CN115356297B - Method and device for identifying slightly green flue-cured tobacco leaves - Google Patents
Method and device for identifying slightly green flue-cured tobacco leaves Download PDFInfo
- Publication number
- CN115356297B CN115356297B CN202211023788.9A CN202211023788A CN115356297B CN 115356297 B CN115356297 B CN 115356297B CN 202211023788 A CN202211023788 A CN 202211023788A CN 115356297 B CN115356297 B CN 115356297B
- Authority
- CN
- China
- Prior art keywords
- tobacco leaves
- spectral
- information
- identifying
- flue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- 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
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/55—Specular reflectivity
-
- 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
- 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
- G01N2021/8466—Investigation of vegetal material, e.g. leaves, plants, fruits
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Manufacture Of Tobacco Products (AREA)
Abstract
本发明提出一种烤烟烟叶微青的鉴别方法及装置，涉及卷烟生产技术设备领域，采集待鉴别烟叶的原始光谱图像，对所述原始光谱图像进行预处理以获得预处理后的光谱信息，通过所述预处理后的光谱信息获取所述待鉴别烟叶其叶面各像元的光谱曲线，计算各像元的光谱曲线的曲率半径，根据各所述曲率半径提取相应的像元并显示为青烟区域。本发明提出的烤烟烟叶微青的鉴别方法及装置能够快速鉴别烟叶中的微青部分。
The present invention proposes a method and device for identifying slightly green tobacco leaves, which relates to the field of cigarette production technology and equipment. The original spectral image of the tobacco leaves to be identified is collected, and the original spectral image is preprocessed to obtain the preprocessed spectral information. The spectral curve of each pixel of the leaf surface of the tobacco leaves to be identified is obtained through the preprocessed spectral information, and the curvature radius of the spectral curve of each pixel is calculated. According to each curvature radius, the corresponding pixel is extracted and displayed as a green tobacco area. The method and device for identifying slightly green tobacco leaves proposed by the present invention can quickly identify the slightly green part in the tobacco leaves.
Description
Technical Field
The invention relates to the technical field of cigarette production equipment, in particular to a method and a device for identifying the slight green of flue-cured tobacco leaves.
Background
In the tobacco leaf production, the primary flue-cured tobacco leaves have slight green phenomena at different degrees due to the baking technology and the like. The micro-strip green tobacco leaves have poor smoking quality, and after the micro-strip green tobacco leaves are mixed with the positive group tobacco leaves, miscellaneous gas and irritation are increased, so that the smoking quality is obviously reduced (blue jade freshness, shallow talk of flue-cured tobacco containing green has influence on tobacco quality [ J ], light industry science and technology.201506-1116-02). Therefore, there is a need to differentiate micro-strip green tobacco leaves, utilizing industry-selected uses and reducing the difficulty of industrial formulations.
The quality evaluation of the appearance of the tobacco leaves mainly takes manpower as a main factor, namely, the grade quality of the tobacco leaves is judged through hand touch and eye observation. In the classification of tobacco leaves, the color of tobacco leaves is an important evaluation index of the quality grade thereof. The basic colors of the tobacco leaves comprise lemon yellow, orange and reddish brown, the non-basic colors of the tobacco leaves comprise cyan yellow, slight cyan and variegated colors, the accurate identification of the colors is very high in requirements on quality of evaluation experts, and the tobacco leaves are easily interfered by various external factors in practical application. In recent years, technicians have made many standardized and data improvements for evaluating flue-cured tobacco quality. For example, chinese patent CN 113010848A proposes a method for processing appearance quality evaluation data of flue-cured tobacco, which performs score addition and subtraction according to leaf green content, and finally performs qualitative and quantitative description on appearance quality of flue-cured tobacco through score division range, but identification of micro-strip green tobacco leaves is most difficult to determine, and most of them also need to rely on other appearance quality factors for comprehensive evaluation.
In view of this, the present inventors have devised a method and apparatus for identifying a slight green of cured tobacco leaves through repeated experiments according to production design experiences conducted in the art and related fields for many years, so as to solve the problems existing in the prior art.
Disclosure of Invention
The invention aims to provide a method and a device for identifying the slight green of flue-cured tobacco leaves, which can rapidly identify the slight green part in the tobacco leaves.
In order to achieve the above-mentioned aim, the invention provides a flue-cured tobacco leaf slight-blue identification method, wherein, the original spectrum image of tobacco leaf to be identified is collected, the original spectrum image is preprocessed to obtain preprocessed spectrum information, the spectrum curve of each pixel of the leaf surface of the tobacco leaf to be identified is obtained through the preprocessed spectrum information, the curvature radius of the spectrum curve of each pixel is calculated, and the corresponding pixel is extracted according to each curvature radius and displayed as a blue-green smoke area.
The invention also provides a device for identifying the slight green of the flue-cured tobacco, wherein the device for identifying the slight green of the flue-cured tobacco comprises:
The identification unit is used for acquiring an original spectrum image of the tobacco leaves to be identified;
the preprocessing unit is used for preprocessing the original spectrum image and obtaining processed spectrum information;
the calculating unit is used for obtaining the spectrum curve of each pixel according to the spectrum information and calculating the curvature radius of the spectrum curve of each pixel;
and the output unit extracts corresponding pixels according to the curvature radius and displays the pixels as a green smoke area.
Compared with the prior art, the method and the device for identifying the slight green of the flue-cured tobacco leaf have the following characteristics and advantages:
The method and the device for identifying the slight green of the flue-cured tobacco leaf can be used for rapidly identifying the slight green part in the tobacco leaf and can be used for rapidly identifying the green distribution in the tobacco leaf on line, thereby avoiding the loss caused by artificial factors, greatly solving the personnel consumption, increasing the selective use width of industry and reducing the difficulty of industrial formulation.
Drawings
The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way. In addition, the shapes, proportional sizes, and the like of the respective components in the drawings are merely illustrative for aiding in understanding the present invention, and are not particularly limited. Those skilled in the art with access to the teachings of the present invention can select a variety of possible shapes and scale sizes to practice the present invention as the case may be.
Fig. 1 is a flow chart of a method for identifying the slight green of the flue-cured tobacco leaves.
FIG. 2 is a raw spectral image of an embodiment of the present invention;
FIG. 3 is a schematic diagram of an embodiment of a micro-patch cyan identification image;
FIG. 4 is a raw spectral image of another embodiment of the present invention;
FIG. 5 is a schematic diagram of a light green identification result image according to another embodiment of the present invention;
Detailed Description
The details of the invention will be more clearly understood in conjunction with the accompanying drawings and description of specific embodiments of the invention. The specific embodiments of the invention described herein are for purposes of illustration only and are not to be construed as limiting the invention in any way. Given the teachings of the present invention, one of ordinary skill in the related art will contemplate any possible modification based on the present invention, and such should be considered to be within the scope of the present invention.
It will be understood that when an element is referred to as being "mounted" or "disposed" on another element, it can be directly on the other element or intervening elements may be present. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or intervening elements may be present.
The invention provides a method for identifying the slight green of flue-cured tobacco leaves, as shown in figure 1, which comprises the steps of collecting original spectrum images of the tobacco leaves to be identified (as shown in figures 2 and 4), preprocessing the original spectrum images to obtain preprocessed spectrum information, obtaining spectrum curves of pixels on the leaf surfaces of the tobacco leaves to be identified through the spectrum information, calculating the curvature radius of the spectrum curves of the pixels, extracting corresponding pixels according to the curvature radius and displaying the pixels as green smoke areas (as shown in figures 3 and 5).
The invention also provides a device for identifying the slight green of the flue-cured tobacco leaves, which comprises:
The identification unit is used for acquiring an original spectrum image of the tobacco leaves to be identified;
The preprocessing unit is used for preprocessing the original spectrum image and obtaining processed spectrum information;
the calculating unit is used for obtaining the spectrum curve of each pixel according to the spectrum information and calculating the curvature radius of the spectrum curve of each pixel;
And the output unit extracts corresponding pixels according to the curvature radius and displays the pixels as a green smoke area.
The method and the device for identifying the slight green of the flue-cured tobacco leaf can be used for rapidly identifying the slight green part in the tobacco leaf and can be used for rapidly identifying the green distribution in the tobacco leaf on line, thereby avoiding the loss caused by artificial factors, greatly solving the personnel consumption, increasing the selective use width of industry and reducing the difficulty of industrial formulation.
According to the method and the device for identifying the slight green of the flue-cured tobacco leaf, provided by the invention, the on-line rapid identification of the slight green part in the tobacco leaf is realized by calculating the curvature radius of part of each pixel spectral curve in the tobacco leaf spectral image and extracting the distribution of the green tobacco region according to the curvature radius in each pixel.
Meanwhile, the curvature radius is a main method for measuring the bending degree of the curve, the curvature radius can be better attached to the curvature radius of the characteristic peak, the curvature radius is measured, priori learning knowledge is not needed, the trouble of manually marking data is avoided, and the quick identification can be automatically carried out.
In an alternative example of the present invention, the identification unit is a hyperspectral imager, and the raw spectral image is acquired by the hyperspectral imager.
The hyperspectral imager can adopt the prior art, can obtain tens to hundreds of wave bands of images in a narrower wave band range by measuring the energy of the radiation emitted or reflected by a sample, can realize the classification and detection of objects, organically combines the traditional two-dimensional imaging technology and the spectrum technology, has spectrum analysis and image processing capability, has higher spectrum resolution and can obtain rich information. The hyperspectral imager is used for acquiring detailed spectral characteristic information and external characteristic information of a sample, so that the measured substance can be evaluated with higher accuracy and stronger comprehensiveness.
In an alternative example of the invention, the model number of the hyperspectral imaging spectrometer is GAIAFIELD-V10E-AZ4, and further the identification unit also comprises a CCD camera, 4 halogen lamps with the power supply of 50W and an electric displacement control platform.
When the original image is acquired, firstly, a sample (tobacco leaf) to be detected is placed on an electric displacement control platform, then, the movement of the electric displacement control platform and the acquisition of hyperspectral images are controlled through a computing unit, and finally, the acquired hyperspectral image data are transmitted to the computing unit (computer) for storage.
In an alternative example, when the hyperspectral imager is used for collecting original spectrum images of tobacco leaves, the tobacco leaves are fully unfolded and placed on an objective table of the hyperspectral imager manually, the map information of the front and the back of the tobacco leaves is collected respectively, and the map information is stored in a preprocessing unit or a computing unit according to digital numbers.
In an alternative embodiment of the invention, the preprocessing process includes data correction of the raw map information using standard black and white plate correction formulas.
The standard black-and-white plate correction formula is:
In the formula, a standard white correction plate obtains a full-white calibration image W, a full-black calibration image B is obtained by scanning a built-in blackboard, an original image I is subjected to correction processing, and the preprocessed map data is R.
Specifically, before the sample to be measured is collected, a standard white correction plate and a built-in blackboard are required to be collected for correcting the collected hyperspectral image, so as to eliminate the influence of instrument noise. W is a standard white correction plate to obtain a full white calibration image, the white plate is moved below a camera to collect and store a hyperspectral image, B is a full black calibration image obtained by scanning a built-in blackboard, a camera lens is covered by a camera cover to collect and store the full black hyperspectral image, I is a sample to be detected and is obtained by collecting the sample on an electric displacement control platform, and R is a spectrum image which is obtained by correcting the black and white plate and is used for eliminating instrument noise and smoothing.
In an optional embodiment of the invention, obtaining the spectrum value of each pixel of the leaf surface of the tobacco leaf to be identified through the pretreated spectrum information comprises the steps of converting the pretreated spectrum information into single-band spectrum data, dividing the single-band spectrum data into tobacco leaf information and background information, dividing the tobacco leaf information into leaf surface information, leaf stalk information and fold information, marking the leaf surface information as an induction area, correcting the spectrum of an interested area, intercepting the leaf surface information in a preset wave band range, and calculating the curvature radius of a spectrum curve of each pixel in the leaf surface information one by one.
In an alternative example of this embodiment, spectrum information with a band of 675.4nm in the pre-processed spectrum information is selected to be converted into binary single-band spectrum data.
In an alternative example of this embodiment, the global thresholding is used to segment the tobacco leaf and background information, and then the leaf surface, leaf stalk and fold information is segmented by the local thresholding, marking the leaf surface portion as the region of interest.
The threshold segmentation method of the image has the characteristics of intuitiveness and easiness in implementation, and plays an important role in image segmentation application. Specifically, the image f (m, n) is composed of two types of regions having different gray levels, i.e., dark objects and bright objects. The bright and dark portions of such an image can be clearly distinguished in the histogram, so a threshold can be selected for distinguishing between bright and dark peaks. Threshold segmentation can be seen as a functional operation:
T=T[m,n,p(m,n),f(m,n)];
Wherein m and n represent the abscissa p (m, n) of the pixel, f (m, n) represents the gray value of the pixel, and the thresholded image is defined as:
wherein the pixel with gray scale labeled 1 corresponds to the object of interest, the pixel labeled 0 corresponds to the background, and there are:
1. if T depends only on f (m, n), the threshold is global;
2. If T depends on f (m, n) and p (m, n), the threshold is local;
3. if T depends on the coordinates (m, n), the threshold is adaptive.
In the above process, the global threshold may be understood as counting the histogram of the whole image, then dividing all pixels in the image by selecting a certain threshold, the local threshold is dividing the image into set small blocks, counting the histogram in the small blocks, selecting a certain threshold, dividing the pixels in the small blocks, and the local threshold dividing is finer.
In an alternative example of this embodiment, the radius of curvature calculation uses the formula:
In the formula, X is a spectral band value, and y is the reflectivity of the corresponding spectral band value.
In an alternative example, the spectrum of the region of interest is corrected by a multi-element scattering correction method, then the curvature radius of the spectrum curve of each pixel is calculated one by one in the 600nm-700nm band range.
The multi-element scattering correction method can effectively eliminate spectrum differences caused by different scattering levels through the baseline translation and offset phenomenon of spectrum data corrected by the average spectrum, thereby enhancing the correlation between the spectrum and the data.
The method comprises (1) obtaining average value of all spectrum data, (2) subjecting spectrum of each sample and average spectrum to unitary linear regression, solving least square problem to obtain baseline shift and offset of each sample, and (3) correcting spectrum of each sample by subtracting the obtained baseline shift and dividing by offset to obtain corrected spectrum
In an optional embodiment of the invention, the extraction process of the green smoke region according to each curvature radius comprises the steps of sequencing the pixels according to the curvature radius from low to high, then normalizing the curvature radius, and setting thermodynamic diagram to display green smoke distribution of leaf surfaces by adopting dividing points as indexes.
The normalization process is to convert the data into decimal of the (0, 1) interval and change the dimensionality expression into non-dimensionality expression to solve the comparability of the data.
In an alternative example of this embodiment, the radius of curvature of each of the sorted pixels is divided into four equal parts by a quantile method, and the equal parts respectively occupy 10%,20% and 30% of the radius of curvature according to the numerical value, which are used as quantiles for green smoke identification. Other percentages may also be used in the quantile method. The selection of the dividing points is different, the displayed green smoke distribution effect is different, the display areas with different green degrees can be understood, if the set dividing points are large, the green areas are displayed more completely, partial misidentification can exist, if the set dividing points are small, the green areas are displayed as areas with larger green degrees, in the embodiment, the selected dividing points are divided by reference according to the standard of the flue-cured tobacco GB 2635-92, and finally, the pixel areas with smaller values than the set dividing points are displayed.
In an alternative embodiment of the invention, the method of authentication further comprises preparing the tobacco leaves to be authenticated, and placing the tobacco leaf sample in an environment of 70% relative humidity and 25 ℃ for 48 hours to form the tobacco leaves to be authenticated.
The detailed explanation of the embodiments described above is only for the purpose of explaining the present invention so as to enable a better understanding of the present invention, but these descriptions should not be construed as limiting the present invention in any way, and in particular, the individual features described in the different embodiments may be arbitrarily combined with each other to constitute other embodiments, and these features should be understood as being applicable to any one embodiment, except as explicitly stated to the contrary, without being limited to only the described embodiment.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211023788.9A CN115356297B (en) | 2022-08-26 | 2022-08-26 | Method and device for identifying slightly green flue-cured tobacco leaves |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211023788.9A CN115356297B (en) | 2022-08-26 | 2022-08-26 | Method and device for identifying slightly green flue-cured tobacco leaves |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115356297A CN115356297A (en) | 2022-11-18 |
| CN115356297B true CN115356297B (en) | 2024-12-24 |
Family
ID=84005590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211023788.9A Active CN115356297B (en) | 2022-08-26 | 2022-08-26 | Method and device for identifying slightly green flue-cured tobacco leaves |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115356297B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115479907B (en) * | 2022-08-26 | 2024-12-27 | 中国烟草总公司郑州烟草研究院 | Digital tobacco leaf and manufacturing method thereof |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4262248B2 (en) * | 2003-01-21 | 2009-05-13 | 日本たばこ産業株式会社 | Application perfume inspection device and cigarette making machine equipped with the same |
| CN103940750B (en) * | 2014-04-21 | 2016-06-01 | 上海交通大学 | Saline soil curve of spectrum visible light wave range remote-sensing inversion construction process |
| CN110542658A (en) * | 2019-09-10 | 2019-12-06 | 中国烟草总公司郑州烟草研究院 | tobacco non-smoke substance classification method based on hyperspectral imaging technology |
| CN113010848A (en) * | 2021-02-23 | 2021-06-22 | 中国农业科学院烟草研究所(中国烟草总公司青州烟草研究所) | Flue-cured tobacco appearance quality evaluation data processing method, system, terminal and computer equipment |
| CN113390799B (en) * | 2021-06-24 | 2023-06-27 | 中国烟草总公司郑州烟草研究院 | Method for identifying and detecting stems in tobacco leaves |
| CN113679082A (en) * | 2021-06-24 | 2021-11-23 | 中国烟草总公司郑州烟草研究院 | Intelligent cigar cutting method and device |
-
2022
- 2022-08-26 CN CN202211023788.9A patent/CN115356297B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 基于高光谱成像的烟叶泛青特征分析与表征;郭文孟等;《烟草科技》;20230508;第56卷(第7期);第84-91页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115356297A (en) | 2022-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112699756B (en) | Hyperspectral image-based tea origin identification method and system | |
| CN113390799B (en) | Method for identifying and detecting stems in tobacco leaves | |
| CN119470828B (en) | Traditional chinese medicine medicinal material true and false identification platform | |
| CN103955938A (en) | Wheat growing status diagnosing method based on mobile internet mode and leaf color analysis | |
| CN110763698A (en) | A hyperspectral citrus leaf disease identification method based on characteristic wavelengths | |
| CN111443043A (en) | A method for quality detection of walnut kernels based on hyperspectral images | |
| CN108734205A (en) | A kind of simple grain for different cultivars wheat seed pinpoints identification technology | |
| CN105466921A (en) | Simultaneous detection method of many samples | |
| CN116559111A (en) | A method for identifying sorghum varieties based on hyperspectral imaging technology | |
| CN113570538B (en) | Blade RGB image bias distribution parameter information acquisition and analysis method | |
| CN120672874B (en) | Food freshness detection method and system based on image recognition | |
| CN115356297B (en) | Method and device for identifying slightly green flue-cured tobacco leaves | |
| CN118067639A (en) | A hyperspectral image acquisition and analysis method for evaluating the aging degree of textiles | |
| CN113610965B (en) | Plant chlorophyll content three-dimensional space stereoscopic distribution visualization method based on point cloud | |
| CN110319933B (en) | Light source spectrum optimization method based on CAM02-UCS color appearance model | |
| CN104463919A (en) | Full-pixel color analysis method for evaluating skin pigments | |
| CN115356275B (en) | Tobacco grade evaluation method | |
| EP3940370B1 (en) | Method for extracting spectral information of object to be detected | |
| CN115479907B (en) | Digital tobacco leaf and manufacturing method thereof | |
| Ciortan et al. | Color and hyperspectral image segmentation for historical documents | |
| CN119131472A (en) | A method for identifying adulteration of moringa seed oil | |
| CN116642827A (en) | Large-scale fabric staining rating method and system based on hyperspectral imaging technology | |
| CN109543757A (en) | A kind of painting and calligraphy painting style identification method based on spectral imaging technology and atlas analysis | |
| CN115147659A (en) | A kind of image analysis method and system for DUS test characters of plant fruit outer surface | |
| CN110807817B (en) | Machine vision method for target color recognition adapting to illumination change |
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 |