CN114877810B - Cotton fiber length detection device based on machine vision and application method - Google Patents
Cotton fiber length detection device based on machine vision and application method Download PDFInfo
- Publication number
- CN114877810B CN114877810B CN202210642908.7A CN202210642908A CN114877810B CN 114877810 B CN114877810 B CN 114877810B CN 202210642908 A CN202210642908 A CN 202210642908A CN 114877810 B CN114877810 B CN 114877810B
- Authority
- CN
- China
- Prior art keywords
- cotton fiber
- cotton
- sampling
- sample
- module
- 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.)
- Active
Links
Classifications
-
- 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
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- 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
-
- 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
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/50—Reuse, recycling or recovery technologies
- Y02W30/66—Disintegrating fibre-containing textile articles to obtain fibres for re-use
Abstract
The application relates to a cotton fiber length detection technology, in particular to a cotton fiber length detection device based on machine vision, which mainly comprises a machine case, wherein a cotton fiber image acquisition module, a cotton fiber sample brushing module, a cotton fiber sampling module, a sliding table module and a central control module are arranged in the machine case; the application method based on the cotton fiber length detection device based on machine vision uses a central control module to process and analyze corresponding data algorithms, and mainly comprises the following steps: preprocessing an image to obtain a gray value normalization result of each pixel point of the image; obtaining a length-frequency distribution curve through programming software, establishing a length-frequency distribution curve model by adopting a machine learning technology, and respectively calculating the value of each length quality index of the cotton sample according to the model. The application is economical, practical and convenient to popularize and apply, and the machine vision method is provided, so that the length of the cotton fiber can be automatically and rapidly detected, the repeatability of the detection length index result is high, and the stability of the detection result is high.
Description
Technical Field
The application relates to a cotton fiber length detection technology, in particular to a cotton fiber length detection device based on machine vision and an application method, and belongs to the field of cotton quality detection mechanical equipment and application.
Background
Cotton is a production material for national folk life. In the cotton production and processing process, quality detection and classification are indispensable, and have a critical influence on cotton distribution quality of spinning enterprises. The quality of cotton is detected and graded, so that the purposes of promoting effective utilization of resources and improving quality and efficiency can be achieved. The existing testing instrument comprehensively reflects the quality of cotton mainly through the detection indexes of length, strength, fineness, color signs and impurity content. The indexes for measuring the cotton length mainly comprise main body length, quality length, average length, short fiber rate, uniformity and the like. The longer the body length, the lower the flock rate and the higher the uniformity, the more advantageous the spinning. Since 1940, the theory of sampling and measuring the length of cotton fibers by Hertel has been widely implemented in various countries, however, the method of sampling and measuring the length of cotton fibers based on the theory has been pointed out that there are defects, which cannot determine some statistical indexes of cotton fibers, and have not been solved yet.
HVI high capacity testers based on the sampling theory of Hertel have become the mainstay for detecting cotton fiber length. For many years, in order to meet the detection speed requirement of the cotton textile industry, the cotton fiber length testing instrument is updated continuously, and the automatic and informationized degree is improved continuously from the original manual mechanical type to the later digital electronic type, to the present intelligent type, and the detection time is shorter and shorter. According to the detection method of the cotton fiber length test instrument, the cotton fiber length test instrument can be classified into three types, namely a root-by-root test type, a grouping test type and a non-grouping test type.
The Y111 type roller length analyzer and the Y121 type comb sheet type length analyzer are representative of mechanical cotton fiber length detectors. The Y111 roller length analyzer belongs to a group test type according to a detection method, but the consistency of test results is poor due to manual operation, fibers are easy to lose in the use process, and the stability depends on the proficiency of operators. The Y121 type comb-sheet length instrument belongs to group test type, and the detection principle is as follows: firstly, a steel needle comb sheet is utilized to finish a sample into cotton bundles with one end flush, then fibers in the cotton bundles are divided into a plurality of groups according to a certain length from long to short, the weights of the groups are respectively weighed, and finally, each length index of the cotton fibers is obtained. The two mechanical cotton fiber length detectors are basically eliminated.
Fiber length cameras were then developed in the period of the rapid development of semiconductor technology in the last century, classified according to the detection method, which were not of the group test type, 6E ginned cotton fiber cameras were the earliest developed one for detecting cotton fiber length, after which us' ukt developed digital fiber cameras of types 530 and 730. The test principle of the photo camera is as follows: after a part of light emitted by an optical slit in the instrument passes through the fiber and is absorbed, the rest part is received by a photoelectric tube, the light intensity is converted into current, the current is input into a computer to obtain a histogram of fiber number and fiber length distribution, an accurate photographic curve is drawn, and then the fiber length and fiber quantity accumulation distribution map is obtained through secondary accumulation, so that various length indexes are calculated. Although the photographic method can finish the length detection of the cotton fiber in a short time, certain limitations exist, that is, only a specific drawing method can be used for obtaining a plurality of specific length indexes such as the average length of the upper half part, the span length and the like, the short staple index in the cotton fiber can not be provided for cotton spinning enterprises, and literature indicates that the photographic curve obtained by the method has limited precision, and more students criticize the error of the basic theory.
After the development of the photographic apparatus, a high-integration high-capacity cotton fiber tester is also developed, and the photographic apparatus based on the Hertel theory is adopted by the length detection module. Foreign large-capacity cotton fiber testers are introduced from domestic enterprises and cotton fiber inspection stations in the last 80 th century, HVI 1000 introduced by Uster corporation in 2004 is introduced in a large amount to date, and the XJ128 large-capacity cotton fiber testers developed by domestic long-term spinning are successfully manufactured by prototype test in 2010, and the performance of the high-capacity cotton fiber testers is the same as that of foreign equipment. The rapid high-capacity fiber detector can not detect the short-staple index in cotton fiber, and can detect the short-staple index only root by root or by adopting a better method for accurately detecting the short-staple index. The conventional cotton fiber root-by-root testing instrument is mainly an Uster AFIS (Advanced Fiber Information System) single fiber rapid testing system, and the AFIS can test the length and diameter data of single cotton fiber one by utilizing infrared light. However, the AFIS is time-consuming and labor-consuming to manufacture cotton sliver, and the requirement of quick inspection cannot be met. In addition, the detailed length distribution condition of the cotton fibers can be obtained by combining the cotton fiber embedding test method of the computer space three-dimensional reconstruction technology, the whole test flow comprises the steps of preparing embedding blocks, cutting cross sections into slices, taking cross section images and reconstructing, and the sample test time is more than one half hour.
In order to solve the problem that the artificial factors have great influence in the purchasing link of cotton length detection by using an experienced master to judge, the rapid detection of the cotton quality by developing and applying a new theory and a new method is urgently needed. Of note, in recent years, cotton fiber length detection theory and methodology in combination with computer image processing techniques is at the corner of the new.
Disclosure of Invention
The application aims to provide the cotton fiber length detection device based on machine vision, which is economical, practical, convenient to popularize and apply, and capable of detecting the cotton fiber length by using a machine vision method, effectively solving the problems of instability and incapability of repeatability of long-term working of a single-channel measurement cotton fiber length of a light intensity type photoelectric detection system and realizing automatic and rapid detection.
The second purpose of the application is to provide a method which is simple and convenient to operate, economical and practical, is not influenced by fluctuation of the light flux in a small range under the condition of non-ideal sampling, and improves the detection efficiency; the method has the advantages of high detection speed, high repeatability of detection length index results, high stability of detection results, low application cost and high universality, and can realize automatic and rapid detection of the cotton fiber length detection device based on machine vision, so as to solve the problems in the background technology.
In order to achieve the above purpose, the present application provides the following technical solutions:
the cotton fiber length detection device based on machine vision mainly comprises a case, wherein a cotton fiber image acquisition module, a cotton fiber sample brushing module, a cotton fiber sampling module, a sliding table module and a central control module are mainly arranged in the case; the case shell is provided with an observation port, a lofting port I and a lofting port II;
the cotton fiber sampling module is used for sampling cotton fibers of a sample and mainly comprises two groups of sampling rollers.
The cotton fiber sample brushing module is used for brushing sample to sample cotton fibers, and the structure of the cotton fiber sample brushing module mainly comprises a sample brushing bracket, wherein a sample laying-out platform and an industrial brush are arranged on the sample brushing bracket.
The cotton fiber image acquisition module is used for carrying out image sampling on the cotton fibers of the sample and uploading image data information to the central control module; the structure mainly comprises a bracket, a bottom plate, a transverse small bracket, a sampling camera, an image acquisition table and a reflecting panel, wherein the bracket is arranged above the bottom plate, the sampling camera is arranged above the bracket, the image acquisition table is arranged on the bottom plate and is opposite to the position of the camera, the image acquisition table is composed of a box body with a square light source, an air suction box body is arranged at the rear side of the image acquisition table, an air suction opening is arranged at the air suction box body, an opening of the air suction opening faces the image acquisition table, and an air suction pipeline is arranged at the rear side of the air suction box body; in the cotton fiber image acquisition module, the cotton fiber air suction box body is provided with an electric push rod II, and the electric push rod II can drive the air suction box body to move forward and backward to carry out straight treatment before photographing cotton fibers. The small transverse bracket is provided with a hinge shaft, the light reflecting panel is provided with a supporting arm, and the supporting arm is connected with the small transverse bracket through the hinge shaft.
The central control module is used for carrying out data algorithm processing analysis on the cotton sample image information obtained by the cotton fiber image obtaining module and uploading the data algorithm processing analysis.
Preferably, each group of sampling roller structure comprises a sampling roller shell, a sampling port, a central rotating shaft, a round hole arc sampling plate, a cotton sample pressing plate, a cylinder, a link mechanism and a servo motor; the cylinder is installed in the sampling cylinder shell and is connected with the connecting rod mechanism, the cotton sample pressing plate is installed in the sampling cylinder, one end of the cotton sample pressing plate is flexibly connected with the sampling cylinder, the flexible connection comprises an elastic rubber sheet and a rubber band, and the other end of the cotton sample pressing plate is propped against the inner wall of the sampling cylinder and can be lifted or slide up and down under the drive of the connecting rod mechanism. Preferably, the sample brushing support in the cotton fiber sample brushing module is provided with a sample setting platform, an industrial brush, a motor, an electric push rod I and an air suction pipe, wherein the industrial brush is close to the sample setting platform, the air suction pipe is arranged at the rear side of the industrial brush, and the electric push rod is arranged on the sample brushing support at the lower part of the industrial brush and can drive the industrial brush to move forwards and backwards.
Preferably, the sliding tables are provided with two groups, comb clamps are arranged on the two groups of sliding tables, and the interval distance between the two groups of sliding tables is matched with the interval distance between the two groups of sampling rollers.
Preferably, a waste cotton fiber collecting box is further arranged in the case, and a cotton fiber tray is arranged below the cotton fiber sampling module; the upper end of a bracket in the cotton fiber image acquisition module is provided with a sliding rail, and the sampling camera is embedded on the sliding rail through a sliding block, so that the sampling camera can slide left and right. Preferably, the outer diameter of the industrial hairbrush adopted in the cotton fiber sample brushing module has a vertical gap of 2-3mm with the contact surface of the lofting platform, so that the industrial hairbrush can peel free fibers and neps, and simultaneously, the requirement of minimum radial traction force of main cotton whisker fibers is met.
Preferably, the sampling camera adopted by the cotton fiber image acquisition module is a microscopic CCD camera with an amplifying function, the optical microscope lens is favorable for finely capturing fine characteristics of fibers, and the microscopic CCD camera can convert acquired picture information into digital information so as to facilitate computer data transmission and post-processing.
The application method of the cotton fiber length detection device based on machine vision comprises the steps of utilizing the cotton fiber image acquisition module in the cotton fiber length detection system to acquire cotton sample pictures through a sampling camera, and performing corresponding data algorithm processing analysis through the central control module of the detection system, wherein the data algorithm processing analysis steps mainly comprise:
(1) extracting cotton sample areas in the images through programming software, and performing image preprocessing, including image cutting, image enhancement, median filtering and gray level conversion; different cotton samples have different maximum gray values, and the result of normalization of the gray values of all pixel points of the image is obtained by calculating the difference between each pixel value and the maximum gray value in the image;
(2) the normalized gray level image is cut into a plurality of rectangular images which accord with the processing in a horizontal average mode along the vertical direction through programming software, and the height of each rectangular image is guaranteed to be the same pixel size; then accumulating gray values, summing and averaging, and calculating the average value of each rectangular gray value after normalization to obtain a corresponding value of the ordinate of the length-frequency distribution curve;
(3) establishing a corresponding relation between pixels and actual length and dimension through programming software, converting a measuring unit into international unit system millimeter, and calculating the corresponding actual length and dimension of each pixel to obtain the corresponding value of the length-frequency distribution curve length actual dimension abscissa;
(4) and (3) carrying out mathematical operation by programming software to fit a length-frequency distribution curve, establishing a length-frequency distribution curve model by adopting a machine learning technology, and respectively calculating the average length, the uniformity index and the short fiber rate of the upper half part of the cotton sample according to the model.
In the structure of the cotton fiber length detection device based on machine vision, the cotton fiber sampling module 8 comprises two groups of sampling rollers, the sampling comb clamps are fixed on the sliding table of the sliding table module, and when sampling is carried out, the sampling rollers rotate, and the comb clamps are fixed, so that cotton fibers in the sampling rollers can be automatically grabbed.
Each group of sampling roller structure comprises a sampling roller shell, a sampling port, a central rotating shaft, a round hole arc sampling plate, a cotton sample pressing plate, a cylinder, a connecting rod mechanism and a servo motor; the cotton sample pressing plate has a pressing function on cotton fibers. The cylinder is installed in the sampling cylinder shell and is connected with the connecting rod mechanism, the cotton sample pressing plate is controlled through the connecting rod mechanism, so that the connecting rod mechanism pushes the cotton sample pressing plate to press down the cotton fiber sample, the cotton fiber penetrates through the circular hole arc sampling plate, and the sampling comb clamp can clamp the cotton fiber to finish sampling.
The cotton fiber sample brushing module 7 and the cotton fiber sampling module 8 are positioned on the same horizontal plane, the sampling comb clamp is fixed on a sliding table of the sliding table module, the sliding table is transversely moved to a lofting platform of the cotton fiber sample brushing module 7 through the sliding table module, the motor drives the industrial brush 39 to brush off free fibers and neps on cotton whiskers of the sampling comb clamp, and meanwhile, cotton whiskers on the sampling comb clamp are further combed neatly and straightened. The outer diameter of the industrial brush adopted in the cotton fiber brush sample module 7 has a vertical clearance of 2-3mm with the contact surface of the lofting platform so as to meet the minimum radial traction requirement of the brush on free fibers and neps stripping without damaging main cotton fiber.
The cotton fiber image acquisition module 2, the cotton fiber sampling module 8 and the cotton fiber sample brushing module 7 are all located on the same horizontal plane, the sampling comb clamp is fixed on the sliding table module, the sliding table module transversely moves the sliding table to the image acquisition table 21 of the cotton fiber image acquisition module 2, the sampling cotton fibers are moved and positioned to the position of the image acquisition table 21, and the sampling camera 18 is triggered to acquire cotton fiber pictures. The cotton fibers are straightened by the specially arranged air suction box body 23 and the cotton fiber air suction opening 22, and the electric push rod II 48 is arranged below the cotton fiber air suction box body 23 in the cotton fiber image acquisition module 2, so that the air suction box body 23 can be driven to move forwards and backwards, and the cotton fibers are straightened before photographing, so that the sampling camera 18 can acquire the optimal cotton fiber pictures. The sampling camera 18 is used for collecting cotton fiber pictures, the cotton fiber pictures are processed and analyzed by the central control module 10, the terminal database keeps the original results of the left sample and the right sample of the cotton sample, the final length index is obtained as a left-right average value, and meanwhile, the length index data are synchronously uploaded to the cloud processor. After the picture is collected, the sampling comb clamp is opened, cotton fibers are sucked by negative pressure air flow of the air suction inlet 22 and conveyed to the waste cotton fiber collecting box 6 through the negative pressure cotton suction pipeline 24. The sampling comb clamp is retracted to the cotton fiber sampling module 8 through the sliding table module to finish resetting.
The cotton fiber image acquisition module 2 acquires cotton sample images through the sampling camera 18, performs corresponding data algorithm processing analysis through the central control module 10, establishes a mathematical model through data analysis, and calculates quality indexes such as body length, average length, short fiber, uniformity and the like respectively.
Compared with the prior art, the cotton fiber length detection device and the application method based on machine vision provided by the application have the following beneficial effects: according to the application, the cotton fiber image acquisition module, the cotton fiber sample brushing module, the cotton fiber sampling module, the sliding table module and the central control module are matched with each other in the detection system, so that the problems that the cotton fiber length detection depends on manual experience, the labor intensity is high, the influence of human factors is large and the detection requirement of high cotton yield is difficult to meet are solved in the cotton fiber length detection process, and the detection system can be used for realizing automatic batch rapid detection of the cotton fiber length.
The detection system and the application method of the application creatively provide a machine vision method for detecting the length of the cotton fiber, and effectively solve the problems of instability and incapability of repeatability of long-term working of the single-channel measurement of the length of the cotton fiber by the light intensity type photoelectric detection system; under the condition of non-ideal sampling, the detection efficiency is improved without being influenced by fluctuation of the light flux in a small range; the method has the characteristics of high detection speed, high repeatability of detection length index results and high stability of detection results, and has the advantages of low cost and high universality.
The method is applied to a cotton purchasing link, can quickly obtain cotton length quality indexes including main body length, average length, short fibers, uniformity and the like, objectively evaluates the quality of purchasing seed cotton, provides scientific reference for reasonable and fair pricing of cotton processing enterprises, greatly reduces labor intensity, eliminates influence of subjective factors, improves detection precision and efficiency, and provides a batch, normalized and standardized detection mode for cotton processing enterprises to detect cotton fiber length indexes. Meanwhile, the detection data is uploaded to the cloud processor in real time, interference of human factors is eliminated, fairness and fairness in purchasing links are promoted, powerful cotton length index data are provided for cotton quality tracing in the purchasing stage. The device has simple structure and low energy consumption, and has great utilization value and popularization value. Further promotes the formation of good-quality and premium situation of cotton.
Drawings
Fig. 1 is a schematic diagram of a front view of a detection system according to the present application.
Fig. 2 is a schematic front view of the detecting device according to the present application with the peripheral casing removed.
Fig. 3 is an enlarged side view of the structure a of the present application.
Fig. 4 is an enlarged view of the structure B according to the present application.
Fig. 5 is an enlarged view of the structure C according to the present application.
Fig. 6 is a schematic perspective view of a cotton fiber image acquisition module in the detection system according to the present application.
FIG. 7 is a schematic diagram of a detection process of the detection system of the present application.
FIG. 8 is a schematic diagram of a length-frequency distribution curve in a method of using the detection system of the present application.
The figure shows: 1 is a machine box, 2 is a cotton fiber image acquisition module, 3 is an observation port, 4 is a lofting port I, 5 is a lofting port II, 6 is a waste cotton fiber collection box, 7 is a cotton fiber sampling module, 8 is a cotton fiber sampling module, 9 is a cotton fiber tray, 10 is a central control module, 11 is a support frame, 12 is a sliding table module, 13 is a sliding table I, 14 is a sliding table II, 15 is a sampling roller I, 16 is a sampling roller II, 17 is a support frame, 18 is a sampling camera, 19 is a light reflecting panel, 20 is a bottom plate, 21 is an image acquisition table, 22 is an air suction port, 23 is an air suction box body, 24 is an air suction pipe, 25 is a transverse small support frame, 26 is a hinge shaft, 27 is a support arm, 28 is a slide rail, 29 is a slide block, 30 is a cloud platform, 31 is a sampling roller shell, 32 is a cotton sample pressing plate, 33 is a central rotating shaft, 34 is a circular hole arc sampling plate, 35 is a sampling port, 36 is a sampling circular hole, 37 is a cylinder, 38 is a connecting rod mechanism, 39 is an industrial brush, 40 is an air discharge platform, 41 is an electric push rod I, 42 is an electric motor, 43 is an air suction pipe, 44 is a support, 45 is an air suction pipe, and 45 is a transmission pipe; 46 is a comb clamp I, 47 is a comb clamp II, and 48 is an electric push rod II.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1:
referring to the drawings, a cotton fiber length detection device based on machine vision mainly comprises a machine case 1, wherein a cotton fiber image acquisition module 2, a cotton fiber sample brushing module 7, a cotton fiber sampling module 8, a sliding table and sliding table module 12 and a central control module 10 are mainly arranged in the machine case 1; the case is provided with an observation port 3, a lofting port I4 and a lofting port II 5.
The cotton fiber sampling module 8 is used for sampling cotton fibers of a sample, and the structure mainly comprises two groups of sampling rollers.
The cotton fiber sample brushing module 7 is used for brushing sample to the cotton fiber sample, and the structure mainly comprises a sample brushing bracket, and a sample setting platform 40 and an industrial brush 39 are arranged on the sample brushing bracket.
The cotton fiber image acquisition module 2 is used for carrying out image sampling on the cotton fiber sample and uploading image data information to the central control module; the structure mainly comprises a bracket 17, a bottom plate 20, a transverse small bracket 25, a sampling camera 18, an image acquisition table 21 and a reflecting panel 19, wherein the bracket 17 is arranged above the bottom plate 20, the sampling camera 18 is arranged above the bracket 17, the image acquisition table 21 is arranged at a position on the bottom plate, which is opposite to the camera, the image acquisition table 21 is formed by a box body serving as a square light source, an air suction box body 23 is arranged at the rear side of the image acquisition table 21, an air suction opening 22 is arranged on the air suction box body, an opening of the air suction opening 22 faces the image acquisition table 21, an electric push rod II 48 is arranged below the cotton fiber air suction box body 23 in the cotton fiber image acquisition module, and can drive the air suction box body to move forwards and backwards to perform direct treatment before photographing cotton fibers; the rear side of the air suction box body 23 is provided with an air suction pipeline 24; the small transverse bracket 25 is provided with a hinge shaft, the light reflecting panel 19 is provided with a supporting arm 27, and the supporting arm 27 is connected with the small transverse bracket 25 through the hinge shaft 26.
The central control module is used for carrying out data algorithm processing analysis on the cotton sample image information obtained by the cotton fiber image obtaining module and uploading the data algorithm processing analysis.
Two groups of sampling rollers in the cotton fiber sampling module 8 structure, wherein each group of sampling roller structure comprises a sampling roller shell 31, a sampling port 35, a central rotating shaft 33, a round hole arc sampling plate 34, a cotton sample pressing plate 32, a cylinder 37, a connecting rod mechanism 38 and a servo motor; the cylinder is installed in the sampling cylinder shell and is connected with the connecting rod mechanism, the cotton sample pressing plate is installed in the sampling cylinder, one end of the cotton sample pressing plate is flexibly connected with the sampling cylinder, the flexible connection comprises an elastic rubber sheet and a rubber band, and the other end of the cotton sample pressing plate is propped against the inner wall of the sampling cylinder and can be lifted or slide up and down under the drive of the connecting rod mechanism.
The sample brushing support 42 in the cotton fiber sample brushing module is provided with a sample setting platform 40, an industrial brush 39, a motor 43, an electric push rod I41 and an air suction pipe 44, wherein the industrial brush is close to the sample setting platform, the air suction pipe is arranged at the rear side of the industrial brush, and the electric push rod is arranged on the sample brushing support at the lower part of the industrial brush and can drive the industrial brush to move forwards and backwards.
The slip table is equipped with two sets of, all is equipped with the comb clamp on two sets of slip tables, the interval of two sets of slip tables is apart from the looks adaptation of interval of two sets of sampling cylinder. The cotton fiber length detection device based on machine vision provided by the embodiment is used for completing the detection of the cotton fiber length index once through sampling, sample brushing, image sampling, recycling and data analysis uploading of the cotton fiber serving as a sample. The networking work is needed for the analysis and uploading of the computer data, and the length index and the distribution curve of the cotton fibers of the sample are analyzed and calculated through internal software. In actual operation, a detector firstly opens an air pump to ensure that the negative pressure air flow of the detection environment of the whole system is smooth, randomly takes 20-30g of cotton samples, equally divides the cotton samples into a left sample and a right sample, respectively puts the cotton samples into a sample discharge port I and a sample discharge port II, presses a start button, the sample discharge port is communicated with a sampling roller, presses the cotton samples through a cotton sample pressing plate in the sampling roller, enables cotton fibers to pass through a circular hole arc sampling plate, drives the circular hole arc sampling plate to rotate around an axle center through the roller, and is provided with a comb clamp on a sliding table module to scrape the cotton fibers so as to finish random automatic sampling; after the delay of the control program control, the comb clamps arranged on the sliding table module drive cotton fibers to move to a lofting platform 40 of the cotton fiber sample brushing module, and the cotton fibers are brushed for 3-4s by an industrial hairbrush 39 to finish automatic sample brushing; through central control program control time delay, cotton fibers mounted on a comb of a sliding table module after moving a sample to reach an image acquisition table 21 of a cotton fiber image acquisition module 2, an electric push rod II 48 is arranged below a cotton fiber air suction box body 23 in the cotton fiber image acquisition module 2, the air suction box body 23 can be driven to move forward and backward, the cotton fibers are subjected to straight treatment before being photographed, the cotton fibers are fixed by a negative pressure air port, and the cotton sample is photographed by a sampling camera 18, so that one-time cotton sample detection is completed. The sliding table module drives the comb clamp to move to the initial sampling position through the cooperation of the central control program control delay and the optical limit sensor.
The detection system in the embodiment creatively provides a machine vision method for detecting the length of the cotton fiber, and effectively solves the problems of instability and incapability of repeatability of long-term working of the single-channel measurement of the length of the cotton fiber by the light intensity type photoelectric detection system; the method has the characteristics of high detection speed, high repeatability of detection length index results and high stability of detection results, and has the advantages of low cost and high universality.
The detection system provided by the embodiment can rapidly realize the detection of the length of the cotton fiber, can realize the deduction and calculation of the quality indexes such as the main body length, the quality length, the average length, the short fiber rate, the uniformity and the like of the cotton fiber by applying corresponding mathematical statistics and algorithms, and solves the problems of complicated detection procedures, slower speed, high labor intensity, strong subjective factors, inaccurate detection results and the like of the length index of the cotton fiber in the purchasing and processing processes of the cotton. The method has a certain practical reference application value in cotton spinning production; meanwhile, the method is applied to the purchasing process, the detected cotton fiber length index data can be uploaded to the cloud processor in real time, the technical problem that the corresponding length index information of a cotton purchasing link, a processing link and a ginned cotton packaging link is lost can be established, and powerful real-time cotton fiber length quality data of the purchasing link is provided for high-quality tracing of cotton.
Example 2:
compared with embodiment 1, this embodiment is different in that:
a waste cotton fiber collecting box is also arranged in the case, and a cotton fiber tray 9 is arranged below the cotton fiber sampling module; the upper end of a bracket in the cotton fiber image acquisition module is provided with a sliding rail, and the sampling camera is embedded on the sliding rail through a sliding block, so that the sampling camera can slide left and right.
Example 3:
compared with embodiment 2, this embodiment is different in that:
the outer diameter of the industrial brush adopted in the cotton fiber sample brushing module is in a vertical gap of 2-3mm with the contact surface of the lofting platform, so that the industrial brush can peel free fibers and neps, and simultaneously, the requirement of minimum radial traction force of main cotton fiber fibers is met.
The sampling camera adopted by the cotton fiber image acquisition module is a microscopic CCD camera with an amplifying function, the optical microscope lens is favorable for finely capturing fine characteristics of fibers, and the microscopic CCD camera can convert acquired picture information into digital information so as to facilitate computer data transmission and post-processing.
Example 4:
according to the application method of the cotton fiber length detection device based on machine vision, the cotton fiber image acquisition module in the cotton fiber length detection system is utilized to acquire a cotton sample picture through the sampling camera, and corresponding data algorithm processing analysis is carried out through the central control module of the detection system, wherein the data algorithm processing analysis steps mainly comprise:
(1) extracting a cotton sample area in an image by programming software python, and performing image preprocessing, including image cutting, image enhancement, median filtering and gray level conversion; different cotton samples have different maximum gray values, and the result of normalization of the gray values of all pixel points of the image is obtained by calculating the difference between each pixel value and the maximum gray value in the image;
(2) the normalized gray level image is cut into a plurality of rectangular images which accord with the processing in a horizontal average mode along the vertical direction through programming software python, and the height of each rectangular image is guaranteed to be the same pixel size; then accumulating gray values, summing and averaging, and calculating the average value of each rectangular gray value after normalization to obtain a corresponding value of the ordinate of the length-frequency distribution curve;
(3) establishing a corresponding relation between pixels and actual length and dimension through programming software python, converting a measuring unit into international unit system millimeter, and calculating the corresponding actual length and dimension of each pixel to obtain the corresponding value of the length-frequency distribution curve length actual dimension abscissa;
(4) and (3) carrying out mathematical operation by programming software python to fit a length-frequency distribution curve, establishing a length-frequency distribution curve model by adopting a machine learning technology, and respectively calculating the average length, the uniformity index and the short fiber rate of the upper half part of the cotton sample according to the model.
The embodiment creatively provides a machine vision method for detecting the length of the cotton fiber, and effectively solves the problems of instability and incapability of repeatability of long-term working of single-channel measurement of the length of the cotton fiber by a light intensity type photoelectric detection system; under the condition of non-ideal sampling, the detection efficiency is improved without being influenced by fluctuation of the light flux in a small range; the method has the characteristics of high detection speed, high repeatability of detection length index results and high stability of detection results, and has the advantages of low cost and high universality.
Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the application, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. The cotton fiber length detection device based on machine vision is characterized by mainly comprising a machine case, wherein a cotton fiber image acquisition module, a cotton fiber sample brushing module, a cotton fiber sampling module, a sliding table module and a central control module are mainly arranged in the machine case; the case shell is provided with an observation port, a lofting port I and a lofting port II;
the cotton fiber sampling module is used for sampling sample cotton fibers and mainly structurally comprises two groups of sampling rollers;
the cotton fiber sample brushing module is used for brushing samples of sample cotton fibers and mainly structurally comprises a sample brushing bracket, wherein a sample laying-out platform and an industrial brush are arranged on the sample brushing bracket;
the cotton fiber image acquisition module is used for carrying out image sampling on the cotton fibers of the sample and uploading image data information to the central control module; the structure mainly comprises a bracket, a bottom plate, a transverse small bracket, a sampling camera, an image acquisition table and a reflecting panel, wherein the bracket is arranged above the bottom plate, the sampling camera is arranged above the bracket, the image acquisition table is arranged on the bottom plate and is opposite to the position of the camera, the image acquisition table is composed of a box body serving as a square light source, an air suction box body is arranged at the rear side of the image acquisition table, an air suction opening is arranged on the air suction box body, an opening of the air suction opening faces the image acquisition table, and an air suction pipeline is arranged at the rear side of the air suction box body; the light reflecting panel is provided with a supporting arm which is connected with the small transverse bracket through the hinge shaft;
the central control module is used for carrying out data algorithm processing analysis on the cotton sample image information obtained by the cotton fiber image obtaining module and uploading the data algorithm processing analysis;
two groups of sampling rollers in the cotton fiber sampling module structure, wherein each group of sampling roller structure comprises a sampling roller shell, a sampling port, a central rotating shaft, a round hole arc sampling plate, a cotton sample pressing plate, a cylinder, a connecting rod mechanism and a servo motor; the cylinder is arranged in the sampling cylinder shell and is connected with the connecting rod mechanism, the cotton sample pressing plate is arranged in the sampling cylinder, one end of the cotton sample pressing plate is flexibly connected with the sampling cylinder, the flexible connection comprises an elastic rubber sheet and a rubber band, and the other end of the cotton sample pressing plate is propped against the inner wall of the sampling cylinder and is lifted or slides up and down under the drive of the connecting rod mechanism.
2. The machine vision-based cotton fiber length detection device according to claim 1, wherein the sample brushing support in the cotton fiber sample brushing module is provided with a sample setting platform, an industrial hairbrush, a motor, an electric push rod I and an air suction pipe, wherein the industrial hairbrush is arranged close to the sample setting platform, the air suction pipe is arranged at the rear side of the industrial hairbrush, and the electric push rod is arranged on the sample brushing support at the lower part of the industrial hairbrush to drive the industrial hairbrush to move forwards and backwards.
3. The machine vision-based cotton fiber length detection device according to claim 2, wherein two groups of sliding tables are provided, comb clamps are arranged on the two groups of sliding tables, and the interval distance between the two groups of sliding tables is matched with the interval distance between the two groups of sampling rollers.
4. The machine vision-based cotton fiber length detection device according to claim 3, wherein a waste cotton fiber collecting box is further arranged in the machine case, and a cotton fiber tray is arranged below the cotton fiber sampling module; the upper end of a bracket in the cotton fiber image acquisition module is provided with a sliding rail, and the sampling camera is embedded on the sliding rail through a sliding block, so that the sampling camera slides left and right; in the cotton fiber image acquisition module, an air suction box body is provided with an electric push rod II, and the electric push rod II drives the air suction box body to move forwards and backwards, so that the cotton fibers are subjected to straight treatment before photographing.
5. The machine vision-based cotton fiber length detection device according to claim 4, wherein the industrial brush adopted in the cotton fiber sample brushing module has a vertical gap of 2-3mm between the outer diameter of the industrial brush and the contact surface of the lofting platform, so that the industrial brush can peel free fibers and neps, and simultaneously, the requirement of minimum radial traction force of main cotton whisker fibers is met.
6. The machine vision-based cotton fiber length detection device according to claim 5, wherein the sampling camera used by the cotton fiber image acquisition module is a microscopic CCD camera with an amplifying function, the optical microscope lens is helpful for capturing fine characteristics of the fiber, and the microscopic CCD camera converts acquired picture information into digital information so as to facilitate computer data transmission and post-processing.
7. The application method of the cotton fiber length detection device based on machine vision as claimed in claim 6, wherein the process comprises the following steps: the cotton fiber image acquisition module in the cotton fiber length detection system is utilized to acquire cotton sample pictures through a sampling camera, and corresponding data algorithm processing analysis is carried out through a central control module of the detection system, wherein the data algorithm processing analysis steps mainly comprise:
(1) extracting cotton sample areas in the images through programming software, and performing image preprocessing, including image cutting, image enhancement, median filtering and gray level conversion; different cotton samples have different maximum gray values, and the result of normalization of the gray values of all pixel points of the image is obtained by calculating the difference between each pixel value and the maximum gray value in the image;
(2) the normalized gray level image is cut into a plurality of rectangular images which accord with the processing in a horizontal average mode along the vertical direction through programming software, and the height of each rectangular image is guaranteed to be the same pixel size; then accumulating gray values, summing and averaging, and calculating the average value of each rectangular gray value after normalization to obtain a corresponding value of the ordinate of the length-frequency distribution curve;
(3) establishing a corresponding relation between pixels and actual length and dimension through programming software, converting a measuring unit into international unit system millimeter, and calculating the corresponding actual length and dimension of each pixel to obtain the corresponding value of the length-frequency distribution curve length actual dimension abscissa;
(4) and (3) carrying out mathematical operation by programming software to fit a length-frequency distribution curve, establishing a length-frequency distribution curve model by adopting a machine learning technology, and respectively calculating the average length, the uniformity index and the short fiber rate of the upper half part of the cotton sample according to the model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210642908.7A CN114877810B (en) | 2022-06-08 | 2022-06-08 | Cotton fiber length detection device based on machine vision and application method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210642908.7A CN114877810B (en) | 2022-06-08 | 2022-06-08 | Cotton fiber length detection device based on machine vision and application method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114877810A CN114877810A (en) | 2022-08-09 |
| CN114877810B true CN114877810B (en) | 2023-09-29 |
Family
ID=82680619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210642908.7A Active CN114877810B (en) | 2022-06-08 | 2022-06-08 | Cotton fiber length detection device based on machine vision and application method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114877810B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116180278B (en) * | 2023-02-28 | 2023-10-03 | 江苏阿代尔新材料科技有限公司 | Regenerated fiber production line |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1632480A (en) * | 2004-12-17 | 2005-06-29 | 张志杰 | Camera shooting type test method and system for chrominance and length of cotton or wool fiber |
| CN201897536U (en) * | 2010-12-06 | 2011-07-13 | 陕西长岭纺织机电科技有限公司 | Cotton fiber length/intensity automatic sampling device |
| CN102419329A (en) * | 2011-08-23 | 2012-04-18 | 刘永浩 | Cotton fiber embedding inspection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7695592B2 (en) * | 2005-04-21 | 2010-04-13 | Honeywell International Inc. | Method and apparatus for measuring fiber orientation of a moving web |
| US7545971B2 (en) * | 2005-08-22 | 2009-06-09 | Honeywell International Inc. | Method and apparatus for measuring the crepe of a moving sheet |
-
2022
- 2022-06-08 CN CN202210642908.7A patent/CN114877810B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1632480A (en) * | 2004-12-17 | 2005-06-29 | 张志杰 | Camera shooting type test method and system for chrominance and length of cotton or wool fiber |
| CN201897536U (en) * | 2010-12-06 | 2011-07-13 | 陕西长岭纺织机电科技有限公司 | Cotton fiber length/intensity automatic sampling device |
| CN102419329A (en) * | 2011-08-23 | 2012-04-18 | 刘永浩 | Cotton fiber embedding inspection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114877810A (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101339185B (en) | Automatic microscopic imager for detecting cast-off cells and detection method | |
| CN107643295A (en) | A kind of method and system of the cloth defect on-line checking based on machine vision | |
| CN102288607A (en) | Woven fabric count detector based on digital microscope | |
| CN104655644A (en) | Method and device for automatically detecting defects of lithium battery pole piece | |
| CN114877810B (en) | Cotton fiber length detection device based on machine vision and application method | |
| CN101063659A (en) | System for detecting paper flaw of paper sheet | |
| US7588438B2 (en) | System, method and apparatus for fiber sample preparation for image analysis | |
| CN112964732A (en) | Spinning cake defect visual detection system and method based on deep learning | |
| 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 | |
| CN112730407A (en) | Detection device and detection method for detecting yarn hairiness index by using linear array digital camera | |
| JP3676878B2 (en) | Pollen automatic collection and analysis system | |
| CN201148561Y (en) | Cotton fibre sampling apparatus for testing cotton fibre length | |
| CN102419329A (en) | Cotton fiber embedding inspection method | |
| CN110261388A (en) | A kind of multi-angle yarn appearance digital test device and test method | |
| CN104778709A (en) | Construction method of electronic blackboard based on yarn sequence images | |
| CN113418919A (en) | Textile fiber component qualitative and quantitative online analysis system and method | |
| CN111189685A (en) | Automatic sampling and sample preparing device for measuring length of loose fiber and manufacturing method thereof | |
| CN201335683Y (en) | On-line image acquisition device | |
| CN215676860U (en) | Medical gauze mask structure size detection device | |
| Dong et al. | Automated quantitative cytology imaging analysis system in cervical cancer screening in Shanxi Province, China | |
| CN115629069A (en) | Visual detection table for multi-data acquisition of tea and tea straightness evaluation method | |
| WO2023226103A1 (en) | Leaf structure measurement method during threshing, redrying and air separation process and leaf outlet amount measurement method of air separator | |
| CN207051206U (en) | A kind of carding machine web quality detecting system | |
| CN217180567U (en) | Bearing end face imaging and collecting device |
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 |