CN116337858A - Device and method for rapidly detecting nep and linter content in cotton - Google Patents

Device and method for rapidly detecting nep and linter content in cotton Download PDF

Info

Publication number
CN116337858A
CN116337858A CN202310293833.0A CN202310293833A CN116337858A CN 116337858 A CN116337858 A CN 116337858A CN 202310293833 A CN202310293833 A CN 202310293833A CN 116337858 A CN116337858 A CN 116337858A
Authority
CN
China
Prior art keywords
cotton
nep
fiber
image
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
Application number
CN202310293833.0A
Other languages
Chinese (zh)
Inventor
张得旺
康宇鹏
安少元
杨虎
杨燕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shaanxi Changling Textile Electromechanical Technology Co ltd
Original Assignee
Shaanxi Changling Textile Electromechanical Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shaanxi Changling Textile Electromechanical Technology Co ltd filed Critical Shaanxi Changling Textile Electromechanical Technology Co ltd
Priority to CN202310293833.0A priority Critical patent/CN116337858A/en
Publication of CN116337858A publication Critical patent/CN116337858A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N1/00—Sampling; Preparing specimens for investigation
    • G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge 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
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30188—Vegetation; Agriculture
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Geometry (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Multimedia (AREA)
  • Preliminary Treatment Of Fibers (AREA)

Abstract

The invention discloses a rapid detection device for nep and linter content in cotton, which comprises: the cotton feeding mechanism is used for transmitting samples to be tested; the carding mechanism comprises a plurality of carding needle rollers, an inlet of the carding mechanism is connected with the tail end of the cotton feeding mechanism, and the sample to be detected is carded into single cotton fibers; the detection mechanism comprises a light-permeable detection channel, an inlet of the detection channel is connected with the tail end of the carding mechanism, and a light source and an imaging mechanism are arranged on one side of the detection channel; the light source is focused at the center of the detection channel, and the imaging mechanism is positioned at the rear side of the light source and is used for imaging and processing single cotton fiber; and the waste cotton collecting system is connected with the tail end of the detection channel and is used for collecting the detected single cotton fibers. The device disclosed by the invention can be used for compressing the work which can be finished by two sets of photoelectric detection systems into one set of image detection system while improving the performance of nep classification detection and short velvet rate detection, so that the accuracy of nep content and short velvet content detection is improved, and the running stability and the use convenience of a nep and short velvet tester are also improved.

Description

Device and method for rapidly detecting nep and linter content in cotton
Technical Field
The invention relates to the field of cotton fiber quality detection devices, in particular to a rapid detection device and a detection method for nep and linter content in cotton.
Background
The rapid nep and linter content detection device is usually installed in a textile mill laboratory and in each cotton fiber detection mechanism to detect nep content and linter content of cotton fibers, and therefore, the nep and linter content detection device must have two main functions of nep detection and linter detection.
At present, the existing nep and linter testers are all detected by adopting two extensions (nep detection extension and linter detection extension), and each extension adopts a large-target-surface analog signal photoelectric sensor (the nep extension adopts a 6mm multiplied by 6mm photoelectric sensor and the linter extension adopts a 50mm multiplied by 6mm photoelectric sensor), analog signal AD sampling and signal amplitude threshold judging technologies to detect nep content and linter content in cotton fibers; the photoelectric sensor and the light source are positioned on two sides of the detected fiber, the detected fiber is utilized to shade different light, analog signals with different amplitudes are generated on the photoelectric sensor, then AD sampling is carried out on the analog signals, and the size and the number of neps are judged according to the different signal amplitudes. This technique is commonly referred to as "photolithography".
As the requirements on cotton fiber quality in international cotton trade and textile export are improved, the short velvet content in cotton fiber indexes is already brought into the detection project of the forced national standard, the nep content is also actively advancing the establishment of national standard documents, and the implementation of national standard detection of nep content indexes and short velvet content indexes needs stable and reliable equipment to finish the detection; the large target surface photoelectric sensor adopted in the photographic detection has low resolution (nep size range is 50-3000 mu m, short velvet measurement accuracy is required to reach 0.1 mm) and is easy to be influenced by temperature to cause detection data change, so that the requirements of high accuracy and stable data in national standard detection can not be well met. In addition, in the detection process, the photoelectric sensor is easy to be interfered by external stray light, and the transmission and sampling of weak analog signals (millivolt level) generated during the measurement of a sample are easy to be interfered by external electromagnetic interference, so that the data distortion is caused, and the requirement of the long-term stability of the detection device cannot be met.
Disclosure of Invention
Aiming at the defects or shortcomings, the invention aims to provide a rapid detection device and a rapid detection method for nep and linter content in cotton, which can simultaneously detect nep content index and linter content index of cotton fibers by using a set of detection system.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
a rapid detection device for nep and linter content in cotton comprises:
the cotton feeding mechanism is used for transmitting samples to be tested;
the carding mechanism comprises a plurality of carding needle rollers, an inlet of the carding mechanism is connected with the tail end of the cotton feeding mechanism, and the sample to be detected is carded into single cotton fibers;
the detection mechanism comprises a light-permeable detection channel, an inlet of the detection channel is connected with the tail end of the carding mechanism, and a light source and an imaging mechanism are arranged on one side of the detection channel; the imaging mechanism is positioned at the rear side of the light source and is used for imaging and processing single cotton fiber;
and the waste cotton collecting system is connected with the tail end of the detection channel and is used for collecting the detected single cotton fibers.
Further, the cotton feeding mechanism comprises: and one end of the conveying belt is provided with a sample conveying plate and a roller, and the sample to be tested is conveyed to the conveying belt after passing through the roller.
The carding mechanism comprises a large carding roller and a small carding roller, and the large carding roller and the small carding roller are connected with a driving motor and a variable speed transmission assembly; the large and small carding rollers are tangentially arranged, the rotating speeds are different, and the relative motion of the circumferential surfaces is generated through the rotating speed difference, so that cotton of a tested sample is carded into a single cotton fiber state.
Further, the detection channel includes: the detection channel is communicated with the carding mechanism and the waste cotton collecting system.
Further, the light source is a zoom converging light source.
Further, the zoom converging light source includes: the movable cylindrical lenses, the grating plates, the white LED lamp plates and the radiator are sequentially arranged between the light source side plates from top to bottom; the lens assembly comprises a cylindrical lens, a light source side plate, a light source and a white LED lamp panel, wherein a displacement adjusting block is arranged on the cylindrical lens, an adjusting fixing screw is arranged at the tail end of the light source side plate, and the distance between the cylindrical lens and the white LED lamp panel is changed through the adjusting fixing screw, so that the convergence focus of the light source is changed.
Further, the imaging mechanism comprises an ultra-high-speed linear array digital camera, an image processing board is connected to the ultra-high-speed linear array digital camera, and the image processing board comprises an FPGA core processing chip and is an embedded system.
Further, the waste cotton collecting system comprises a waste cotton box communicated with the detection channel, and a fan is arranged at the tail end of the waste cotton box.
A method for rapidly detecting nep and linter content in cotton comprises the following steps:
continuously collecting single fiber images passing through each test channel and generating continuous image stream data;
carrying out gray scale processing on each image in the image stream data to obtain gray scale value data of each image;
and comparing the fiber gray value in each gray image with the background gray value to finish nep detection statistics and short velvet content detection statistics and generate a report for output.
Further, the nep detection and statistics function specifically includes:
presetting a background gray preset range value (setting a threshold value 1 and setting a threshold value 2), and comparing the fiber gray value in each gray image with the background gray preset range value: when the gray value of the fiber is larger than the set threshold value 1, the fiber is a fiber nep; when the gray value of the fiber is smaller than the set threshold value 2, the fiber is a seed cotton knot; when the gray value of the fiber is larger than the set threshold value 2 and smaller than the set threshold value 1, the fiber is a single cotton fiber.
Comparing the fiber gray value in each gray image with the background gray value, specifically:
1.1, designing a symmetrical gradient operator mask matrix T for extracting a nep edge:
Figure SMS_1
1.2, taking the peripheral neighborhood of each pixel in the gray level image to form a 5X 5 sliding window, wherein the pixel to be judged is a central point P 3,3 Recorded as matrix W:
Figure SMS_2
wherein P is i,j I is 1,2 … … 5,j is 1,2 … … for a pixel in the sliding window.
1.3, calculating a gradient value S=sum (T×W) of each pixel point in the image, comparing the gradient value S with a fiber nep threshold Th1 and a seed nep threshold Th2, and judging a matrix central point P according to a comparison result 3,3 Is a nep type of (c).
Judging a matrix central point P according to the comparison result 3,3 The nep type of (2) is specifically:
when S is>Th1, then P 3,3 Is a fiber nep;
when Th2<S<Th1, then P 3,3 Background or normal fibers;
when S is<Th2, then P 3,3 Is a seed cotton knot.
1.4, measuring the size of each nep, dividing and classifying the nep according to the nep size standard in the industry standard, and counting the number in a classified way, thus finishing report output.
Further, the measuring the size of each nep, dividing and classifying the statistical quantity according to the nep size standard in the industry standard, and completing report output specifically comprises:
2.1, after obtaining the nep type, generating a seed chip nep binary pattern Z and a fiber nep binary pattern X;
2.2, carrying out corrosion and expansion filtering treatment on the generated seed chip nep binary pattern Z and the generated fiber nep binary pattern X to eliminate random discrete noise;
2.3, calculating the seed cotton knot binary pattern Z and the fiber cotton knot binary pattern X to generate a new binary pattern Filter1, and obtaining an output binary pattern Filter2 by filtering the new binary pattern Filter 1:
2.4, using standard BLOB operator to obtain length L and width w of external rectangle of every white point in output binary image Filter2, then according to length and width of said rectangle obtaining nep diameter
Figure SMS_3
The size of each nep can be obtained according to the image resolution of 25 mu m/pixel;
and 2.5, classifying and counting the number according to the nep size standard division in the industry standard, and reporting the counting result to a computer through a communication interface to finish report output.
Further, the specific algorithm of the step 2.3 step includes:
each pixel point and 8 neighborhood thereof in the binary image form a matrix A, the matrix A is multiplied by a corrosion operator F1, the value of a center point is judged according to a corrosion filtering threshold Th3, and then a new binary image Filter1 is generated; each pixel point and 8 neighborhood thereof in the new binary image form a matrix B, the matrix B is multiplied by an expansion operator F2, the value of a center point is judged according to an expansion filtering threshold Th4, and the output result of the filtering module is Filer2:
Figure SMS_4
Figure SMS_5
further, the continuous acquisition of the images of each fiber passing through the test channel and the generation of continuous image stream data further comprise the step of dynamically updating and storing the continuous image stream data according to a format of 4096 columns x 1024 rows for searching or identifying the image reserved for examination when neps are found.
Further, the length calculation and statistics of the single cotton fiber specifically include:
5.1, comparing the fiber gray value in each gray image with the background gray value, and extracting fiber characteristics to obtain a curve communicated by a single pixel;
5.2, calculating the pixel length of each fiber by utilizing an edge extraction, filtering and cyclic recursion algorithm, and then calculating the length of each fiber according to the pixel resolution;
and 5.3, counting the length according to the classification standard of the national standard.
Further, the calculating the length of each fiber according to the pixel resolution by using the edge extraction, filtering and cyclic recursion algorithm specifically includes:
6.1 designing a symmetrical gradient operator mask matrix T f For extracting cotton fiber edges:
Figure SMS_6
6.2, taking the peripheral neighborhood of each pixel in the image to form a 3X 3 sliding window, wherein the pixel to be judged is a central point P 2,2 Recorded as matrix RM.
Figure SMS_7
6.3, calculating each center pixel point P in the image 2,2 Gradient value S of (2) f ＝sum(T f X RM), comparing the gradient value S with the fiber brightness threshold Th5 issued by the computer, and judging the central point P of the matrix according to the comparison result 2,2 Whether cotton is the pixel or not and generating a Fiber binary image Fiber, wherein the size of the binary image is 4096 multiplied by 512.
Figure SMS_8
And 6.4, performing expansion and corrosion filtering to eliminate salt and pepper noise, wherein the filtering adopts 8-neighborhood convolution filtering. Forming a matrix C by using the Fiber binary image for each pixel point and 8 adjacent domains thereof, and performing matrix multiplication operation on the matrix C and an expansion operator Fb1 to generate a new binary image Fb_Filter1; each pixel point and 8 neighborhood thereof in the new binary image form a matrix D, and the matrix D and the expansion operator Fb2 are subjected to matrix multiplication operation to generate a new binary image Fb_Filter2, and the new binary image Fb_Filter2 is output to a statistics module:
Figure SMS_9
Figure SMS_10
Figure SMS_11
Figure SMS_12
Figure SMS_13
Figure SMS_14
and 6.5, obtaining the circumscribed rectangle of each continuous fiber in the Filter2 binary image by utilizing a BLOB standard operator, obtaining pixels and values in the rectangle, obtaining the real length of each cotton fiber according to the image resolution of 25 mu m/pixel, counting the number according to three cotton fiber dividing standards in industry standards, dividing the three classified counting numbers by the total number to obtain the percentage of the three fiber lengths, and reporting the calculation result to a computer through a communication interface to finish report output.
Further, the gray scale processing is carried out on the fiber image stream data to obtain gray scale value data of each image;
counting dynamic big data of the gray values of the image to form a statistical graph with gray values of 0-255 on the abscissa and the statistical times of occurrence of each gray value on the ordinate;
positioning the abscissa corresponding to the maximum value in the statistical curve to obtain a background statistical value in the current image;
comparing the background statistic value with a standard background value interval (55-65) set by a computer: if the statistic value is smaller than the lower limit 55 of the standard value interval, the brightness of the light source is lower, and the light source needs to be turned on; if the statistical value is greater than the upper limit 65 of the standard value interval, the brightness of the light source is higher, and the light source needs to be dimmed; if the statistical value is within the standard value interval (55-65), the brightness of the light source is kept unchanged without adjustment.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a rapid detection device for nep and linter content in cotton, which can realize automatic detection and analysis by arranging a cotton feeding mechanism, a carding mechanism, a detection mechanism and a waste cotton collecting system, rapidly and accurately complete detection of nep content and linter content by imaging treatment and analysis of the detection mechanism, solves the defects that the existing nep and linter testers are low in resolution, are easily influenced by temperature change and cannot consider various nep detection and linter detection, and simultaneously compresses the work which can be completed by two sets of photoelectric detection systems into one set of image detection system while improving nep classification detection and linter rate detection performance, thereby not only improving the accuracy of nep content and linter content detection, but also improving the operation stability and the use convenience of the nep and linter testers, and providing reliable detection instrument equipment for national standard detection implementation.
Furthermore, the detection device provided by the invention adopts a high-brightness white LED zooming convergence light source, and has the characteristics of high luminous efficiency, strong light, long service life and small light attenuation. The fiber flow formed by single cotton fibers flying at high speed is formed into a visual digital image by adopting an ultra-high-speed linear array digital camera, various neps can be conveniently detected and the length of each single fiber can be calculated by utilizing an image processing technology while the problem that analog signals are easily interfered is solved, and the linter rate of a sample is obtained through statistics.
The invention also provides a rapid detection method of nep and linter content in cotton, which uses an image acquisition system and an image processing system to acquire cotton fiber flow images illuminated by a light source under the influence of gray background through ultra-white transparent glass, and judges the nep number and the single cotton fiber length in the cotton fiber flow through an image detection and identification algorithm; the detection result is transmitted to the computer through the communication interface, and the computer outputs, stores and prints the detection result, so that the whole detection method is quick and accurate, and the working efficiency and accuracy are greatly improved.
Drawings
FIG. 1 is a schematic diagram of a rapid detection device for nep and linter content in cotton according to the present invention;
FIG. 2 is a schematic view of the structure of an LED zoom converging light source according to the invention;
FIG. 3 is a schematic diagram of the operation of the image detection system of the present invention;
FIG. 4 is a flow chart of a method for rapid detection of nep and linter content in cotton in accordance with the present invention;
fig. 5 is a schematic diagram of the gray scale comparison of the image of the present invention.
In the figure, 1-cotton feeding mechanism; 2-a large carding roll; 3-small comb needle roller; 4-carding mechanism; 5-a detection channel; 6-Gray background plate; 7-a light source; 8-an imaging mechanism; 9-an image processing board; 10-a cotton waste collection system; 11-a light source side plate; 12-a cylindrical lens; 13-a displacement adjusting block; 14-grating plates; 15-a white LED lamp panel; 16-a heat sink; 17-adjusting the set screw; 18-single cotton fiber; 19-a sample to be tested; 1-a conveyor belt; 1-2-sample feeding; 1-3-roller; 10-1, a waste cotton box; 10-2, a fan.
Detailed Description
The present invention now will be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
As shown in fig. 1, the rapid detection device for nep and linter content in cotton according to the present invention comprises:
the cotton feeding mechanism 1 is used for conveying a sample 19 to be tested;
the carding mechanism 4 comprises a plurality of carding needle rollers, wherein an inlet of the carding mechanism 4 is connected with the tail end of the cotton feeding mechanism 1, and a sample 19 to be detected is carded into single cotton fibers 18;
the detection mechanism comprises a light-permeable detection channel 5, an inlet of the detection channel 5 is connected with the tail end of the carding mechanism 4, and a light source 7 and an imaging mechanism 8 are arranged on one side of the detection channel 5; wherein, the light source 7 focuses on the center of the detection channel 5, and the imaging mechanism 8 is positioned at the rear side of the light source 7 and is used for imaging and processing the single cotton fiber 18;
and the waste cotton collecting system 10 is connected with the tail end of the detection channel 5 and is used for collecting detected single cotton fibers 18.
Wherein, cotton feeding mechanism 1 includes: and one end of the conveyor belt 1-1 is provided with a sample feeding plate 1-2 and a roller 1-3, and the sample 19 to be tested is fed to the conveyor belt 1-1 after passing through the roller 1-3.
The carding mechanism 4 comprises a large carding roller 2 and a small carding roller 3, and the large carding roller 2 and the small carding roller 3 are connected with a driving motor and a variable speed transmission assembly; the large carding needle roller 2 and the small carding needle roller 3 are tangentially arranged, the rotating speeds are different, and the relative motion of the circumferential surfaces is generated through the rotating speed difference, so that cotton of a tested sample is carded into a single cotton fiber state.
The detection channel 5 comprises: an ultra-white transparent glass 5-1 on the same side as the light source and a gray background plate (6) on the other side, wherein the detection channel 5 is communicated with the carding mechanism 4 and the waste cotton collecting system 10. The detection channel 5 is in a regular quadrangular frustum pyramid structure, the lower bottom of the detection channel is in a rectangle with the size of 5mm multiplied by 100mm, the upper bottom of the detection channel is in a rectangle with the size of 10mm multiplied by 100mm, and the upper bottom and the lower bottom of the detection channel are not closed, so that the detection channel is used as a flight channel of a cotton fiber flow to be detected. The ultra-white transparent glass 5-1 is used as the front side surface of the quadrangular frustum, the gray background plate 6 is used as the rear side surface, and the left side surface and the right side surface are made of ferrous metal materials; the joints between the four sides are sealed and smooth. The lower bottom surface is connected with an outlet of the carding mechanism 4, and the upper bottom surface is connected with an inlet of the waste cotton collecting system 10; the carding mechanism 4 generates relative motion on the circumferential surface through the rotation speed difference of the large carding needle roller 2 and the small carding needle roller 3, the cotton of the tested sample is carded into a single cotton fiber state, and single fibers are driven by negative pressure air to pass through a detection channel to form single fiber flow.
The waste cotton collecting system 10 comprises a waste cotton box 10-1 communicated with the detection channel 5, wherein a fan 10-2 is arranged at the tail end of the waste cotton box 10-1, and the waste cotton collecting system consists of the waste cotton box and the fan.
Further, as shown in fig. 2, the light source 7 is an LED zoom converging light source. The zoom converging light source includes: a light source side plate 11, wherein a movable cylindrical lens 12, a grating plate 14, a white LED lamp panel 15 and a radiator 16 are sequentially arranged between the light source side plates 11 from top to bottom; the cylindrical lens 12 is provided with a displacement adjusting block 13, the tail end of the light source side plate 11 is provided with an adjusting and fixing screw 17, and the converging focus of the light source is changed through the adjusting and fixing screw 17.
As shown in fig. 2, the specific assembly mode of the LED zoom convergence light source is as follows: the length of the white LED lamp panel 15 is 12cm, 21 LED lamp beads with the size of 1W are designed, the color temperature of the lamp beads is 5000K, and a 16-24V direct current PWM power supply is adopted for power supply; the white LED lamp panel 15 is arranged on the radiator 16, and heat conduction silicone grease with the thickness of 0.1mm is uniformly smeared between contact surfaces; the grating plate 14 is positioned at the position 1cm in front of the LED lamp beads, and diverges the point light sources emitted by the LED lamp beads into uniform linear light sources; the cylindrical lens 12 after cutting is assembled with the 2 displacement adjusting blocks 13 and then is arranged in front of the grating plate; finally, the radiator, the white LED lamp panel, the grating plate, the cylindrical lens and the displacement adjusting block are combined together through the light source side plate 11; the two ends of the light source are sealed through end cover plates; the position of the displacement adjusting block in the chute is adjusted by adjusting the fixing screw 17, so that the distance between the axis of the cylindrical lens and the LED light source is changed, and the purpose of changing the converging focus of the light source is achieved.
In addition, the imaging mechanism 8 includes an ultra-high-speed linear array digital camera, and the high-speed linear array digital camera is connected with an image processing board 9, as shown in fig. 3, where the image processing board 9 includes an FPGA core processing chip and is an embedded system. The ultra-high-speed linear array digital camera 8 and the LED zooming and converging light source 7 are arranged on one side of transparent glass of the detection channel; adjusting the convergence focus of the convergence light source to enable the focus to be positioned in the middle of the thickness of the detection channel, and enabling the two convergence light sources to form a shadowless lamp effect to illuminate cotton fiber flow in the channel; the photosensitive unit of the ultra-high-speed linear array digital Camera is 4096 pixels×1 row, the output data format is Mono8, the exposure frequency is set to 150KHz, reflected light rays of cotton fiber flows are collected in real time, optical signals are converted into digital signals in the Camera, the Camera works in a FULL mode, and the digital signals are transmitted to the image processing board 9 through two camera_links.
As shown in fig. 3, the image processing board 9 has at least 3 interfaces, one is camera_link, for receiving digital signals output by the Camera; one is a gigabit ethernet interface for communicating with a computer; one is PWM control interface, which is connected with two light sources to supply power to the light sources and control the brightness of the light sources; the image processing board adopts an embedded design, takes an FPGA as a core chip, is internally provided with image processing software, and automatically runs after the image processing board is electrified to finish the functions of nep detection, short velvet detection and automatic control of a light source.
The software design is designed by adopting a Verilog HDL language, runs on an image processing board taking a CYCLONE V series FPGA as a core processing chip, realizes real-time high-speed processing of image data by utilizing a large-scale parallel operation technology of the FPGA, and completes the functions of nep content detection, short velvet content detection statistics, image quality monitoring, light source brightness adjustment, communication with a computer and the like; the software adopts a modularized design and comprises an image acquisition module, an image storage module, an image brightness monitoring module, a light source brightness adjustment judging module, a PWM signal generator module, a seed cotton knot identification module, a fiber cotton knot identification module, a cotton knot classification statistics module, a fiber length calculation statistics module, a filtering module, a communication module and the like.
The image acquisition module generates continuous image stream data with the transverse resolution of 4096 pixels/line by using 4.8Gbps (512K multiplied by 150K multiplied by 8) data continuously output by a camera in real time through algorithms such as parallel reconstruction, clock synchronous isolation, dynamic framing and the like, wherein each pixel occupies 1 BYTE, and the gray value is between 0 and 255. And then the data stream is output to an image storage module, an image brightness monitoring module, two nep detection modules and a flock detection statistics module in parallel.
The image storage module dynamically updates and stores the input continuous image stream data according to a format of 4096 columns multiplied by 1024 rows, and is used for inquiring and looking up or identifying the image reserved for backup when a nep is observed by a computer, and the module does not participate in nep detection and short velvet detection calculation.
The image brightness monitoring module forms a statistical graph with gray values of 0-255 on the abscissa and the statistics times of each gray value on the ordinate through dynamic big data statistics of input image stream data, then positions the abscissa corresponding to the maximum value in the statistical graph according to the phenomenon that the fiber data occupancy rate is far smaller than the background occupancy rate in the image stream data, then obtains the background statistical value in the current image, and then transmits the statistical value to the light source brightness adjustment judging module.
In addition, as shown in fig. 4, the invention also provides a rapid detection method for nep and linter content in cotton, comprising the following steps:
s1, continuously collecting fiber images of each passing test channel, and generating continuous image stream data;
firstly, adjusting an LED zooming converging light source in the device:
the LED zooming converging light source is arranged on one side of transparent glass of the detection channel, the converging focal point of the converging light source is adjusted to enable the focal point to be located in the middle of the thickness of the detection channel, the two converging light sources form a shadowless lamp effect, and cotton fiber flow in the channel is illuminated;
then, light source brightness adjustment is performed:
the light source brightness adjustment judging module compares the input background statistic value with a standard background value interval (55-65) set by a computer, and if the statistic value is smaller than the lower limit 55 of the standard background value interval, the light source brightness is lower and the light source needs to be turned on; if the statistical value is greater than the upper limit 65 of the standard value interval, the brightness of the light source is higher, and the light source needs to be dimmed; if the statistical value is within the standard value interval (55-65), the brightness of the light source is kept unchanged without adjustment; next, an adjustment amplitude value is calculated, the adjustment amplitude value divides the image from the darkest gray value of 0 to the brightest gray value of 255 into 50 step intervals, and the adjustment amplitude value is calculated in the following manner:
Figure SMS_15
and finally, transmitting the light source adjusting direction (dimming is 1 and dimming is 2) and the adjusting amplitude value to the PWM signal generator module.
The PWM signal generator module is used for generating PWM modulation signals and carrying out PWM signal adjustment according to adjustment commands. The clock frequency of the PWM signal generator module is 120Mhz, in order to prevent the light source from stroboscopic when the camera collects images, the PWM pulse frequency is more than 2 times of the camera collection frequency, so the PWM pulse frequency is set to 300KHz, and each PWM pulse period consists of 400 clocks; in the module, the brightness of the light source is subdivided into 100 step sizes (namely 0% -100%) from darkest to brightest, and each step size is 1 and corresponds to 4 clocks; when the PWM module is started initially, reading a pulse value which is regulated normally last time as an initial value, if data which needs to regulate the brightness of a light source are received, firstly determining an adding algorithm and a subtracting algorithm of the initial value according to the regulating direction value of the light source, and then obtaining a result value which needs to be output according to the regulating amplitude value:
Figure SMS_16
and finally, updating the initial value into a result value, converting the result value into square waves with different space occupation ratios through a PWM (pulse-width modulation) chip and a hardware circuit, performing LC smoothing treatment, converting into different direct current voltages, and outputting the different direct current voltages to the light source to finish the adjustment of the brightness of the light source.
When each single cotton fiber is carded by the large carding needle roller 2 and the small carding needle roller 3, a single fiber stream is formed through a detection channel under the drive of negative pressure air, and then photographing is carried out by an ultra-high speed linear array digital camera, so that continuous image stream data are generated.
S2, carrying out gray scale processing on each image in the image stream data to obtain gray scale value data of each image;
s3, presetting a background gray preset range value (setting a threshold value 1 and setting a threshold value 2), and comparing the fiber gray value in each gray image with the background gray preset range value: when the gray value of the fiber is larger than the set threshold value 1, the fiber is a fiber nep; when the gray value of the fiber is smaller than the set threshold value 2, the fiber is a seed cotton knot; when the gray value of the fiber is larger than the set threshold value 2 and smaller than the set threshold value 1, the fiber is a single cotton fiber. As shown in fig. 5.
The method specifically comprises the following steps:
judging the type of cotton fiber according to the difference between the gray value of the cotton fiber and the gray value of the background;
because the cotton fiber is white or yellowish white, the background is gray, the cotton fiber is whiter than the background, and the fiber neps are generated by knotting a plurality of cotton fibers, the gray value of the fiber neps in the image is larger than the gray value of the background; the cotton seed hulls are black or dark brown, so that the cotton seed hulls are darker than the background, and the gray value of the cotton seed hulls in the image is much smaller than the gray value of the background, and the cotton seed hulls are basically black. The specific calculation method comprises the following steps:
1.1, designing a symmetrical gradient operator mask matrix T for extracting a nep edge:
Figure SMS_17
1.2, taking the peripheral neighborhood of each pixel in the gray level image to form a 5X 5 sliding window, wherein the pixel to be judged is a central point P 3,3 Recorded as matrix W:
Figure SMS_18
wherein P is i，j I is 1,2 … … 5,j is 1,2 … … for a pixel in the sliding window.
1.3, comparing the fiber gray value in each gray image with the background gray value, and when the fiber gray value is larger than the background gray value, the fiber is fiber nep; when the gray value of the fiber is smaller than the gray value of the background, the fiber is a seed cotton knot; when the gray value is equal to the background gray value, the fiber is a single cotton fiber. Calculating a gradient value S=sum (T×W) of each pixel point in the image by using a gradient mask algorithm, comparing the gradient value S with a fiber nep threshold Th1 and a seed nep threshold Th2 issued by a computer, and judging a matrix center point P according to a comparison result 3,3 And respectively generating a seed chip nep binary pattern Z and a fiber nep binary pattern X, wherein the size of the binary pattern is 4096 multiplied by 512. Judging a matrix central point P according to the comparison result 3,3 The nep type of (2) is specifically:
when S is>Th1, then P 3,3 Is a fiber nep;
when Th2<S<Th1, then P 3,3 Background or normal fibers;
when S is<Th2, then P 3,3 Is a seed cotton knot.
S4, calculating and counting the length of the single cotton fiber; or measuring the size of each nep, dividing and classifying the nep according to the nep size standard in the industry standard, and counting the number in a classified way, so as to finish report output.
2. When the single fiber is calculated to be neps, measuring the size of each nep, and dividing and classifying the neps according to the nep size standard in the industry standard to count the number:
the method specifically comprises the following steps:
2.1, after obtaining the nep type, generating a seed chip nep binary pattern Z and a fiber nep binary pattern X;
2.2, carrying out corrosion and expansion filtering treatment on the generated seed chip nep binary pattern Z and the generated fiber nep binary pattern X to eliminate random discrete noise;
2.3, calculating the seed cotton knot binary pattern Z and the fiber cotton knot binary pattern X to generate a new binary pattern Filter1, and obtaining an output binary pattern Filter2 by filtering the new binary pattern Filter 1:
the specific algorithm of the step 2.3 comprises the following steps:
each pixel point and 8 neighborhood thereof in the binary image form a matrix A, the matrix A is multiplied by a corrosion operator F1, the value of a center point is judged according to a corrosion filtering threshold Th3, and then a new binary image Filter1 is generated; each pixel point and 8 neighborhood thereof in the new binary image form a matrix B, the matrix B is multiplied by an expansion operator F2, the value of a center point is judged according to an expansion filtering threshold Th4, and the output result of the filtering module is Filer2:
Figure SMS_19
Figure SMS_20
2.4, using standard BLOB operator to obtain length L and width w of external rectangle of every white point in output binary image Filter2, then according to length and width of said rectangle obtaining nep diameter
Figure SMS_21
The size of each nep can be obtained according to the image resolution of 25 mu m/pixel;
and 2.5, classifying and counting the number according to the nep size standard division in the industry standard, and reporting the counting result to a computer through a communication interface to finish report output.
3. When the single fiber is calculated to be single cotton fiber, measuring the size of each single cotton fiber, and dividing and classifying the single cotton fiber according to the nep size standard in the industry standard to count the number:
length calculation and statistics are carried out on single cotton fiber:
because the diameter of a single cotton fiber is about 20 mu m, the diameter direction of the single cotton fiber occupies only 1 pixel in an image, and because the fiber is continuous, a curve communicated by a single pixel is formed in the image, the pixel length of each fiber can be obtained by utilizing an edge extraction, filtering and cyclic recursion algorithm, then the real length of each fiber is calculated according to the resolution of the image of 25 mu m/pixel, or the size of each nep is measured, and then the nep size and length are divided according to the standard of the nep size and length in the industry standard national standard, the statistical quantity is classified, finally, the statistical result is reported to a computer through a communication interface, and the computer outputs the result according to the report requirement, thereby completing report output.
The length calculation and statistics of the single cotton fiber specifically comprise:
3.1, processing each image by utilizing edge extraction to obtain a cotton fiber edge curve communicated by single pixels:
3.1.1, designing a symmetrical gradient operator mask matrix Tf for extracting cotton fiber edges:
Figure SMS_22
3.1.2, taking the peripheral neighborhood of each pixel in the image to form a 3X 3 sliding window, wherein the pixel to be judged is a central point P 2,2 Recorded as matrix RM.
Figure SMS_23
3.1.3 calculating each center pixel point P in the image 2,2 Gradient value sf=sum (tf×rm), and the gradient value S is compared with the computer-issued valueThe fiber brightness threshold Th5 is compared, and the central point P of the matrix is judged according to the comparison result 2,2 Whether cotton is the Fiber pixel or not and generating a Fiber edge binary image Fiber, wherein the size of the binary image is 4096 multiplied by 512.
Figure SMS_24
3.2, calculating the pixel length of each fiber through a filtering algorithm and a cyclic recursion algorithm, and then calculating the actual length of each fiber according to the pixel resolution:
and 3.2.1, performing expansion and corrosion filtering to eliminate random discrete noise, wherein the filtering adopts 8-neighborhood convolution filtering. Forming a matrix C by using the Fiber binary image for each pixel point and 8 adjacent domains thereof, and performing matrix multiplication operation on the matrix C and an expansion operator Fb1 to generate a new binary image Fb_Filter1; each pixel point and 8 neighborhood thereof in the new binary image form a matrix D, and the matrix D and the expansion operator Fb2 are subjected to matrix multiplication operation to generate a new binary image Fb_Filter2, and the new binary image Fb_Filter2 is output to a statistics module:
Figure SMS_25
Figure SMS_26
Figure SMS_27
Figure SMS_28
Figure SMS_29
Figure SMS_30
and 3.2.2, calculating the circumscribed rectangle of each continuous fiber in the Filter2 binary image by utilizing a BLOB standard operator, calculating pixels and values in the rectangle, and then obtaining the actual length of each cotton fiber according to the image resolution of 25 mu m/pixel.
And 3.3, respectively counting the root numbers of the different lengths according to three cotton fiber dividing standards in national standards, and dividing the three classified counted root numbers by the total root number to obtain the percentage of the three fiber lengths.
And 3.4, finally reporting the calculation result to a computer through a communication interface to finish report output.
It will be apparent to those skilled in the art that the foregoing is merely illustrative of the preferred embodiments of this invention, and that certain modifications and variations may be made in part of this invention by those skilled in the art, all of which are shown and described with the understanding that they are considered to be within the scope of this invention.

Claims (18)

1. A rapid detection device for nep and linter content in cotton, comprising:
the cotton feeding mechanism (1) is used for transmitting a sample (19) to be tested;
the carding mechanism (4) comprises a plurality of carding needle rollers, an inlet of the carding mechanism (4) is connected with the tail end of the cotton feeding mechanism (1), and a sample (19) to be detected is carded into single cotton fibers (18);
the detection mechanism comprises a light-permeable detection channel (5), an inlet of the detection channel (5) is connected with the tail end of the carding mechanism (4), and a light source (7) and an imaging mechanism (8) are arranged on one side of the detection channel (5); the light source (7) is focused at the center of the detection channel (5), and the imaging mechanism (8) is positioned at the rear side of the light source (7) and is used for imaging and processing single cotton fibers (18);
and the waste cotton collecting system (10) is connected with the tail end of the detection channel (5) and is used for collecting detected single cotton fibers (18).
2. The rapid detection device for nep and linter content in cotton according to claim 1, characterized in that the cotton feeding mechanism (1) comprises: and one end of the conveyor belt (1-1) is provided with a sample conveying plate (1-2) and a roller (1-3), and the sample (19) to be tested is conveyed to the conveyor belt (1-1) after passing through the roller (1-3).
3. The rapid detection device for nep and linter content in cotton according to claim 1, characterized in that the carding mechanism (4) comprises a large carding needle roller (2) and a small carding needle roller (3), wherein the large carding needle roller (2) and the small carding needle roller (3) are connected with a driving motor and a variable speed transmission assembly; the large carding needle roller (2) and the small carding needle roller (3) are tangentially arranged, the rotating speeds are different, and the relative motion of the circumferential surfaces is generated through the rotating speed difference, so that cotton of a tested sample is carded into a single cotton fiber state.
4. A rapid detection device of nep and linter content in cotton according to claim 1, characterized in that the detection channel (5) comprises: the detection channel (5) is communicated with the carding mechanism (4) and the waste cotton collecting system (10).
5. The rapid detection device for nep and linter content in cotton according to claim 1, characterized in that the light source (7) is a zoom converging light source.
6. The rapid detection apparatus for nep and linter content in cotton of claim 5, wherein the zoom converging light source comprises: the LED lamp comprises light source side plates (11), wherein a movable cylindrical lens (12), a grating plate (14), a white LED lamp plate (15) and a radiator (16) are sequentially arranged between the light source side plates (11) from top to bottom; the cylindrical lens (12) is provided with a displacement adjusting block (13), the tail end of the light source side plate (11) is provided with an adjusting fixing screw (17), and the converging focus of the light source is changed through the adjusting fixing screw (17).
7. The rapid detection device for nep and linter content in cotton according to claim 1, wherein the imaging mechanism (8) comprises an ultra-high speed linear array digital camera, an image processing board (9) is connected to the high speed linear array digital camera, and the image processing board (9) comprises an FPGA core processing chip and is an embedded system.
8. The rapid detection device for nep and linter content in cotton according to claim 1, characterized in that the waste cotton collection system (10) comprises a waste cotton box (10-1) in communication with the detection channel (5), the end of the waste cotton box (10-1) being provided with a fan (10-2).
9. A method for rapidly detecting nep and linter content in cotton is characterized by comprising the following steps:
continuously collecting single fiber images passing through each test channel and generating continuous image stream data;
carrying out gray scale processing on each image in the image stream data to obtain gray scale value data of each image;
presetting a background gray preset range value (setting a threshold value 1 and setting a threshold value 2), and comparing the fiber gray value in each gray image with the background gray preset range value: when the gray value of the fiber is larger than the set threshold value 1, the fiber is a fiber nep; when the gray value of the fiber is smaller than the set threshold value 2, the fiber is a seed cotton knot; when the gray value of the fiber is larger than the set threshold value 2 and smaller than the set threshold value 1, the fiber is a single cotton fiber.
10. The method for rapid detection of nep and linter content in cotton according to claim 9, wherein: comparing the fiber gray value in each gray image with the background gray value, specifically:
10.1, designing a symmetrical gradient operator mask matrix T for extracting nep edges:
Figure FDA0004142472060000031
10.2, taking the peripheral neighborhood of each pixel in the gray level image to form a 5X 5 sliding window, wherein the pixel to be judged is a central point P 3,3 Recorded as matrix W:
Figure FDA0004142472060000032
wherein P is i，j I is 1,2 … … 5,j is 1,2 … … for a pixel in the sliding window.
10.3, calculating a gradient value S=sum (T×W) of each pixel point in the image, comparing the gradient value S with a fiber nep threshold Th1 and a seed nep threshold Th2, and judging a matrix central point P according to a comparison result 3,3 Is a nep type of (c).
11. The method for rapid detection of nep and linter content in cotton according to claim 10 wherein the matrix center point P is determined based on the comparison 3,3 The nep type of (2) is specifically:
when S is>Th1, then P 3,3 Is a fiber nep;
when Th2<S<Th1, then P 3,3 Background or normal fibers;
when S is<Th2, then P 3,3 Is a seed cotton knot.
12. The method for rapid detection of nep and linter content in cotton of claim 10 wherein comparing the fiber gray value in each gray image to the background gray value further comprises:
calculating and counting the length of a single cotton fiber;
and measuring the size of each nep, dividing and classifying the nep according to the nep size standard in the industry standard, and counting the number to finish report output.
13. The method for rapid detection of nep and linter content in cotton according to claim 12, wherein the measuring the size of each nep and classifying the statistical quantity according to nep size criteria in industry standards, and completing report output specifically comprises:
13.1, after obtaining the nep type, generating a seed chip nep binary pattern Z and a fiber nep binary pattern X;
13.2, carrying out corrosion and expansion filtering treatment on the generated seed cotton knot binary pattern Z and the generated fiber cotton knot binary pattern X to eliminate random discrete noise;
13.3, calculating the seed cotton knot binary pattern Z and the fiber cotton knot binary pattern X to generate a new binary pattern Filter1, and filtering the new binary pattern Filter1 to obtain an output binary pattern Filter 2;
13.4, utilizing standard BLOB operator to obtain length L and width w of external rectangle of every white point in output binary image Filter2, then according to length and width of said rectangle obtaining nep diameter
Figure FDA0004142472060000041
The size of each nep can be obtained according to the image resolution of 25 mu m/pixel;
13.5, classifying and counting the number according to the nep size standard division in the industry standard, and reporting the counting result to a computer through a communication interface to finish report output.
14. The method for rapid detection of nep and linter content in cotton of claim 12 wherein the calculating and counting the length of individual cotton fibers specifically comprises:
14.1, processing each image to obtain a curve communicated by a single pixel;
14.2, calculating the pixel length of each fiber by utilizing an edge extraction, filtering and cyclic recursion algorithm, and then calculating the length of each fiber according to the pixel resolution;
14.3, counting the length according to the classification standard of the national standard.
15. The method for rapid detection of nep and linter content in cotton according to claim 14 wherein the calculating the length of each fiber by the edge extraction, filtering and cyclic recursion algorithm comprises:
15.1, designing a symmetrical gradient operator mask matrix Tf for extracting cotton fiber edges:
Figure FDA0004142472060000051
15.2, taking the peripheral neighborhood of each pixel in the image to form a 3X 3 sliding window, wherein the pixel to be judged is a central point P 2,2 Recorded as matrix RM.
Figure FDA0004142472060000052
15.3, calculating each center pixel point P in the image 2,2 Gradient value sf=sum (tf×rm), comparing the gradient value S with a fiber brightness threshold Th5 issued by a computer, and judging a matrix center point P according to the comparison result 2,2 Whether cotton is the pixel or not and generating a Fiber binary image Fiber, wherein the size of the binary image is 4096 multiplied by 512.
Figure FDA0004142472060000053
15.4, performing expansion and corrosion filtering to eliminate salt and pepper noise, wherein the filtering adopts 8-neighborhood convolution filtering. Forming a matrix C by using the Fiber binary image for each pixel point and 8 adjacent domains thereof, and performing matrix multiplication operation on the matrix C and an expansion operator Fb1 to generate a new binary image Fb_Filter1; each pixel point and 8 neighborhood thereof in the new binary image form matrix D, and the matrix D is multiplied by an expansion operator Fb2 to generate
New binary image fb_filter2 is formed and output to the statistics module:
Figure FDA0004142472060000054
Figure FDA0004142472060000055
Figure FDA0004142472060000056
Figure FDA0004142472060000061
Figure FDA0004142472060000062
Figure FDA0004142472060000063
15.5, calculating the external rectangle of each continuous fiber in the Filter2 binary image by utilizing a BLOB standard operator, calculating pixels and values in the rectangle, obtaining the real length of each cotton fiber according to the image resolution of 25 mu m/pixel, counting the number according to three cotton fiber dividing standards in industry standards, dividing the three classified counting numbers by the total number to obtain the percentage of the three fiber lengths, and reporting the calculation result to a computer through a communication interface to finish report output.
16. The method for rapid detection of nep and linter content in cotton according to claim 13, wherein the specific algorithm of step 13.3 comprises:
each pixel point and 8 neighborhood thereof in the binary image form a matrix A, the matrix A is multiplied by a corrosion operator F1, the value of a center point is judged according to a corrosion filtering threshold Th3, and then a new binary image Filter1 is generated; each pixel point and 8 neighborhood thereof in the new binary image form a matrix B, the matrix B is multiplied by an expansion operator F2, the value of a center point is judged according to an expansion filtering threshold Th4, and the output result of the filtering module is Filer2:
Figure FDA0004142472060000064
Figure FDA0004142472060000065
17. the method for rapid detection of nep and linter content in cotton of claim 9 wherein after continuously capturing images of each fiber passing through the test channel and generating continuous image stream data, further comprising dynamically updating the continuous image stream data in a 4096 column x 1024 row format for inquiry or for open-end review of nep images.
18. The method for rapidly detecting nep and linter content in cotton according to claim 9, wherein the gray processing the image stream data of each fiber to obtain gray value data of each image comprises;
counting dynamic big data of the gray values of the image to form a statistical graph with gray values of 0-255 on the abscissa and the statistical times of occurrence of each gray value on the ordinate;
positioning the abscissa corresponding to the maximum value in the statistical curve to obtain a background statistical value in the current image;
comparing the background statistic value with a standard background value interval (55-65) set by a computer: if the statistic value is smaller than the lower limit 55 of the standard value interval, the brightness of the light source is lower, and the light source needs to be turned on; if the statistical value is greater than the upper limit 65 of the standard value interval, the brightness of the light source is higher, and the light source needs to be dimmed; if the statistical value is in the standard background value interval, the current light source brightness is kept unchanged.
CN202310293833.0A 2023-03-24 2023-03-24 Device and method for rapidly detecting nep and linter content in cotton Pending CN116337858A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310293833.0A CN116337858A (en) 2023-03-24 2023-03-24 Device and method for rapidly detecting nep and linter content in cotton

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310293833.0A CN116337858A (en) 2023-03-24 2023-03-24 Device and method for rapidly detecting nep and linter content in cotton

Publications (1)

Publication Number Publication Date
CN116337858A true CN116337858A (en) 2023-06-27

Family

ID=86881783

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310293833.0A Pending CN116337858A (en) 2023-03-24 2023-03-24 Device and method for rapidly detecting nep and linter content in cotton

Country Status (1)

Country Link
CN (1) CN116337858A (en)

Similar Documents

Publication Publication Date Title
CN102353519B (en) Resolving power evaluation method for three-generation dim light image intensifier
CN104198324B (en) Computer vision-based method for measuring proportion of cut leaves in cut tobacco
CN101063659A (en) System for detecting paper flaw of paper sheet
CN103984979B (en) The algae automatic detection counting device and method being imaged without Lenses Diffractive
CN104646315A (en) Intelligent agricultural product sorting machine with aflatoxin detection function
Behera Image-processing in textiles
CN112285504A (en) Multispectral solar-blind narrow-band ultraviolet imager and method for detecting different discharge states by using same
CN1236302C (en) Automatic multispectral cell smear and analyzing instrument and method used for cervical cell analysis
CN104198325B (en) Stem ratio measuring method in pipe tobacco based on computer vision
CN112130316B (en) Multi-channel multi-spectral-band optical filter structure and application and method thereof
CN102288607A (en) Woven fabric count detector based on digital microscope
CN201255634Y (en) On-line rubber bolt external shape quality inspection machine
CN204699963U (en) A kind of Intelligent agricultural product separator with aflatoxins measuring ability
EP0965393B1 (en) Method for discrimination of product units and apparatus therefor
CN109724950A (en) Dynamic super-resolution fluorescence imaging technique with adaptively sampled frame per second
CN103344583A (en) Praseodymium-neodymium (Pr/Nd) component content detection system and method based on machine vision
CN1111736C (en) Fiber quality monitor
CN100508628C (en) Digitized test method for CCD camera signal-to-noise ratio
CN2531393Y (en) Automatic comprehensive testing instrument for yarn
CN108200425B (en) A kind of multi-direction biography letter detection system and method based on TDI linear array detector
CN214211370U (en) Detection system and sorter based on time-sharing coaxial illumination imaging
CN110864882A (en) Multispectral ultraviolet sensitivity detection system and method
CN116337858A (en) Device and method for rapidly detecting nep and linter content in cotton
CN202548063U (en) System for automatically acquiring light-transmitting image of optic fiber plate
CN109100364A (en) A kind of tunnel defect monitoring system and monitoring method based on spectrum analysis

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