CN113393460B - Cut tobacco quality parameter detection method and system based on image processing - Google Patents
Cut tobacco quality parameter detection method and system based on image processing Download PDFInfo
- Publication number
- CN113393460B CN113393460B CN202110933900.1A CN202110933900A CN113393460B CN 113393460 B CN113393460 B CN 113393460B CN 202110933900 A CN202110933900 A CN 202110933900A CN 113393460 B CN113393460 B CN 113393460B
- Authority
- CN
- China
- Prior art keywords
- tobacco
- cut tobacco
- length
- shred
- shreds
- 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
Images
Classifications
-
- 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
- G06T7/0004—Industrial image inspection
-
- 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
- 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
- G01N9/00—Investigating density or specific gravity of materials; Analysing materials by determining density or specific gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pathology (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Manufacturing Of Cigar And Cigarette Tobacco (AREA)
Abstract
The invention provides a tobacco shred quality parameter detection method and system based on image processing and aiming at tobacco shred density and size distribution, which can quickly measure a large amount of scattered tobacco shreds, establish a tobacco shred density and tobacco shred size distribution interval model, and have the advantages of high measurement speed and high precision. The tobacco shred quality parameter detection method based on image processing comprises the following steps: building a physical system, obtaining pictures, calibrating units, reading a plurality of tobacco shred images, inputting the quality, thickness and pixel size ratio of tobacco shreds, preprocessing the images, outputting the volume, density and size of the tobacco shreds, outputting a tobacco shred density distribution curve and the like; during image processing, the minimum external rectangles of different tobacco shreds are marked by dividing the connected domains of the scattered tobacco shreds, the length of the diagonal line is calculated according to the length and the width of the rectangles, corresponding pixel threshold values are set, the size intervals of the whole tobacco shreds, the middle tobacco shreds and the broken tobacco shreds are obtained, and the quality indexes of the tobacco shreds are evaluated according to the comprehensive performance evaluation indexes of the tobacco shreds.
Description
Technical Field
The invention belongs to the technical field of physical experiments and tobacco mechanical equipment, and particularly relates to a tobacco shred size and density interval model calculation method based on an image processing algorithm.
Background
The length and width shape, the stem content, the whole tobacco shred rate, the medium tobacco shred rate, the broken tobacco shred rate and the tobacco shred density of the tobacco shreds are important cigarette process indexes and have important influence on the evaluation of the cigarette quality and the air separation performance. Most of the traditional measurement modes are manual measurement, the labor intensity is high, the sampling period is long, the evaluation objectivity is poor, and the traditional measurement modes are not in accordance with the modern design concept of tobacco enterprises. The interval models of various tobacco attributes do not form a uniform enterprise standard, and only the method for measuring the width and the density of the tobacco shreds is researched.
The width of the tobacco shreds is generally measured based on a glue stick method and a projector measurement method. The first method is that the rubber rod is put at the entrance of a filament cutter and cut into sheets, and the width of the sheet rubber sheet is measured by a micrometer; and secondly, the cut tobacco is placed on a workpiece table of a projector, a screw rod is manually adjusted to be aligned with the datum line of the starting point and the end point of the cut tobacco to be measured, and the mean value is calculated after multi-point measurement. Because the number of the tobacco shreds to be measured is large and the shape is irregular, the labor intensity of a measurer is high, the speed is low, the result stability is poor, and the traditional measuring method cannot accurately reflect the width of the tobacco shreds. At present, the method for measuring the width of the cut tobacco in China is relatively advanced CCD image measurement, but only the width of the cut tobacco is analyzed, and a measuring method aiming at the size distribution interval of the cut tobacco is not provided.
Aiming at the tobacco shred density measurement, a relatively advanced measurement method in China is a microwave measurement method. When the tobacco shreds pass through the microwave resonant cavity, the sample sections with different densities and moisture cause the change of the amplitude-frequency characteristics of the microwave resonant cavity, the unit density is calculated based on different amplitude values, and the average density and the distribution uniformity can be counted. However, the microwave measurement method can only be applied to the density of a formed cigarette containing a cigarette paper, and the density of the cut tobacco after packaging changes to some extent, so that the cut tobacco cannot be equivalent to the cut tobacco in a natural state. For the traditional manual density measurement, the measurement error is relatively larger because the shape of the cut tobacco is irregular and the volume is difficult to measure.
The method is based on an image processing technology, and the size and density interval of the cut tobacco produced in a natural state are effectively measured. Under the condition of meeting the requirement of precision, the experimental cost is lower, the device arrangement is simple, and the measuring method is simpler and more universal.
Disclosure of Invention
The invention aims to provide a nondestructive measurement system for tobacco shred density and size distribution, which can quickly measure a large amount of scattered tobacco shreds, establish a tobacco shred density and tobacco shred size distribution interval model, and has high measurement speed and high precision.
In order to realize the purpose of the invention, the technical scheme adopted by the invention is as follows:
the cut tobacco quality parameter detection method based on image processing comprises the following steps:
acquiring a tobacco shred image; the tobacco shreds in the tobacco shred images are not overlapped with each other;
secondly, preprocessing and segmenting the tobacco shred images to obtain tobacco shred areas;
step three, obtaining the minimum external rectangle of each tobacco shred area, and taking the length of the diagonal line of the minimum external rectangle as the length of the tobacco shreds;
according to the length of the cut tobacco and the preset size intervals of the whole cut tobacco, the middle cut tobacco and the broken cut tobacco, respectively obtaining the quantity of the cut tobacco in the size areas of the whole cut tobacco, the middle cut tobacco and the broken cut tobacco in the cut tobacco image to obtain the whole cut tobacco rate, the middle cut tobacco rate and the broken cut tobacco rate of the cut tobacco;
calculating the evaluation indexes of the introduced mean value, the variation coefficient, the skewness and the kurtosis according to the whole tobacco shred rate, the medium tobacco shred rate and the broken tobacco shred rate obtained by combining the lengths of the tobacco shreds;
and sixthly, calculating the comprehensive performance evaluation index of the tobacco shreds according to the introduced average value, variation coefficient, skewness and kurtosis evaluation index, and obtaining a result.
Further, in the first step, the method for obtaining the tobacco shred image comprises the following steps:
the method comprises the following steps of arranging a digital camera, an LED light source and a test bed, wherein the digital camera is electrically connected with an image acquisition computer, and the image acquisition computer is electrically connected with a display; the test bed is positioned below the digital camera, a horizontal ruler is fixed on one side above the test bed, and a white board is fixed on the other side; the LED light sources are circumferentially arranged on two sides of the digital camera in a surrounding manner;
and the digital camera uploads the cut tobacco image obtained by shooting to an image acquisition computer.
In the third step, the length of the diagonal line of the minimum circumscribed rectangle is obtained by the following method:
step a, placing a known actual area ofS 0 As a basis for unit calibration, the object of (1) takes an imageP 0 ；
B, reading and storing the image shot by the digital camera by the computer;
c, in the computer calibration system, reading the shot calibration pictureP 0 Obtaining the vertical projection pixel area of the objectS；
D, obtaining the area size ratio of the pixelsSca：
E, according to the area size ratio of the pixelsScaObtaining the pixel length dimension ratioLca：
F, multiplying the pixel length corresponding to the minimum circumscribed rectangle diagonal by the size ratio of the pixel lengthLcaAnd obtaining the length of the minimum circumscribed rectangle diagonal.
In the step one, before the cut tobacco image is obtained, the cut tobacco image is weighed to obtain the second stepiQuality of cut tobaccoM i So that the quality of each group of cut tobacco has relative errorcThe following conditions are met:
wherein ΔmThe precision of the precision balance is obtained.
In step f, the total area of the tobacco shreds in each group of pictures is countedS i And the quality of the cut tobacco and the average thickness of the cut tobacco are comparedhAnd pixel size ratio, transmitting to image acquisition computer, calculatingiDensity of cut tobacco group
And average density of all tobacco shreds
Outputting the result and drawing a density distribution curve, the firstiDensity of cut tobacco group
And average density of all tobacco shreds
The calculation formula of (a) is as follows:
wherein
Is as followsiThe density of the tobacco shred group,
Is the average density of all tobacco shreds,M i Is the quality of the tobacco shred,S i The total area of the tobacco shreds in each group of pictures,hThe average thickness of the tobacco shreds and n is the number of the tobacco shreds.
In a further improvement, the average thickness of the cut tobaccohObtained by manual measurement.
In the fifth step, the introduced evaluation indexes of mean value, coefficient of variation, skewness and kurtosis are as follows: analyzing the centralized position, the dispersion degree and the distribution of the length of the cut tobacco size distribution, such as normal distribution or skewed distribution, and the like, wherein the calculation formula is as follows:
mean value:
coefficient of variation:
skewness:
kurtosis:
wherein
Is the average value of the length of the tobacco shreds,x i The length of the ith tobacco shred, n is the total quantity of the tobacco shreds,s 2 Is the length variance of the tobacco shreds,sIs the standard deviation of the length of the cut tobacco,CVIs a coefficient of variation,G 1 Is a deviation degree,G 2 Is the kurtosis,u k Is a samplekThe central moment of the step,
Is the 3-order central moment of the sample,
The order of 4-order central moment of the sample is taken, and k is the order value of the several-order central moment of the sample.
Further, in the sixth step, the comprehensive performance evaluation indexPJZBThe calculation formula of (a) is as follows:
whereinx b The target length of the tobacco shreds and the corresponding weight of different indexes are set as i =1, 2, 3 and 4;CVis a coefficient of variation,G 1 Is a deviation degree,G 2 Is the kurtosis,
Is the average value of the length of the cut tobacco.
The improved tobacco shred quality evaluation device comprises an image acquisition unit, an image processing unit, an image measuring unit and a tobacco shred quality evaluation unit, and has the following functions:
the image acquisition unit is used for acquiring a tobacco shred image;
the image processing unit is used for preprocessing the tobacco shred images to obtain each tobacco shred area;
the image measuring unit is used for measuring the area of the tobacco shred;
the tobacco shred quality evaluation unit is used for obtaining the evaluation indexes of the density, the whole tobacco shred rate, the medium tobacco shred rate, the broken tobacco rate and the comprehensive performance of the tobacco shredsPJZB。
Compared with the prior art, the invention has the advantages that:
the density can be measured without damaging the tobacco shreds in the natural state of the tobacco shreds, an effective density interval model is established, and the method is high in precision, simple to operate and high in universality.
(2) Different threshold intervals can be set according to the length and the size of the cut tobacco in the image, the component proportion, the whole tobacco shred rate, the medium tobacco shred rate and the broken tobacco shred rate of cut tobacco, cut stems and thin slices in a cigarette or mixed cut tobacco sample are counted, and an effective size interval model is established.
Drawings
FIG. 1 is a cut tobacco density and length interval measuring system based on image processing;
FIG. 2 shows a technical scheme for measuring the tobacco density and length intervals;
FIG. 3 is a cut tobacco sample binaryzation;
FIG. 4 is a density profile;
FIG. 5 cut tobacco density probability distribution;
FIG. 6 is a minimum circumscribed rectangle of tobacco shreds;
FIG. 7 probability density distribution of mixed tobacco length;
FIG. 8 is a frequency distribution of mixed tobacco lengths;
in the figure: 1. the device comprises a display, 2, an image acquisition computer, 3, an LED light source, 4, a white board, 5, a digital camera, 6, a test bed, 7, a level bar, 8 and a tobacco shred sample.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined purpose, the following detailed description will be made on the method and system for detecting tobacco shred quality parameters based on image processing according to the present invention with reference to the accompanying drawings and preferred examples as follows:
example 1
Fig. 1 shows a tobacco density and size distribution interzone measuring system based on image processing. The two LED light sources are wide in illumination range, high in intensity and good in uniformity, and are the basis for improving the tobacco leaf imaging quality and the measurement accuracy, and the white board is arranged below the light sources, so that the tobacco leaf color difference is increased, and the tobacco leaf imaging quality can be improved. The image acquisition computer can carry out analysis and measurement to the experimental picture of gathering, accomplishes size distribution's measurement, combines precision balance's experiment can realize the range statistics of pipe tobacco density.
The flow of the tobacco shred density and size interval measuring system is as follows: the method comprises the steps of establishing a physical system, obtaining pictures, calibrating units, reading in a plurality of groups of tobacco shred images, inputting the quality, thickness and pixel size ratio of tobacco shreds, preprocessing the images, outputting the volume, density and size of the tobacco shreds, outputting a tobacco shred density distribution curve and the like. The specific implementation steps are as follows:
(a) an experiment system is built according to the scheme of the figure 1, and a digital camera and a computer measuring system are connected.
Firstly, the LED light sources are symmetrically arranged at the left and right sides and are lower than the digital camera, so that the camera can prevent the LED light from being blocked, and the brightness of the background white board is uniform without shadow. Secondly, the mirror surface of the digital camera is adjusted to be parallel to the white board, so that the acquired tobacco shred image is in a certain proportion to the actual tobacco shred. The center of the digital camera should coincide with the center of the whiteboard, so that the focus is consistent.
(b) Fixing the digital camera and the white board, and placing an area on the white boardS 0 As a basis for unit calibration, the object of (1) takes an imageP 0 。
Preparation of vertical projection areaS 0 The object is placed on the white board, the center point of the object is superposed with the mark point of the white board, one or more pictures are shot by the digital camera, and the image with better shooting effect is selectedP 0 Mark unit, obtaining pixel and smokeThe actual filament size ratio.
(c) Randomly arranging a certain amount of tobacco shreds on a precision balance, and recording the mass of the tobacco shredsM i 。
Firstly, the precision balance is horizontally placed on a test bed, a nut at the bottom of the balance is adjusted to a horizontal position by combining a level gauge, and the balance is calibrated and peeled by a calibration weight. Then, a certain amount of cut tobacco is put on a precision balance, and the total mass of the cut tobacco is recordedM i (g) In that respect At this time, the quantity of the cut tobacco cannot be too small, so that the relative error of the quality of the cut tobacco is prevented from being larger. Because the tobacco shreds are susceptible to moisture, the experiment needs to be performed in a dry environment or the tobacco shreds are stored in sealed bags. Relative errorcComprises the following steps:
△mthe relative error should be less than or equal to 1% for precision balance precision, and the precision Delta of the precision balance of the systemm=10-3g。
(d) And (c) flatly paving the cut tobacco measured in the step (c) in the central area of the white board, wherein the cut tobacco cannot be overlapped, and shooting an image Pi (the density of the cut tobacco is increased after the cut tobacco is easy to wet, so that the cut tobacco cannot be reused) by using a digital camera.
(e) Repeating the steps (c) and (d) for n times to obtain the quality of n groups of tobacco shreds and corresponding images.
(f) Reading and storing the image shot by the digital camera by the computer;
(g) in a computer unit calibration system, a captured calibration picture is readP 0 Obtaining the vertical projection pixel area of the objectS。
According to the obtained pixel areaSCalculating the area size ratioSca：
Length to size ratioLcaComprises the following steps:
(h) and reading the N taken pictures, P1 … PN, by the aid of the written density and size interval measuring system. Recording size ratioLcaThe quality M of the tobacco leaves in each group of imagesiAverage thickness of tobacco shredsh. The calculation process of the tobacco shred density and length interval measuring system is as follows:
(1) and running a program to preprocess the image.
(2) Graying the picture, namely converting the color image into a grayed image with a small memory;
(3) removing local noise points in the gray level image through a median filtering denoising technology;
(4) the gray scale image is converted into a binary image, and background removal is performed. At this time, the shot object is black, and the background is white;
(5) performing binary image inversion processing on the image, wherein the object is displayed as white and the background is black;
(6) correcting the tobacco shred outline and the middle hole seam through morphological closed operation;
(7) setting a pixel threshold value, deleting a small-area pixel part, and removing irrelevant fine objects such as dust and the like;
(8) dividing tobacco shred areas by multi-connected areas;
(9) counting the total area S of tobacco shreds in each group of picturesiAnd calculating the density and the average density and outputting the result, wherein the calculation formula is as follows:
(10) selecting one or more groups of tobacco shred length measurement samples for measurement;
(11) and calculating the length of the diagonal line of the minimum external rectangle of each tobacco shred, and taking the length as the length of the tobacco shred.
Fig. 2 is a flow chart of tobacco density and length interval measurement. The tobacco shred density and length interval measuring system based on image processing carries out nondestructive measurement on the tobacco shred density and length, avoids interference of human factors, can quickly acquire related data and has high reliability.
The length of the cut tobacco is obtained through image processing and measurement, corresponding pixel threshold values are set according to the size intervals of the whole cut tobacco, the middle cut tobacco and the broken cut tobacco, and the whole cut tobacco rate, the middle cut tobacco rate and the broken cut tobacco rate of the cut tobacco can be counted. In order to evaluate the tobacco shred size distribution model, evaluation indexes such as a mean value, a variation coefficient, skewness, kurtosis and the like are introduced, and the centralized position, the dispersion degree, the normal distribution or the skewed distribution of the tobacco shred length distribution and the like are analyzed.
(1) Mean value
Assuming that n cut tobacco length observations are analyzed, the arithmetic mean is called the sample mean and can be expressed as
In addition, common sample variance of variation degree for describing size data of all tobacco shredss 2 Or standard deviation of the samplesCan be represented as
(2) Coefficient of variation
Using coefficient of variationCVDifferent indexes in the same sample or the same index in different samples can be compared according toCVCan rank the degree of variation of the index.
(3) Skewness and kurtosis
Skewness and kurtosis are measures of skewness and tail weight of the characterization data. The skewness is calculated by the formula
Whereinu k The k-th order central moment of the sample. Skewness is an index that characterizes the symmetry of data. Skewness of data symmetrical about meanG 1 =0, more dispersed data on right side (i.e. right tail length) skewness is positive (G 1 >0) The left-side more dispersed data (left tail length) skewness is negative (G 1 <0）。
The kurtosis is calculated by the formula
Kurtosis when the overall distribution of data is normalG 2 Is approximately 0; when the distribution is more dispersed than the tail of the normal distribution, the kurtosis is positive (G 2 >0) Otherwise, the kurtosis is negative (G 2 <0). When kurtosis is positive (G 2 >0) Meanwhile, the data of the extreme ends on the two sides are more (thick tail); when kurtosis is negative (G 2 <0) When there is less data on both sides (thin tail).
The implementation mode is as follows:
1. unit calibration
The tobacco shred physical property measuring system based on image processing reads a calibration object image with a known area and a real object area S0=106×74=7844mm2. Operating the system, inputting the actual area of the object, performing binarization processing on the calibration object, and outputting the pixel area of the calibration objectS=430458 pixel × pixel and binary map. Calculating the pixel length size ratio of the pixelLca=0.135。
2. Measurement of tobacco shred Density
In the measurement of the tobacco shred density, the tobacco shreds which are measured in the precision balance are flattened and placed on a white board, the tobacco shreds cannot be overlapped, a picture is shot by a digital camera, and the picture is recorded as Pi. This experiment measured the quality of 9 groups of tobacco shreds, and each group of tobacco shredsThe filaments were randomly laid 3 times on the white board, so that an image of 9 × 3=27 cut tobacco was obtained. Photographs of 27 tobacco shreds were takenP1…P27Calculating the average size ratio to be 0.091, wherein the mass of each group is 0.18 and 0.181, and the average thickness of the cut tobacco is 0.12 mm; based on the tobacco density measuring system, simultaneously reads the imageP1…P27And inputting relevant parameters, and calculating the volume, density distribution curve, average density and the like of the cut tobacco, as shown in figures 3, 4 and 5.
3.3. The whole rate, medium rate and broken rate of the mixed cut tobacco are as follows:
the minimum external rectangles of different tobacco shreds are marked by dividing the connected domains of the scattered tobacco shreds, as shown in figure 6; the length of the diagonal line is calculated by the length and the width of the rectangle, and corresponding pixel thresholds are set according to the size intervals of the whole tobacco, the medium tobacco and the broken tobacco, so that the whole tobacco rate, the medium tobacco rate and the broken tobacco rate of the tobacco can be counted: it is calculated that =8.5079, which is,s=7.4938,CV=88.08,G 1 =1.487,G 2 =5.1562. prepared fromG 1 >0, the data on the right side is more scattered, and the right tail is long; from G2>0, the data of the two side ends are more, and the tail is thick;CVlarger, the degree of variation of the index is large.
The four evaluation indexes of the tobacco mean value, the variation coefficient, the skewness and the kurtosis are different in function, the tobacco mean value is a size index, the other three indexes are difference indexes, and the difference between the processed tobacco and the target tobacco cannot be intuitively reflected. In order to quantify the tobacco shred quality parameters, a comprehensive performance evaluation index (standard tobacco shred deviation degree) calculation formula is provided:
in the formula (I), the compound is shown in the specification,x b the target length of the cut tobacco is set as,
are the corresponding weights of the different indexes. The weight is sequentially as follows because the mean value and the coefficient of variation have relatively great influence on the quality parameters of the cut tobacco
. The skewness and kurtosis are less affected, the weights are sequentially
。
Evaluation index of comprehensive performancePJZBThe closer to 0, the closer the actual tobacco shred quality processed by the tobacco factory is to the target tobacco shred quality. Assuming that 5mm is the target length of the cut tobacco, the comprehensive performance evaluation index (standard cut tobacco deviation) value of the current cut tobacco sample is as follows:
drawing a probability density distribution curve of the tobacco shreds, and fitting the probability density curve of the tobacco shreds into a normal distribution function curve as shown in FIG. 7:
wherein a is a standard deviation of the values of a,
is taken as the mean value of the average value,xis an independent variable.
By referring to the chemical characteristics of cut tobacco, cut tobacco can be divided into three categories: the cut tobacco with the length of more than 6.00mm is gathered into one type (whole tobacco), the cut tobacco with the length of 3.00-6.00mm is gathered into one type (medium tobacco), and the cut tobacco with the length of less than 3.00mm is gathered into one type (broken tobacco). Three areas are divided in the cut tobacco length frequency distribution diagram, as shown in fig. 8, the broken tobacco rate is 0.265, the medium tobacco rate is 0.220, and the whole tobacco rate is 0.515.
The above description is only a preferred embodiment of the present invention, and should not be taken as limiting the invention in any way, and any simple modification, equivalent change and modification made by those skilled in the art according to the technical spirit of the present invention are still within the technical scope of the present invention without departing from the technical scope of the present invention.
Claims (6)
1. The cut tobacco quality parameter detection method based on image processing is characterized by comprising the following steps: the method comprises the following steps:
acquiring a tobacco shred image; the tobacco shreds in the tobacco shred images are not overlapped with each other;
secondly, preprocessing and segmenting the tobacco shred images to obtain tobacco shred areas;
step three, obtaining the minimum external rectangle of each tobacco shred area, and taking the length of the diagonal line of the minimum external rectangle as the length of the tobacco shreds;
according to the length of the cut tobacco and the preset size intervals of the whole cut tobacco, the middle cut tobacco and the broken cut tobacco, respectively obtaining the quantity of the cut tobacco in the size areas of the whole cut tobacco, the middle cut tobacco and the broken cut tobacco in the cut tobacco image to obtain the whole cut tobacco rate, the middle cut tobacco rate and the broken cut tobacco rate of the cut tobacco;
calculating the evaluation indexes of the introduced mean value, the variation coefficient, the skewness and the kurtosis according to the length of the cut tobacco; the introduced evaluation indexes of mean value, coefficient of variation, skewness and kurtosis are as follows: analyzing whether the distribution of the centralized position, the dispersion degree and the length of the cut tobacco size distribution is normal distribution or skewed distribution, wherein the calculation formula is as follows:
mean value:
coefficient of variation:
wherein
Is the length of the cut tobaccoThe average value of degree,x i The length of the ith tobacco shred, n is the total quantity of the tobacco shreds,s 2 Is the length variance of the tobacco shreds,sIs the standard deviation of the length of the cut tobacco,CVIs a coefficient of variation,G 1 Is a deviation degree,G 2 Is the kurtosis,u k Is a samplekThe central moment of the step,
Is the 3-order central moment of the sample,
Taking the 4-order central moment of the sample, and taking k as the order value of the several-order central moment of the sample;
calculating the comprehensive performance evaluation index of the tobacco shreds according to the introduced average value, variation coefficient, skewness and kurtosis evaluation indexes, and obtaining a result;
evaluation index of comprehensive performancePJZBThe calculation formula of (a) is as follows:
whereinx b The target length of the tobacco shred,
I =1, 2, 3, 4 for the corresponding weights of the different indices;CVis a coefficient of variation,G 1 Is a deviation degree,G 2 Is the kurtosis,
Is the average value of the length of the cut tobacco.
2. The image processing-based tobacco shred quality parameter detection method according to claim 1, wherein the image processing-based tobacco shred quality parameter detection method comprises the following steps: in the first step, the method for acquiring the tobacco shred image comprises the following steps:
the method comprises the following steps of arranging a digital camera, an LED light source and a test bed, wherein the digital camera is electrically connected with an image acquisition computer, and the image acquisition computer is electrically connected with a display; the test bed is positioned below the digital camera, a horizontal ruler is fixed on one side above the test bed, and a white board is fixed on the other side; the LED light sources are circumferentially arranged on two sides of the digital camera in a surrounding manner;
and the digital camera uploads the cut tobacco image obtained by shooting to an image acquisition computer.
3. The image processing-based tobacco shred quality parameter detection method according to claim 2, wherein the image processing-based tobacco shred quality parameter detection method comprises the following steps: in the third step, the method for obtaining the length of the minimum external rectangle diagonal line is as follows:
step a, placing a known actual area ofS 0 As a basis for unit calibration, the object of (1) takes an imageP 0 ；
B, reading and storing the image shot by the digital camera by the computer;
c, in the computer calibration system, reading the shot calibration pictureP 0 Obtaining the vertical projection pixel area of the objectS；
D, obtaining the area size ratio of the pixelsSca：
E, according to the area size ratio of the pixelsScaObtaining the pixel length dimension ratioLca：
F, multiplying the pixel length corresponding to the minimum circumscribed rectangle diagonal by the size ratio of the pixel lengthLcaAnd obtaining the length of the minimum circumscribed rectangle diagonal.
4. As claimed in claim 3The tobacco shred quality parameter detection method based on image processing is characterized by comprising the following steps: in the first step, before the cut tobacco image is obtained, the second step is obtained by weighingiQuality of cut tobaccoM i So that the quality of each group of cut tobacco has relative errorcThe following conditions are met:
wherein ΔmThe precision of the precision balance is obtained.
5. The image processing-based tobacco shred quality parameter detection method according to claim 4, wherein the image processing-based tobacco shred quality parameter detection method comprises the following steps: in step f, counting the total area of tobacco shreds in each group of picturesS i And the quality of the cut tobacco and the average thickness of the cut tobacco are comparedhAnd pixel size ratio, transmitting to image acquisition computer, calculatingiDensity of cut tobacco groupρ i And average density of all tobacco shredsρOutputting the result and drawing a density distribution curve, the firstiDensity of cut tobacco groupρ i And average density of all tobacco shredsρThe calculation formula of (a) is as follows:
wherein
Is as followsiThe density of the tobacco shred group,
Is the average density of all tobacco shreds,M i Is as followsiThe quality of the tobacco shred group,S i Is the total area of the tobacco shreds in the ith group of pictures,hIs the average thickness of the tobacco shreds, and n is the number of tobacco shreds.
6. The image processing-based tobacco shred quality parameter detection method according to claim 5, wherein the image processing-based tobacco shred quality parameter detection method comprises the following steps: the average thicknesshObtained by manual measurement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110933900.1A CN113393460B (en) | 2021-08-16 | 2021-08-16 | Cut tobacco quality parameter detection method and system based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110933900.1A CN113393460B (en) | 2021-08-16 | 2021-08-16 | Cut tobacco quality parameter detection method and system based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113393460A CN113393460A (en) | 2021-09-14 |
| CN113393460B true CN113393460B (en) | 2021-11-30 |
Family
ID=77622663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110933900.1A Active CN113393460B (en) | 2021-08-16 | 2021-08-16 | Cut tobacco quality parameter detection method and system based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113393460B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115836976B (en) * | 2023-02-23 | 2023-04-25 | 四川新荷花中药饮片股份有限公司 | Production method of intelligent control system for producing toxic decoction pieces of Araceae |
| CN116223440A (en) * | 2023-05-08 | 2023-06-06 | 四川威斯派克科技有限公司 | Near infrared detection device for tobacco raw material proportioning |
| CN117974642B (en) * | 2024-03-28 | 2024-06-11 | 富钛金属科技(昆山)有限公司 | Real-time monitoring system and method for automatic pouring quality of sealant |
| CN118080280B (en) * | 2024-04-24 | 2024-07-09 | 青岛英诺包装科技有限公司 | Monitoring system for film processing based on Internet of things |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5353357A (en) * | 1992-12-04 | 1994-10-04 | Philip Morris Incorporated | Methods and apparatus for inspecting the appearance of substantially circular objects |
| EP1695070A1 (en) * | 2003-12-12 | 2006-08-30 | Focke & Co. (GmbH & Co. KG) | Method and device for testing the surface of a transported material strand in the tobacco processing industry |
| CN102829730A (en) * | 2012-09-05 | 2012-12-19 | 江苏中烟工业有限责任公司 | Measuring method for shredding width |
| CN103454200A (en) * | 2013-08-15 | 2013-12-18 | 中国石油天然气股份有限公司 | Method and device for measuring core pore parameters by using computed tomography technology |
| CN106052571A (en) * | 2016-07-21 | 2016-10-26 | 云南中烟工业有限责任公司 | Device and method for intelligent rapid detection of length and distribution of leaf silk for cigarette |
| CN106770303A (en) * | 2017-03-31 | 2017-05-31 | 河南农业大学 | Cigarette shreds structure characterization methods based on graphical analysis |
| CN110782440A (en) * | 2019-10-22 | 2020-02-11 | 华中农业大学 | Crop grain character measuring method |
| WO2020151823A1 (en) * | 2019-01-24 | 2020-07-30 | Hauni Maschinenbau Gmbh | Quality control of rod-shaped products of the tobacco processing industry |
-
2021
- 2021-08-16 CN CN202110933900.1A patent/CN113393460B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5353357A (en) * | 1992-12-04 | 1994-10-04 | Philip Morris Incorporated | Methods and apparatus for inspecting the appearance of substantially circular objects |
| EP1695070A1 (en) * | 2003-12-12 | 2006-08-30 | Focke & Co. (GmbH & Co. KG) | Method and device for testing the surface of a transported material strand in the tobacco processing industry |
| CN102829730A (en) * | 2012-09-05 | 2012-12-19 | 江苏中烟工业有限责任公司 | Measuring method for shredding width |
| CN103454200A (en) * | 2013-08-15 | 2013-12-18 | 中国石油天然气股份有限公司 | Method and device for measuring core pore parameters by using computed tomography technology |
| CN106052571A (en) * | 2016-07-21 | 2016-10-26 | 云南中烟工业有限责任公司 | Device and method for intelligent rapid detection of length and distribution of leaf silk for cigarette |
| CN106770303A (en) * | 2017-03-31 | 2017-05-31 | 河南农业大学 | Cigarette shreds structure characterization methods based on graphical analysis |
| WO2020151823A1 (en) * | 2019-01-24 | 2020-07-30 | Hauni Maschinenbau Gmbh | Quality control of rod-shaped products of the tobacco processing industry |
| CN110782440A (en) * | 2019-10-22 | 2020-02-11 | 华中农业大学 | Crop grain character measuring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113393460A (en) | 2021-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113393460B (en) | Cut tobacco quality parameter detection method and system based on image processing | |
| CN106770303B (en) | Image analysis-based cigarette cut tobacco structure characterization method | |
| CN104198325B (en) | Stem ratio measuring method in pipe tobacco based on computer vision | |
| CN104256882A (en) | Method for measuring proportion of reconstituted tobacco in cut tobacco on basis of computer vision | |
| CN112560355A (en) | Method and device for predicting Mach number of wind tunnel based on convolutional neural network | |
| CN113376170B (en) | Calibration method and calibration block of product appearance detection equipment | |
| CN102800096A (en) | Robustness estimation algorithm of camera parameter | |
| CN111986201B (en) | Method and device for measuring and calculating density of surface algae of insulator | |
| CN102914620A (en) | Determination method for ration of reconstituted tobacco finished product | |
| CN111415349A (en) | Method for detecting polyester filament yarn based on image processing technology | |
| CN113884492B (en) | Ringelmann blackness calibration and detection method and device for motor vehicle exhaust | |
| CN109030755A (en) | The detection device and method of grain moisture content, mass of 1000 kernel and bulk density based on image processing techniques | |
| CN205538538U (en) | Papermaking method reconstituted tobacco product ration spot test appearance | |
| CN113838123B (en) | Method for measuring appearance characteristics of cut tobacco based on image processing | |
| CN116164818A (en) | Determination method, device, equipment and storage medium for measuring uncertainty | |
| CN113989513A (en) | Method for recognizing reading of square pointer type instrument | |
| CN117274168A (en) | Tobacco shred detection method based on machine vision | |
| CN117451580B (en) | Method for analyzing extrusion stability of wiredrawing protein | |
| JPS60205683A (en) | Method and device for discrminating external appearance | |
| CN114047285B (en) | Method and standard suite for measuring parameters of cigarette combustion cone | |
| CN115236105A (en) | Nondestructive testing method for cigarette end content | |
| CN116433603A (en) | Method for detecting tobacco shred morphology in cigarette | |
| CN113139951B (en) | Method, system and equipment for characterizing attributes of tobacco lamina and computer readable storage medium | |
| CN117876783A (en) | Wool velvet grading method based on machine vision | |
| CN115931902A (en) | Food texture characteristic detection method and device based on air needle-structured light imaging |
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 |