
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Treatment Of Fiber Materials (AREA)
Abstract
The invention relates to the technical field of textile fiber analysis, and discloses a textile fiber component qualitative and quantitative online analysis system which comprises a textile fiber processing unit, a fiber image acquisition unit, a communication unit, an AI component analysis unit and a result feedback unit. The invention also discloses a corresponding analysis method. The invention can reduce the requirements on the experience of inspectors, shorten the testing time and improve the accuracy of the testing result.
Description
Technical Field
The invention relates to the technical field of textile fiber analysis, in particular to a qualitative and quantitative online analysis system and method for textile fiber components.
Background
Textile fibers are basic constituent materials of textiles and clothing products, and the fiber content is an important inspection item for monitoring the quality of the textiles. The qualitative analysis of the fiber components is the primary step of fiber content detection, and the quantitative analysis can be carried out only after the fiber types in the textile yarns are qualitatively identified.
The commonly used quantitative analysis methods include a manual resolution method, a chemical dissolution method and a microscopic method, and the microscopic method mainly depends on the microscopic observation of the fiber morphology for species identification, counting and measurement. Therefore, qualitative techniques and quantitative techniques by microscopy of fiber components have been regarded as key operation techniques by textile testing practitioners. However, for a long time, qualitative analysis and quantitative analysis by microscopy can only adopt traditional manual operation, and depend too much on experience and judgment capability of technicians, especially quantitative analysis by microscopy, a test sample usually needs 2-3 hours, and meanwhile, hands, eyes and brain are tense and cooperate, so that not only is the efficiency low, but also the long-term operation can have adverse effects on occupational health.
The existing qualitative technology is mainly based on FZ/T01057.1-4-2007 part 1-4 of the identification test method of textile fibers, the specific category of the fibers is comprehensively judged by manual operation through observing test phenomena of a combustion method, a microscope method and a dissolution method and combining practical experience, and the specific operation flow is shown in figure 1.
The existing quantitative technology of the microscope method mainly comprises the steps of preparing sample slices according to the standards of GB/T16988-2013 'determination of content of a mixture of special animal fibers and sheep wool', 'FZ/T30003-2009' quantitative analysis method of a flax and cotton blended product microprojection method ', and' FZ/T01101-2008 'determination of textile fiber content' and the like, placing the prepared slices on a stage of a microscopic projector or a fiber fineness analyzer, moving a glass slide, and observing various fibers entering a visual field through an eyepiece or a screen. The type of the fiber is identified according to the morphological and structural characteristics of the fiber, the number of the fiber is respectively recorded, the diameter of the fiber is measured, and finally the content of each fiber is calculated through a formula. The whole test process requires special operation of a specially-assigned person, and other work tasks cannot be performed simultaneously. The specific test flow is shown in figure 2.
In the existing operation process, because the mastering degree of technicians on various fiber forms is different, part of chemical fibers can be accurately identified by combining a combustion method or a dissolution method. The technicians need training and experience accumulation for 3-6 months or even longer period for learning and mastering the morphological characteristics of various fibers, and the culture cost is high. The difference of the abilities of the operators causes that technical experts with abundant experience need to consult or seek help in actual detection, communication and verification are performed remotely through means such as screen capture, voice, video and the like, the efficiency is low, and correct results cannot be obtained due to the difference of information understanding or the limitation of the abilities of the persons seeking help.
In addition, in the quantitative process of the microscopy method, the type identification needs to be carried out by operators according to experience to select the type, the diameter measurement needs to be carried out by focusing clearly and then scribing is carried out, manual operation cannot be carried out in each step, the test time is long, and fatigue is easy to occur.
Aiming at the problems, system equipment for recognizing high-definition images of textile fibers through an AI image processing technology appears in the market, an automatic analysis scheme integrating image acquisition, recognition, measurement, statistics and calculation is realized through an automatic microscope and artificial intelligence software, and the problems that the testing process depends on personnel and experience and the testing efficiency is low are solved.
However, for the laboratory demander, not only the investment of the equipment is huge, but also the operator is trained professionally, the measurement result is different due to different levels of the operator, and for some samples which are difficult to process through AI, the effective detection cannot be obtained.
Disclosure of Invention
In order to solve the problems, the invention provides a qualitative and quantitative online analysis system and method for textile fiber components, which reduce the requirements on the experience of inspectors, shorten the test time and improve the accuracy of test results.
The technical scheme adopted by the invention is as follows:
a textile fiber component qualitative and quantitative online analysis system is characterized by comprising a textile fiber processing unit, a fiber image acquisition unit, a communication unit, an AI component analysis unit and a result feedback unit, wherein a demand party carries out fiber raveling and slicing through the textile fiber processing unit, then sends the fiber raveling and slicing to the fiber image acquisition unit for image acquisition according to the regulations, sends the acquired image to the AI component analysis unit through the communication unit, and the AI component analysis unit analyzes the image and feeds the analysis result back to the demand party through the result feedback unit.
The online analysis system further comprises an expert auditing unit, wherein during qualitative analysis, the AI component analysis unit judges the images, and when the type prediction probability obtained by the AI component analysis unit is smaller than a preset value, the expert auditing unit analyzes the acquired images.
A qualitative and quantitative online analysis method for textile fiber components is characterized by comprising the following steps:
(1) the demand party carries out the fiber raveling and slicing through a textile fiber processing unit;
(2) acquiring images of the slices by a fiber image acquisition unit according to the specification;
(3) sending the collected image to an AI component analysis unit through a communication unit;
(4) the AI component analysis unit analyzes the image;
(5) and feeding back the analysis result to the demand side through a result feedback unit.
Further, in the step (4), during qualitative analysis, when the type prediction probability obtained by the AI component analysis unit is smaller than a predetermined value, the acquired image is analyzed by the expert auditing unit.
Further, the analysis process of the expert audit unit includes an operation of requiring the demander to repeat steps (2) - (4).
Furthermore, the analysis method is qualitative analysis, the analysis process of the expert auditing unit further comprises the step that a demanding party obtains the chemical characteristic information of the textile fibers by a chemical method of a combustion test or a dissolution test and sends the chemical characteristic information to the expert auditing unit by the communication unit, and the expert auditing unit performs analysis by combining the image information and the chemical characteristic information to obtain a qualitative result.
Further, the analysis method is qualitative analysis, and the fiber image acquisition unit focuses on longitudinal morphology and surface features of the images acquired by the slices.
Further, the analysis method is quantitative analysis, and the fiber image acquisition unit focuses on the number and geometric characteristics of fibers in a section of the acquired image.
Further, the step (5) further comprises a step of saving the analysis result in a background database.
The invention has the beneficial effects that:
(1) the culture period and the culture cost of testers are reduced;
(2) the testing error opportunity is effectively reduced, the labor intensity is reduced, and the management cost is reduced;
(3) the tester can conveniently communicate with deep experts in the industry directly for learning, and the rapid growth of the technical ability of the tester is facilitated;
(4) the equipment investment cost is reduced;
(5) the test time is shortened, and the efficiency is improved by more than 100%.
Drawings
FIG. 1 is a flow diagram of a prior art manual qualitative analysis;
FIG. 2 is a flow chart of prior art manual microscope quantitative analysis;
FIG. 3 is a block diagram of an analysis system according to the present invention;
FIG. 4 is a flow chart of a qualitative test of textile fibers;
FIG. 5 is a flow chart of a quantitative test of textile fibers.
Detailed Description
The following describes in detail specific embodiments of the qualitative and quantitative online analysis system and method for textile fiber components according to the present invention with reference to the accompanying drawings.
Referring to the attached figure 3, the textile fiber component qualitative and quantitative online analysis system comprises a textile fiber processing unit, a fiber image acquisition unit, a communication unit, an AI component analysis unit and a result feedback unit, all the units work cooperatively, a demand party sends fiber to the fiber image acquisition unit for image acquisition according to the regulations after the fiber is raveled and sliced by the textile fiber processing unit, the acquired image is sent to the AI component analysis unit by the communication unit, the AI component analysis unit analyzes the image, and the analysis result is fed back to the demand party by the result feedback unit.
Textile fiber processing unit and fibre image acquisition unit are located the demand side, and the demand side is each detecting element or research institution etc. and textile fiber processing unit is the unit module of handling textile fiber sample, and operating personnel can adopt traditional cutting equipment and method to carry out the section processing to textile fiber sample, and the demand according to qualitative or quantitative determination is different, carries out different processings to textile fiber. The fiber image acquisition unit adopts equipment such as a high-definition industrial camera, a large depth-of-field microscope and the like to complete image acquisition on the slices. The slice processing and image acquisition parameters may be defined according to predetermined rules to make the AI component analysis result more desirable.
The AI component analysis unit is arranged on a testing side and mainly comprises an image recognition technology, and artificial intelligence software is used for automatically recognizing, measuring and counting the fiber images. The analysis of the slice image can be performed by referring to related equipment in the prior art or directly borrowing the recognition function of the related equipment. The AI component analysis unit also has learning ability, and can perform more intelligent identification and analysis on the known slice images, so that the test result is more ideal.
The communication unit can adopt an internet communication protocol, so that the photographing equipment of the demand party is directly interconnected with the AI component analysis unit of the test party, images photographed by other equipment can also be transmitted to the test party through a network, and the test party transmits the images to the AI component analysis unit in a leading-in mode and the like. And after the AI component analysis unit completes the analysis, the result is fed back to the demand party through the result feedback unit.
The AI component analysis unit is in an automatic process of image data processing, so that the quality of image shooting directly influences the detection result, and the parameters of the image such as size, resolution and the like are appointed in the image acquisition process to ensure the reliability of the detection result.
During qualitative analysis, for the condition that the quality of a slice image is not high or a shot slice is determined to be difficult to carry out AI analysis, an AI component analysis unit firstly carries out prediction on the class probability of the image and presets a probability value, and when the class prediction probability obtained by the AI component analysis unit is smaller than a preset value, the acquired image can be analyzed by means of a configured expert auditing unit. The expert auditing unit consists of qualified experts in the industry and detection operators, has rich detection experience, can require the demander to slice and take pictures again to overcome the basic defects in the early stage, is consulted by the experts, and can complete detection analysis by combining physicochemical operating parameters of the demander.
The invention is further illustrated below by means of a qualitative analysis process and a quantitative analysis process, respectively, of the textile fibre composition.
Referring to fig. 4, the qualitative analysis process is performed by image analysis to determine the type of components in the textile fibers. The specific process is as follows:
(1) the demand party carries out the fiber raveling and slicing through a textile fiber processing unit;
(2) acquiring images of the slices by a fiber image acquisition unit according to the specification;
(3) sending the collected image to an AI component analysis unit through a communication unit;
(4) the AI component analysis unit analyzes the image;
(5) the AI component analysis unit obtains the type prediction probability, and when the type prediction probability is smaller than a preset value, the acquired image is analyzed through the expert auditing unit;
(6) and feeding back the analysis result to the demand side through a result feedback unit.
The requesting party may be required to repeat the operations of steps (2) - (4) while the expert reviews the analysis of the unit, the captured image being focused on longitudinal morphology and surface features. The demanding party can also be required to obtain the chemical characteristic information of the textile fibers by a chemical method and send the information to the expert auditing unit through the communication unit. The chemical method comprises a combustion test and a dissolution test, wherein the traditional chemical method is used for reference, related data are transmitted to an expert auditing unit according to requirements, and consultation is carried out by matching with an expert group.
Referring to fig. 5, for the quantitative analysis process, the composition ratio of the components in the textile fiber is determined by image analysis, typically after the qualitative analysis is completed. The specific process is as follows:
(1) the demand party carries out the fiber raveling and slicing through a textile fiber processing unit;
(2) acquiring images of the slices by a fiber image acquisition unit according to the specification;
(3) sending the collected image to an AI component analysis unit through a communication unit;
(4) the AI component analysis unit analyzes the image;
(5) and feeding back the analysis result to the demand side through a result feedback unit.
The requesting party may be required to repeat the operations of steps (2) - (4) while the expert reviews the analysis of the unit, the acquired image being focused on the number of fibers and the geometric characteristics of the section.
The AI component analysis unit is independent of each demand side, can accept the detection demands of different units, can acquire a larger number of slice images, and enables the AI component analysis unit to identify the detection result to be closer to the actual value on the basis of learning. The expert team can process difficult and complicated cases in a centralized way, so that the demander uses high-quality resources with little investment. Through the wireless communication technology, the data transmission and detection can be completed across regions, and the detection processing of the whole day is realized.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (9)
1. The utility model provides a textile fiber composition qualitative and quantitative on-line analysis system which characterized in that: the automatic fiber opening and cutting device comprises a textile fiber processing unit, a fiber image acquisition unit, a communication unit, an AI component analysis unit and a result feedback unit, wherein a demand party opens fibers through the textile fiber processing unit and slices, then sends the fibers to the fiber image acquisition unit for image acquisition according to the regulations, sends acquired images to the AI component analysis unit through the communication unit, and the AI component analysis unit analyzes the images and feeds analysis results back to the demand party through the result feedback unit.
2. The qualitative and quantitative on-line analysis system of textile fiber composition according to claim 1, characterized in that: the online analysis system further comprises an expert auditing unit, wherein during qualitative analysis, the AI component analysis unit judges the images, and when the AI component analysis unit obtains that the type prediction probability is smaller than a preset value, the expert auditing unit analyzes the acquired images.
3. A textile fiber component qualitative and quantitative online analysis method is characterized in that: the method comprises the following steps:
(1) the demand party carries out the fiber raveling and slicing through a textile fiber processing unit;
(2) acquiring images of the slices by a fiber image acquisition unit according to the specification;
(3) sending the collected image to an AI component analysis unit through a communication unit;
(4) the AI component analysis unit analyzes the image;
(5) and feeding back the analysis result to the demand side through a result feedback unit.
4. A method for qualitative and quantitative on-line analysis of the composition of textile fibres according to claim 3, characterized in that: in the step (4), during qualitative analysis, when the type prediction probability obtained by the AI component analysis unit is smaller than a predetermined value, the acquired image is analyzed by the expert auditing unit.
5. The method of qualitative and quantitative on-line analysis of the composition of textile fibers according to claim 4, characterized in that: the analysis process of the expert audit unit includes an operation of requiring the demander to repeat steps (2) - (4).
6. The method of qualitative and quantitative on-line analysis of the composition of textile fibers according to claim 5, characterized in that: the analysis method is qualitative analysis, the analysis process of the expert auditing unit further comprises the step that a demanding party obtains chemical characteristic information of the textile fibers through a chemical method of a combustion test or a dissolution test and sends the chemical characteristic information to the expert auditing unit through the communication unit, and the expert auditing unit performs analysis by combining image information and the chemical characteristic information to obtain a qualitative result.
7. The method for qualitative, quantitative, on-line analysis of the composition of textile fibers according to any of claims 3 to 5, characterized in that: the analysis method is qualitative analysis, and the fiber image acquisition unit focuses on longitudinal morphology and surface features of the images acquired by the slices.
8. The method for qualitative, quantitative, on-line analysis of the composition of textile fibers according to any of claims 3 to 5, characterized in that: the analysis method is quantitative analysis, and the fiber image acquisition unit focuses on the number and geometric characteristics of fibers in a section of the acquired image.
9. The method for qualitative, quantitative, on-line analysis of the composition of textile fibers according to any of claims 3 to 5, characterized in that: the step (5) further comprises the step of storing the analysis result in a background database.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480355.5A CN113418919A (en) | 2021-04-30 | 2021-04-30 | Textile fiber component qualitative and quantitative online analysis system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480355.5A CN113418919A (en) | 2021-04-30 | 2021-04-30 | Textile fiber component qualitative and quantitative online analysis system and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113418919A true CN113418919A (en) | 2021-09-21 |
Family
ID=77712016
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110480355.5A Pending CN113418919A (en) | 2021-04-30 | 2021-04-30 | Textile fiber component qualitative and quantitative online analysis system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113418919A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114414523A (en) * | 2021-10-25 | 2022-04-29 | 上海布眼人工智能科技有限公司 | Textile fiber component qualitative method based on automatic waveband selection |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102628759A (en) * | 2012-03-26 | 2012-08-08 | 中纺标(北京)检验认证中心有限公司 | Preparation and detection method of textile fiber digitized slice and microscopic examination simulation method |
CN102912613A (en) * | 2012-10-31 | 2013-02-06 | 江南大学 | Embedded collecting, processing and displaying method for fiber amplified images |
CN107561078A (en) * | 2017-10-17 | 2018-01-09 | 北京和众视野科技有限公司 | Textile fiber detection method |
CN108519382A (en) * | 2018-03-20 | 2018-09-11 | 北京和众视野科技有限公司 | The evaluation method of textile fabric recognition result |
CN209764816U (en) * | 2018-12-28 | 2019-12-10 | 仲恺农业工程学院 | Banana disease online identification system based on artificial intelligence |
CN111982908A (en) * | 2020-09-07 | 2020-11-24 | 中纺标(深圳)检测有限公司 | Fiber automatic analysis equipment based on visual analysis technology and analysis method thereof |
CN112304229A (en) * | 2019-08-02 | 2021-02-02 | 武汉舜立软件有限公司 | Automatic analysis method and system for textile fiber components |
-
2021
- 2021-04-30 CN CN202110480355.5A patent/CN113418919A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102628759A (en) * | 2012-03-26 | 2012-08-08 | 中纺标(北京)检验认证中心有限公司 | Preparation and detection method of textile fiber digitized slice and microscopic examination simulation method |
CN102912613A (en) * | 2012-10-31 | 2013-02-06 | 江南大学 | Embedded collecting, processing and displaying method for fiber amplified images |
CN107561078A (en) * | 2017-10-17 | 2018-01-09 | 北京和众视野科技有限公司 | Textile fiber detection method |
CN108519382A (en) * | 2018-03-20 | 2018-09-11 | 北京和众视野科技有限公司 | The evaluation method of textile fabric recognition result |
CN209764816U (en) * | 2018-12-28 | 2019-12-10 | 仲恺农业工程学院 | Banana disease online identification system based on artificial intelligence |
CN112304229A (en) * | 2019-08-02 | 2021-02-02 | 武汉舜立软件有限公司 | Automatic analysis method and system for textile fiber components |
CN111982908A (en) * | 2020-09-07 | 2020-11-24 | 中纺标(深圳)检测有限公司 | Fiber automatic analysis equipment based on visual analysis technology and analysis method thereof |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114414523A (en) * | 2021-10-25 | 2022-04-29 | 上海布眼人工智能科技有限公司 | Textile fiber component qualitative method based on automatic waveband selection |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2222554C (en) | Method and apparatus for continuously monitoring and forecasting slide and specimen preparation for a biological specimen population | |
CN111722714A (en) | Digital substation metering operation inspection auxiliary method based on AR technology | |
EP1706724B1 (en) | Method and apparatus for testing fibres | |
CN104459172B (en) | A kind of automatization's sample of bone marrow processing means and automated analysis, diagosis method | |
CA3030226A1 (en) | System and method for combined automatic and manual inspection | |
CN104256882B (en) | Based on reconstituted tobacco ratio measuring method in the pipe tobacco of computer vision | |
CN114062366B (en) | Online detection method and system for winnowing stem-removing quality in tobacco shred production | |
CN108414409B (en) | System and method for collecting spectrum sheet image of rotary ferrograph | |
CN108627520A (en) | A kind of on-line detecting system and method for heterogeneous solid material presentation quality | |
CN107024478A (en) | A kind of part processes automatic defect identifying system | |
CN102662048A (en) | Method and device for analyzing urine | |
CN110738630A (en) | Training method and detection system of recursive deep learning system | |
DE112019005951T5 (en) | Centralized analysis of multiple visual inspection devices | |
CN116128881B (en) | Lens detection system and method for defect product synchronous marking | |
CN112964732A (en) | Spinning cake defect visual detection system and method based on deep learning | |
CN111896513A (en) | Big data statistical analysis system for residual detection of toxic and harmful substances | |
CN113418919A (en) | Textile fiber component qualitative and quantitative online analysis system and method | |
CN112712514A (en) | Product detection system and auxiliary detection method | |
CN108875777B (en) | Method for identifying fiber types and blending proportion in textile fabric based on double-path neural network | |
CN111665243B (en) | Textile fiber identification and component detection system | |
Baron et al. | Evaluation of the Magiscan image analyzer for asbestos fiber counting | |
CN211699002U (en) | Automatic identification and sorting device for floor surface plates | |
CN102103090A (en) | Device and method for detecting quality of web of carding machine on line based on computer vision | |
CN212321459U (en) | Automatic image identification equipment of fibre | |
CN213456679U (en) | Automatic fiber analysis equipment based on visual analysis technology |
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 |