CN117705754A - Textile polyester fiber content online detection method based on hyperspectral imaging - Google Patents
Textile polyester fiber content online detection method based on hyperspectral imaging Download PDFInfo
- Publication number
- CN117705754A CN117705754A CN202311632307.9A CN202311632307A CN117705754A CN 117705754 A CN117705754 A CN 117705754A CN 202311632307 A CN202311632307 A CN 202311632307A CN 117705754 A CN117705754 A CN 117705754A
- Authority
- CN
- China
- Prior art keywords
- polyester fiber
- textile
- pixel points
- fiber content
- value
- 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
- 239000000835 fiber Substances 0.000 title claims abstract description 82
- 239000004753 textile Substances 0.000 title claims abstract description 82
- 229920000728 polyester Polymers 0.000 title claims abstract description 81
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- 238000000701 chemical imaging Methods 0.000 title claims abstract description 34
- 230000003595 spectral effect Effects 0.000 claims abstract description 95
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000002310 reflectometry Methods 0.000 claims abstract description 23
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 12
- 238000001228 spectrum Methods 0.000 claims description 50
- 238000009499 grossing Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 229920002334 Spandex Polymers 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003960 organic solvent Substances 0.000 description 1
- 239000004759 spandex Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to the technical field of textile component content detection, and discloses a textile polyester fiber content online detection method based on hyperspectral imaging, which comprises intermittently collecting infrared spectrum of a textile in an infrared band of 1000-1700nmAn image; characteristic interception is carried out on the infrared spectrogram image, and an ROI area in the infrared spectrogram image is determined; scanning and analyzing all pixel points in the ROI area, extracting a spectral reflectance value in the band range of each pixel point, and determining a spectral reflectance average value of the ROI area; the average value of the spectral reflectivity is used for deriving the wavelength to obtain the second derivative R of the spectral reflectivity of the reflectivity to the wavelength 2 The method comprises the steps of carrying out a first treatment on the surface of the By R 2 As input and output, the content of the polyester fiber is used for constructing an intelligent prediction model of the content of the polyester fiber based on a neural network; leading the trained prediction model into a detection system, and inputting a second derivative R of the spectral reflectivity 2 Obtaining the polyester fiber content. According to the invention, the ROI area can be intelligently selected, and the content of the polyester fiber can be accurately detected.
Description
Technical Field
The invention relates to the technical field of textile component content detection, in particular to a textile polyester fiber content online detection method based on hyperspectral imaging.
Background
In textile detection, detection means and instruments are required to have the characteristics of accuracy, rapidness and environmental protection, so that the requirements of the market can be met. The existing methods for detecting the content of the textile products are characterized in that after the qualitative detection of the blended products, the detected textile products are dissolved by reagents such as strong acid, organic solvent and the like, so that samples are destroyed, the time consumption is long, the efficiency is low, the online detection cannot be carried out on a large scale, the detection is inaccurate, the error is large, and the detection requirements of quality and trade are difficult to meet.
In the actual identification process, the existing detection system does not cut and separate the invalid region in the textile fabric spectral image, so that the system cannot accurately identify on line and cannot select the intelligent ROI region.
Therefore, how to provide a textile polyester fiber content online detection method based on hyperspectral imaging, which can intelligently select an ROI region and has high detection precision and high efficiency, is a problem to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the invention provides an online detection method for the polyester fiber content of a textile based on hyperspectral imaging, which aims to solve the problems that the existing infrared identification technology cannot intelligently select proper regions from ROI regions for spectral feature extraction and cannot accurately detect the polyester fiber content.
In order to solve the technical problems, the invention adopts the following technical scheme:
a method for detecting the content of polyester fibers in textiles based on hyperspectral imaging on line comprises the following steps:
intermittently collecting infrared spectrogram images of the textile in an infrared band of 1000-1700 nm;
characteristic interception is carried out on the infrared spectrogram image, and an ROI area in the infrared spectrogram image is determined;
scanning and analyzing all pixel points in the ROI area, extracting a spectral reflectance value of each pixel point in a wave band range of 1000-1700nm, and determining a spectral reflectance average value R of the ROI area;
the average value R of the spectral reflectivity is used for deriving the wavelength to obtain the second derivative R of the spectral reflectivity of the reflectivity to the wavelength 2 ；
By R 2 As input, the polyester fiber content Q is output, and an intelligent polyester fiber content prediction model based on a neural network is constructed;
leading the trained prediction model into a detection system, and taking the second derivative R of the spectral reflectance of the textile 2 For input, the polyester fiber content of the textile being tested was measured.
Preferably, in the above-mentioned method for detecting the polyester fiber content of textiles based on hyperspectral imaging, the method for intermittently collecting infrared spectrogram images of textiles in the infrared band of 1000-1700nm comprises:
the preset spectral resolution is 9nm, and the spectral range is 1000nm-1700nm;
and collecting infrared spectrum information of the textile on the conveyor belt through a hyperspectral camera, and storing the infrared spectrum information once every 1s to obtain a plurality of infrared spectrum images.
Preferably, in the above method for online detecting the polyester fiber content of a textile based on hyperspectral imaging, the feature interception is performed on the infrared spectrogram, and determining the ROI area in the infrared spectrogram comprises:
analyzing the infrared spectrogram images to obtain a plurality of first spectral image areas, wherein the first spectral image areas comprise areas occupied by each textile in the infrared spectrogram images;
and carrying out feature interception on the plurality of first spectrum image areas to determine the ROI area in the first spectrum image areas.
Preferably, in the above method for online detecting the polyester fiber content of a textile based on hyperspectral imaging, the method for analyzing a plurality of infrared spectrograms to obtain a plurality of first spectral image areas includes:
edge detection is carried out on a plurality of infrared spectrogram images, so that a first area occupied by each textile in the infrared spectrogram images within 1s is obtained;
and cutting the first area occupied by each textile in the infrared spectrum image to obtain a plurality of first spectrum image areas.
Preferably, in the above-mentioned method for online detecting the polyester fiber content of textile based on hyperspectral imaging, the method for determining the ROI area in the first spectral image area includes:
scanning and analyzing the first spectrum image area, detecting gray values of pixel points in the first spectrum image area, integrating the gray values of the pixel points in the same spectrum image area, and generating a gray value record set;
comparing the gray values in the gray value record set to determine the maximum gray value in the gray value record set;
screening the corresponding pixel points in the first spectrum image area by taking a first proportion value smaller than the maximum gray value as a screening condition, and identifying the screened pixel points as pixel points to be removed;
and if the plurality of pixel points to be removed meet a preset aggregation screening rule, determining the area to be removed, which is mapped by the pixel points to be removed, in the first spectrum image area as an invalid area, and removing the invalid area in the first spectrum image area to obtain the ROI area.
Preferably, in the above-mentioned method for online detecting the content of polyester fibers in textiles based on hyperspectral imaging, the method for removing the invalid region in the first spectral image region to obtain the ROI region includes:
uniformly dividing a first spectrum image area into a plurality of first blocks, and determining pixel points in the first blocks according to the total pixel points in the first spectrum image area;
scanning and analyzing the pixel points in the first blocks to obtain a plurality of first pixel points with gray values larger than 60% of the maximum gray value, calculating the proportion of all the first pixel points in each first block to the total pixel points in the first spectrum image area to obtain a second proportion value, and limiting the second proportion value to be smaller than a preset proportion value;
scanning a plurality of pixel points in a first block, detecting gray values, integrating the gray values of the pixel points in the same first block, generating a plurality of gray value data sets, and associating with the first block;
calculating the gray average value of all pixel points in the first block, marking the gray average value as a block gray expression value, and if the block gray expression value is smaller than 60% of the maximum gray value and the difference value between the block gray expression value and the maximum gray value is larger than a preset difference value, marking the corresponding first block to be removed;
and removing the first block marked with the to-be-removed area in the first spectrum image area to obtain the ROI area.
Preferably, in the above-mentioned method for online detecting the polyester fiber content of textile based on hyperspectral imaging, the method for determining the average value R of the spectral reflectance of the ROI area includes:
acquiring an ROI (region of interest) of each textile in an infrared spectrogram image;
scanning all pixel points in the ROI to obtain a corresponding relation between the spectral reflectance value and the wavelength, wherein the corresponding relation between the spectral reflectance value and the wavelength comprises a plurality of spectral reflectance values and corresponding wavelengths in the ROI;
extracting spectral reflectance values of the pixel points in the wavelength range of 1000-1700nm, and averaging the spectral reflectance values in the wavelength range of 1000-1700nm to obtain the spectral average reflectance R of the ROI.
Preferably, in the above-mentioned method for online detecting the content of polyester fibers in textiles based on hyperspectral imaging, the average value R of the spectral reflectance is derived from the wavelength to obtain the second derivative R of the reflectance with respect to the wavelength 2 The method of (1) comprises:
performing SG smoothing on the spectrum average reflectivity R to obtain a smoothed spectrum reflectivity R 1 ；
Will smooth the spectral reflectance R 1 Obtaining the second derivative R of the reflectivity with respect to the wavelength by obtaining the second derivative R of the wavelength with respect to the wavelength 2 。
Preferably, in the above-mentioned method for detecting the polyester fiber content of the textile based on hyperspectral imaging, the method for detecting the polyester fiber content of the textile to be detected comprises the following steps:
taking 600 textiles with different polyester fiber contents as training samples, collecting the average spectral reflectance R, smoothing SG, and smoothing the average spectral reflectance R 1 Obtaining R by obtaining second derivative 2 ；
R obtained with 600 training samples 2 The Levenberg-Marquardt algorithm is adopted as input to select 10 layers of the neural network and output the polyester fiber content Q to obtain the second derivative R of the spectrum average reflectivity 2 A neural network-based intelligent prediction model of the polyester fiber content with the polyester fiber content Q;
leading the trained intelligent prediction model based on the polyester fiber content of the neural network into an identification system, and inputting R obtained by extraction and calculation of infrared spectrogram images 2 And outputting and obtaining the polyester fiber content Q.
The invention provides a hyperspectral imaging-based on-line detection method for the content of textile polyester fibers, which has the beneficial effects that compared with the prior art, the method is as follows:
(1) The online detection method for the polyester fiber content of the textile based on hyperspectral imaging realizes the intelligent selection of the ROI area of the textile in the infrared spectrogram based on hyperspectral imaging, predicts the polyester fiber content Q based on the intelligent polyester fiber content prediction model of the neural network, realizes the online accurate detection of the polyester fiber content of the textile, has high detection precision and high efficiency, and solves the problems that the prior infrared identification technology cannot intelligently select the ROI area and cannot accurately detect the polyester fiber content;
(2) The method for detecting the content of the textile polyester fiber on line based on hyperspectral imaging is simple, high in accuracy and high in treatment efficiency, ensures the accuracy and reliability of detecting the content of the textile polyester fiber, does not need manual intervention in the whole process, can realize fully-automatic operation, and reduces labor cost.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for on-line detection of the polyester fiber content of a textile based on hyperspectral imaging;
FIG. 2 is a schematic diagram of a method for on-line detection of the polyester fiber content of a textile based on hyperspectral imaging;
FIG. 3 is a spectral image of a textile acquired by a hyperspectral camera provided by the present invention;
fig. 4 is a plot of spectral reflectance versus wavelength extracted from the textile of fig. 3 at different regions.
Detailed Description
The following describes in further detail the embodiments of the present invention with reference to the drawings and examples. The following examples are illustrative of the invention and are not intended to limit the scope of the invention.
In the description of the present application, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate description of the present application and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present application.
The terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present application, unless otherwise indicated, the meaning of "a plurality" is two or more.
In the description of the present application, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the terms in this application will be understood by those of ordinary skill in the art in a specific context.
As shown in fig. 1-4, the invention provides an on-line detection method for the content of textile polyester fiber based on hyperspectral imaging, which comprises the following steps:
s100: and intermittently collecting infrared spectrogram images of the textiles in the infrared band of 1000-1700nm, wherein the infrared spectrogram images are collected by a hyperspectral camera, so that the information of each textile in the infrared spectrogram images can be collected quickly and efficiently.
S200: the infrared spectral image is subjected to feature interception, the ROI area in the infrared spectral image is determined, the infrared spectral image often contains a large amount of information, key parts in the image can be highlighted through feature interception, the complexity of data processing can be reduced by selecting the interested ROI area, the processing time and the calculated amount can be reduced by intelligently selecting the ROI area, and the processing efficiency is greatly improved.
S300: and scanning and analyzing all pixel points in the ROI area, extracting a spectral reflectance value of each pixel point in a wave band range of 1000-1700nm, determining a spectral reflectance average value R of the ROI area, and reducing the influence of noise and improving the precision of a processing result by utilizing the spectral reflectance average value R of the ROI.
S400: the average value R of the spectral reflectivity is used for deriving the wavelength to obtain the second derivative R of the spectral reflectivity of the reflectivity to the wavelength 2 By the second derivative method, baseline shift of the spectral reflectance data can be eliminated, resolution of the spectral reflectance data can be increased, and noise can be effectively reduced and suppressed.
S500: by R 2 The intelligent prediction model for the polyester fiber content based on the neural network is constructed by taking the polyester fiber content Q as an output, so that the accuracy of predicting the polyester fiber content Q can be improved, and the adaptability is high.
S600: leading the trained intelligent prediction model based on the polyester fiber content of the neural network into an identification system, and taking the second derivative R of the spectral reflectance of the textile 2 For inputting, the polyester fiber content Q of the textile to be detected is detected, wherein the components of the textile to be detected are 97% polyester fiber and 3% spandex, so that the treatment efficiency and the treatment precision are greatly improved, the whole process does not need manual intervention, the full-automatic operation can be realized, and the labor cost is reduced.
In some embodiments of the present invention, the method for intermittently acquiring infrared spectral images of a textile in the infrared band of 1000-1700nm in step S100 comprises:
s110: the preset spectral resolution is 9nm, and the spectral range is 1000nm-1700nm;
s120: and collecting infrared spectrum information of the textile on the conveyor belt through a hyperspectral camera, and storing the infrared spectrum information once every 1s to obtain a plurality of infrared spectrum images.
In some embodiments of the present invention, the method for performing feature extraction on an infrared spectral image in step S200 to determine an ROI area in the infrared spectral image includes:
s210: analyzing the infrared spectrogram images to obtain a plurality of first spectral image areas, wherein the first spectral image areas comprise areas occupied by each textile in the infrared spectrogram images;
s211: performing edge detection on a plurality of infrared spectrogram images to obtain a first area occupied by each textile in the infrared spectrogram images within 1s, wherein the edge detection method is performed based on a gradient algorithm, as image information obtained by a hyperspectral camera is gray value information, gradient can be directly calculated, a threshold range is set, and a part exceeding the threshold is taken as an edge and then cut to obtain a spectral image of the first area;
s212: cutting a first area occupied by each textile in the infrared spectrogram image to obtain a plurality of first spectral image areas;
s220: feature interception is carried out on a plurality of first spectrum image areas, and ROI areas in the first spectrum image areas are determined;
s221: scanning and analyzing the first spectrum image area, detecting gray values of pixel points in the first spectrum image area, integrating the gray values of the pixel points in the same spectrum image area, and generating a gray value record set;
s222: comparing the gray values in the gray value record set to determine the maximum gray value in the gray value record set;
s223: screening the corresponding pixel points in the first spectrum image area by taking a first proportion value smaller than the maximum gray value as a screening condition, and identifying the screened pixel points as pixel points to be removed;
s224: and if the plurality of pixel points to be removed meet a preset aggregation screening rule, determining the area to be removed, which is mapped by the pixel points to be removed, in the first spectrum image area as an invalid area, and removing the invalid area in the first spectrum image area to obtain the ROI area.
In some embodiments of the present invention, the method for removing the invalid region from the first spectral image region in step S224 to obtain the ROI region includes:
s225: the first spectrum image area is uniformly divided into a plurality of first blocks, and pixel points in the first blocks are determined according to the total pixel points in the first spectrum image area, so that the operation process is reduced, and the processing efficiency is improved;
s226: scanning and analyzing the pixel points in the first blocks to obtain a plurality of first pixel points with gray values larger than 60% of the maximum gray value, calculating the proportion of all the first pixel points in each first block to the total pixel points in the first spectrum image area to obtain a second proportion value, limiting the second proportion value to be smaller than a preset proportion value, and limiting the size of each first block so that each first block cannot be excessively large, and facilitating subsequent calculation;
s227: scanning pixel points in a plurality of first blocks, detecting gray values, integrating the gray values of the pixel points in the same first block, generating a plurality of gray value data sets, correlating the gray value data sets with the first blocks, and facilitating intelligent extraction of the first blocks corresponding to the gray value data sets, so that the processing efficiency is improved;
s228: calculating the gray average value of all pixel points in the first block, marking the first block to be removed if the gray average value of the first block is less than 60% of the maximum gray value and the difference value between the gray average value and the maximum gray value is greater than the preset difference value, and reserving the first block corresponding to the difference value between the gray average value of the first block and the 60% of the maximum gray value which is less than the preset difference value to effectively mark the block to be removed;
s229: and removing the first block marked with the to-be-removed area in the first spectrum image area to obtain the ROI area, namely intelligently selecting the ROI area, and improving the processing efficiency.
In some embodiments of the present invention, the method of determining the spectral reflectance average R of the ROI region in step S300 includes:
s310: acquiring an ROI (region of interest) of each textile in an infrared spectrogram image;
s320: scanning all pixel points in the ROI to obtain a corresponding relation between the spectral reflectance value and the wavelength, wherein the corresponding relation between the spectral reflectance value and the wavelength comprises a plurality of spectral reflectance values and corresponding wavelengths in the ROI;
s330: extracting spectral reflectance values of the pixel points in the wavelength range of 1000-1700nm, and averaging the spectral reflectance values in the wavelength range of 1000-1700nm to obtain the spectral average reflectance R of the ROI.
In some embodiments of the present invention, the average value of the spectral reflectance R is derived from the wavelength in step S400 to obtain the second derivative R of the reflectance with respect to the wavelength 2 The method of (1) comprises:
s410: performing SG smoothing on the spectrum average reflectivity R to obtain a smoothed spectrum reflectivity R 1 The SG smoothing processing can filter noise while keeping the shape and the width of the signal unchanged, and can inhibit the noise by a moving average method, so that the information extraction precision is improved, and meanwhile, most of fine characteristics of infrared spectrogram images can be reserved, and the accuracy and the stability of the spectrum average reflectivity R can be effectively improved, so that the method can be better applied to analysis and application based on spectrums;
s420: will smooth the spectral reflectance R 1 Obtaining the second derivative R of the reflectivity with respect to the wavelength by obtaining the second derivative R of the wavelength with respect to the wavelength 2 The method can highlight the subtle changes in the spectrum curve, better extract the characteristics of the textile, effectively extract useful information and reduce the influence of noise, thereby deeply understanding the inherent characteristics of the spectrum data and providing valuable references for subsequent analysis.
In some embodiments of the present invention, constructing the neural network-based intelligent prediction model of the polyester fiber content in step S500 includes:
s510: taking 600 textiles with different polyester fiber contents as training samples, collecting the average spectral reflectance R, smoothing SG, and smoothing the average spectral reflectance R 1 Obtaining R by obtaining second derivative 2 ；
S520: r obtained with 600 training samples 2 The Levenberg-Marquardt algorithm is adopted as input to select 10 layers of the neural network and output the polyester fiber content Q to obtain the second derivative R of the spectrum average reflectivity 2 With polyestersThe intelligent prediction model of the polyester fiber content based on the neural network with the fiber content Q can accurately predict the polyester fiber content through training of a large amount of data, and has strong applicability.
The working principle of the textile polyester fiber content online detection method based on hyperspectral imaging is as follows: firstly intermittently collecting infrared spectral images of textiles in the infrared band of 1000-1700nm, determining first spectral image areas based on the infrared spectral images, scanning and analyzing all pixel points in each first spectral image area to obtain a gray value record set, determining the maximum gray value, uniformly dividing each first spectral image area into a plurality of first blocks, scanning and analyzing all pixel points in each first block to obtain the gray average value of all pixel points in each first block, recording the gray average value as a block gray expression value, if the gray average value of the blocks is smaller than 60% of the maximum gray value and the difference value between the gray average value and the gray average value is larger than a preset difference value, marking the corresponding first blocks to be removed, removing the first blocks to be removed to obtain an ROI area, scanning all pixel points in the ROI area to obtain a plurality of spectral reflectance values in the wavelength range of 1000-1700nm and corresponding wavelengths, obtaining the spectral average reflectance R of the ROI area, deriving the spectral average reflectance R to the second order of the wavelength to obtain the derivative reflectance R of the wavelength 2 R is taken as 2 And inputting the trained intelligent prediction model of the polyester fiber content based on the neural network to obtain the polyester fiber content Q, and effectively solving the problems that the ROI area cannot be intelligently selected for spectral feature extraction and the polyester fiber content cannot be accurately detected in the existing infrared identification technology.
According to the embodiment, the invention provides the textile polyester fiber content online detection method based on hyperspectral imaging, which can intelligently select the ROI, has high polyester fiber content detection precision and high treatment efficiency.
In summary, the hyperspectral imaging-based online detection method for the polyester fiber content of the textile can not only realize rapid and efficient intelligent selection of interested ROI (region of interest) and reduce processing time and calculation amount, greatly improve processing efficiency, but also predict the polyester fiber content Q based on the intelligent prediction model of the polyester fiber content of a neural network, realize online accurate detection of the polyester fiber content of the textile, reduce the influence of noise, improve the precision of a processing result, eliminate manual intervention in the whole process, realize fully-automatic operation and reduce labor cost.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives, and variations may be made in the above embodiments by those skilled in the art without departing from the spirit and principles of the invention.
Claims (9)
1. The method for detecting the content of the textile polyester fiber on line based on hyperspectral imaging is characterized by comprising the following steps of:
intermittently collecting infrared spectrogram images of the textile in an infrared band of 1000-1700 nm;
characteristic interception is carried out on the infrared spectrogram image, and an ROI area in the infrared spectrogram image is determined;
scanning and analyzing all pixel points in the ROI area, extracting a spectral reflectance value of each pixel point in a wave band range of 1000-1700nm, and determining a spectral reflectance average value R of the ROI area;
the average value R of the spectral reflectivity is used for deriving the wavelength to obtain the reflectivitySecond derivative of spectral reflectance R with respect to wavelength 2 ；
By R 2 As input, the polyester fiber content Q is output, and an intelligent polyester fiber content prediction model based on a neural network is constructed;
leading the trained prediction model into a detection system, and taking the second derivative R of the spectral reflectance of the textile 2 For input, the polyester fiber content of the textile being tested was measured.
2. The method for detecting the polyester fiber content of the textile based on hyperspectral imaging according to claim 1, wherein the method for intermittently collecting the infrared spectrogram image of the textile in the infrared band of 1000-1700nm comprises the following steps:
the preset spectral resolution is 9nm, and the spectral range is 1000nm-1700nm;
and collecting infrared spectrum information of the textile on the conveyor belt through a hyperspectral camera, and storing the infrared spectrum information once every 1s to obtain a plurality of infrared spectrum images.
3. The method for online detection of the polyester fiber content of the textile based on hyperspectral imaging according to claim 1, wherein the feature interception of the infrared spectrogram image and the determination of the ROI area in the infrared spectrogram image comprise:
analyzing the infrared spectrogram images to obtain a plurality of first spectral image areas, wherein the first spectral image areas comprise areas occupied by each textile in the infrared spectrogram images;
and carrying out feature interception on the plurality of first spectrum image areas to determine the ROI area in the first spectrum image areas.
4. The method for online detection of the polyester fiber content of the textile based on hyperspectral imaging according to claim 3, wherein the method for analyzing the infrared spectrograms to obtain the first spectrogram areas comprises the following steps:
edge detection is carried out on a plurality of infrared spectrogram images, so that a first area occupied by each textile in the infrared spectrogram images within 1s is obtained;
and cutting the first area occupied by each textile in the infrared spectrum image to obtain a plurality of first spectrum image areas.
5. A method for online detection of textile polyester fiber content based on hyperspectral imaging as claimed in claim 3 wherein the method for determining ROI areas in the first spectral image area comprises:
scanning and analyzing the first spectrum image area, detecting gray values of pixel points in the first spectrum image area, integrating the gray values of the pixel points in the same spectrum image area, and generating a gray value record set;
comparing the gray values in the gray value record set to determine the maximum gray value in the gray value record set;
screening the corresponding pixel points in the first spectrum image area by taking a first proportion value smaller than the maximum gray value as a screening condition, and identifying the screened pixel points as pixel points to be removed;
and if the plurality of pixel points to be removed meet a preset aggregation screening rule, determining the area to be removed, which is mapped by the pixel points to be removed, in the first spectrum image area as an invalid area, and removing the invalid area in the first spectrum image area to obtain the ROI area.
6. The method for online detection of the polyester fiber content of the textile based on hyperspectral imaging according to claim 5, wherein the method for removing the invalid region in the first spectral image region to obtain the ROI region comprises the following steps:
uniformly dividing a first spectrum image area into a plurality of first blocks, and determining pixel points in the first blocks according to the total pixel points in the first spectrum image area;
scanning and analyzing the pixel points in the first blocks to obtain a plurality of first pixel points with gray values larger than 60% of the maximum gray value, calculating the proportion of all the first pixel points in each first block to the total pixel points in the first spectrum image area to obtain a second proportion value, and limiting the second proportion value to be smaller than a preset proportion value;
scanning a plurality of pixel points in a first block, detecting gray values, integrating the gray values of the pixel points in the same first block, generating a plurality of gray value data sets, and associating with the first block;
calculating the gray average value of all pixel points in the first block, marking the gray average value as a block gray expression value, and if the block gray expression value is smaller than 60% of the maximum gray value and the difference value between the block gray expression value and the maximum gray value is larger than a preset difference value, marking the corresponding first block to be removed;
and removing the first block marked with the to-be-removed area in the first spectrum image area to obtain the ROI area.
7. The method for on-line detection of the polyester fiber content of textiles based on hyperspectral imaging according to claim 1, wherein the method for determining the average value R of the spectral reflectance of the ROI area comprises:
acquiring an ROI (region of interest) of each textile in an infrared spectrogram image;
scanning all pixel points in the ROI to obtain a corresponding relation between the spectral reflectance value and the wavelength, wherein the corresponding relation between the spectral reflectance value and the wavelength comprises a plurality of spectral reflectance values and corresponding wavelengths in the ROI;
extracting spectral reflectance values of the pixel points in the wavelength range of 1000-1700nm, and averaging the spectral reflectance values in the wavelength range of 1000-1700nm to obtain the spectral average reflectance R of the ROI.
8. The method for on-line detection of the polyester fiber content of the textile based on hyperspectral imaging as claimed in claim 1, wherein the average value R of the spectral reflectance is derived from the wavelength to obtain the second derivative R of the reflectance with respect to the wavelength 2 The method of (1) comprises:
performing SG smoothing on the spectrum average reflectivity R to obtain smoothingSpectral reflectance R 1 ；
Will smooth the spectral reflectance R 1 Obtaining the second derivative R of the reflectivity with respect to the wavelength by obtaining the second derivative R of the wavelength with respect to the wavelength 2 。
9. The method for detecting the polyester fiber content of the textile based on hyperspectral imaging according to claim 1, wherein the method for detecting the polyester fiber content of the textile to be detected comprises the following steps:
taking 600 textiles with different polyester fiber contents as training samples, collecting the average spectral reflectance R, smoothing SG, and smoothing the average spectral reflectance R 1 Obtaining R by obtaining second derivative 2 ；
R obtained with 600 training samples 2 The Levenberg-Marquardt algorithm is adopted as input to select 10 layers of the neural network and output the polyester fiber content Q to obtain the second derivative R of the spectrum average reflectivity 2 A neural network-based intelligent prediction model of the polyester fiber content with the polyester fiber content Q;
leading the trained intelligent prediction model based on the polyester fiber content of the neural network into an identification system, and inputting R obtained by extraction and calculation of infrared spectrogram images 2 And outputting and obtaining the polyester fiber content Q.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311632307.9A CN117705754A (en) | 2023-11-30 | 2023-11-30 | Textile polyester fiber content online detection method based on hyperspectral imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311632307.9A CN117705754A (en) | 2023-11-30 | 2023-11-30 | Textile polyester fiber content online detection method based on hyperspectral imaging |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117705754A true CN117705754A (en) | 2024-03-15 |
Family
ID=90145332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311632307.9A Pending CN117705754A (en) | 2023-11-30 | 2023-11-30 | Textile polyester fiber content online detection method based on hyperspectral imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117705754A (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101256670A (en) * | 2008-03-20 | 2008-09-03 | 华南师范大学 | Method and apparatus for three-dimensional visualization of sequence image |
| CN103674864A (en) * | 2013-11-12 | 2014-03-26 | 浙江大学 | Fish water content distribution detection method based on hyperspectral imaging technology |
| CN105021617A (en) * | 2015-07-14 | 2015-11-04 | 华中农业大学 | Hyperspectral imaging-based measuring equipment of chlorophyll content of whole rice plant and method thereof |
| CN106226247A (en) * | 2016-07-15 | 2016-12-14 | 暨南大学 | A kind of cell detection method based on EO-1 hyperion micro-imaging technique |
| US20170124386A1 (en) * | 2015-10-30 | 2017-05-04 | Xiaomi Inc. | Method, device and computer-readable medium for region recognition |
| CN108872091A (en) * | 2018-03-20 | 2018-11-23 | 浙江理工大学 | A kind of detection method of the vegetable pesticide residue concentration based on high light spectrum image-forming |
| US20200034972A1 (en) * | 2018-07-25 | 2020-01-30 | Boe Technology Group Co., Ltd. | Image segmentation method and device, computer device and non-volatile storage medium |
| CN110766679A (en) * | 2019-10-25 | 2020-02-07 | 普联技术有限公司 | Lens contamination detection method and device and terminal equipment |
| CN113390797A (en) * | 2021-07-20 | 2021-09-14 | 中国热带农业科学院湛江实验站 | Method for estimating total phosphorus content of pineapple leaves based on hyperspectral data |
| CN114863294A (en) * | 2022-05-10 | 2022-08-05 | 巴州新矿测绘有限责任公司 | Water quality parameter monitoring method and device suitable for brackish water lake |
| KR20230049378A (en) * | 2021-10-06 | 2023-04-13 | 한국식품연구원 | Imaging device for diagnosing soft rot of chinese cabbage based on non-destructive analysis technology and method for manufacturing the same |
| CN115998246A (en) * | 2021-10-22 | 2023-04-25 | 南京微纳科技研究院有限公司 | Control method, device, equipment, medium and program product for imaging system |
| TWI810103B (en) * | 2022-09-07 | 2023-07-21 | 大陸商環旭電子股份有限公司 | Method, system, storage medium and intelligent terminal for detecting wafers on tray |
| US11727669B1 (en) * | 2022-03-17 | 2023-08-15 | Institute of Facility Agriculture, Guangdong Academy of Agricultural Science | Intelligent recognition method of hyperspectral image of parasites in raw fish |
| CN116664565A (en) * | 2023-07-28 | 2023-08-29 | 江苏森标科技有限公司 | Hidden crack detection method and system for photovoltaic solar cell |
| CN117110157A (en) * | 2023-02-14 | 2023-11-24 | 中国科学院西安光学精密机械研究所 | Turbid water body total suspended particulate matter inversion method suitable for stationary orbit satellite GOCI |
-
2023
- 2023-11-30 CN CN202311632307.9A patent/CN117705754A/en active Pending
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101256670A (en) * | 2008-03-20 | 2008-09-03 | 华南师范大学 | Method and apparatus for three-dimensional visualization of sequence image |
| CN103674864A (en) * | 2013-11-12 | 2014-03-26 | 浙江大学 | Fish water content distribution detection method based on hyperspectral imaging technology |
| CN105021617A (en) * | 2015-07-14 | 2015-11-04 | 华中农业大学 | Hyperspectral imaging-based measuring equipment of chlorophyll content of whole rice plant and method thereof |
| US20170124386A1 (en) * | 2015-10-30 | 2017-05-04 | Xiaomi Inc. | Method, device and computer-readable medium for region recognition |
| CN106226247A (en) * | 2016-07-15 | 2016-12-14 | 暨南大学 | A kind of cell detection method based on EO-1 hyperion micro-imaging technique |
| CN108872091A (en) * | 2018-03-20 | 2018-11-23 | 浙江理工大学 | A kind of detection method of the vegetable pesticide residue concentration based on high light spectrum image-forming |
| US20200034972A1 (en) * | 2018-07-25 | 2020-01-30 | Boe Technology Group Co., Ltd. | Image segmentation method and device, computer device and non-volatile storage medium |
| CN110766679A (en) * | 2019-10-25 | 2020-02-07 | 普联技术有限公司 | Lens contamination detection method and device and terminal equipment |
| CN113390797A (en) * | 2021-07-20 | 2021-09-14 | 中国热带农业科学院湛江实验站 | Method for estimating total phosphorus content of pineapple leaves based on hyperspectral data |
| KR20230049378A (en) * | 2021-10-06 | 2023-04-13 | 한국식품연구원 | Imaging device for diagnosing soft rot of chinese cabbage based on non-destructive analysis technology and method for manufacturing the same |
| CN115998246A (en) * | 2021-10-22 | 2023-04-25 | 南京微纳科技研究院有限公司 | Control method, device, equipment, medium and program product for imaging system |
| US11727669B1 (en) * | 2022-03-17 | 2023-08-15 | Institute of Facility Agriculture, Guangdong Academy of Agricultural Science | Intelligent recognition method of hyperspectral image of parasites in raw fish |
| CN114863294A (en) * | 2022-05-10 | 2022-08-05 | 巴州新矿测绘有限责任公司 | Water quality parameter monitoring method and device suitable for brackish water lake |
| TWI810103B (en) * | 2022-09-07 | 2023-07-21 | 大陸商環旭電子股份有限公司 | Method, system, storage medium and intelligent terminal for detecting wafers on tray |
| CN117110157A (en) * | 2023-02-14 | 2023-11-24 | 中国科学院西安光学精密机械研究所 | Turbid water body total suspended particulate matter inversion method suitable for stationary orbit satellite GOCI |
| CN116664565A (en) * | 2023-07-28 | 2023-08-29 | 江苏森标科技有限公司 | Hidden crack detection method and system for photovoltaic solar cell |
Non-Patent Citations (4)
| Title |
|---|
| DAIKOS, OLESYA ET AL.: "In-line monitoring of the residual moisture in impregnated black textile fabrics by hyperspectral imaging", PROGRESS IN ORGANIC COATINGS, vol. 163, 28 February 2022 (2022-02-28) * |
| 叶军立: "小麦植株表型性状高通量获取与分析***关键技术研究", 中国博士学位论文全文数据库 农业科技辑, no. 02, 15 February 2023 (2023-02-15), pages 12 - 13 * |
| 朱梦远;杨红兵;李志伟;: "高光谱图像和叶绿素含量的水稻纹枯病早期检测识别", 光谱学与光谱分析, no. 06, 15 June 2019 (2019-06-15) * |
| 胡晟煜 等: "基于特征波长积分吸收光谱法的 H2S 浓度监测", 激光与光电子学进展, 13 March 2023 (2023-03-13) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097034B (en) | Intelligent face health degree identification and evaluation method | |
| EP3662484B1 (en) | Automatic assay assessment and normalization for image processing | |
| CN107665492B (en) | Colorectal panoramic digital pathological image tissue segmentation method based on depth network | |
| EP1470411B1 (en) | Method for quantitative video-microscopy and associated system and computer software program product | |
| EP1484595B1 (en) | Color space transformations for use in identifying objects of interest in biological specimens | |
| US10761003B2 (en) | Method and system for analyzing cuttings coming from a wellbore | |
| CN114549522A (en) | Textile quality detection method based on target detection | |
| WO1981003080A1 (en) | Method and apparatus for measuring mean cell volume of red blood cells | |
| CN108682008A (en) | A kind of Leukocyte Image clarity evaluation method and device | |
| JPH08504522A (en) | Method and apparatus for identifying objects using a regular sequence of boundary pixel parameters | |
| CN116228764B (en) | Neonate disease screening blood sheet acquisition quality detection method and system | |
| CN111160451A (en) | Flexible material detection method and storage medium thereof | |
| CN113109240B (en) | Method and system for determining imperfect grains of grains implemented by computer | |
| CN109285140A (en) | A kind of printed circuit board image registration appraisal procedure | |
| JPH0418267B2 (en) | ||
| CN106780347A (en) | A kind of loquat early stage bruise discrimination method based on OCT image treatment | |
| CN114184599B (en) | Single-cell Raman spectrum acquisition number estimation method, data processing method and device | |
| Tu et al. | AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds | |
| CN106204523A (en) | A kind of image quality evaluation method and device | |
| CN117541763A (en) | Tea fermentation degree determination method and system based on image recognition | |
| CN117705754A (en) | Textile polyester fiber content online detection method based on hyperspectral imaging | |
| JP6246978B2 (en) | Method for detecting and quantifying fibrosis | |
| JPH10185911A (en) | Device for analyzing cell and method therefor | |
| CN113450383B (en) | Quantitative analysis method, device, equipment and medium of immunochromatographic test paper | |
| CN115656063A (en) | Target pollutant content analysis method and device based on hyperspectral data |
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 |