Feeding uniformity detection method and system based on hyperspectrum
Through hyperspectral imaging technology, the hyperspectral data set and intelligent interpretation model before and after feeding is established, real-time and accurate monitoring of feeding uniformity detection is achieved, the problems of detection complexity and hysteresis in the existing technology are solved, and the feeding process is improved. level of control.
Patent Information
- Application Number
- CN202411261693.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-03
AI Technical Summary
The existing feed uniformity detection methods have problems such as complex operation, long detection cycle, and lag in effect feedback, making it difficult to achieve real-time, accurate and quantitative monitoring of feeding effects.
The hyperspectral imaging method is used to establish the hyperspectral data set of smoke sheets or leaf filaments before and after feeding, and the feeding uniformity evaluation model is constructed based on the intelligent interpretation method of combining map information. The spectral and imaging information after feeding are captured by the hyperspectral industrial camera. Enter the model to obtain the feed uniformity evaluation results.
Real-time, accurate and quantitative monitoring of feeding effects is achieved, the control level of feeding process is improved, and the complexity and hysteresis problems of detection in the existing technology is solved.
Smart Images

Figure CN119064292A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of tobacco leaf quality detection, and more specifically, to a method and system for detecting feed uniformity based on a hyperspectral spectrum. Background Art
[0002] Adding material is the process of spraying liquid on the surface of tobacco leaves / leaf shreds, and is one of the key steps in the shred process. The uniformity of liquid application and the effect of material absorption of liquid play an important role in the quality stability of cigarette products. At present, the industry's evaluation method for adding material uniformity mainly adopts YC / T 353-2010 "Determination of Cigarette Adding Material Uniformity". This method is an offline chemical analysis method with high accuracy of test results, but it has disadvantages such as complex operation, long test cycle, and delayed feedback of results.
[0003] Therefore, there is an urgent need for a method and system for detecting uniformity of feed materials based on hyperspectral. Summary of the invention
[0004] The purpose of the present invention is to provide a method and system for detecting uniformity of feeding based on hyperspectral, so as to solve the above-mentioned problems in the prior art and realize real-time, accurate and quantitative monitoring of feeding effect.
[0005] The present invention provides a method for detecting uniformity of feeding material based on hyperspectral, which comprises:
[0006] A hyperspectral imaging method is used to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials;
[0007] Based on the hyperspectral data set, construct a feed uniformity assessment model;
[0008] Based on the hyperspectral data of the tobacco slices or shreds before and after adding the material and the feeding uniformity evaluation model, the feeding uniformity of the tobacco slices or shreds after adding the material is evaluated.
[0009] As described above, the method for detecting uniformity of feeding based on hyperspectral, wherein preferably, the hyperspectral imaging method is used to establish a hyperspectral dataset of tobacco leaves or shreds before and after feeding, specifically comprising:
[0010] Using a hyperspectral industrial camera, a hyperspectral dataset of tobacco leaves or shreds before and after adding materials is established.
[0011] The method for detecting uniformity of feeding based on hyperspectral as described above, wherein preferably, the hyperspectral data set of tobacco slices or shredded leaves before and after feeding is established by means of a hyperspectral industrial camera, specifically comprising:
[0012] Select the designated brand of tobacco flakes or shreds before adding the material, the liquid material, and the material after adding the material;
[0013] A hyperspectral industrial camera is used to capture the spectrum and imaging information of tobacco slices or leaf shreds before and after adding materials, so as to establish a hyperspectral dataset of tobacco slices or leaf shreds before and after adding materials.
[0014] The method for detecting uniformity of feeding based on hyperspectral as described above, wherein preferably, the method of using a hyperspectral industrial camera to capture the spectrum and imaging information of the tobacco slices or leaf shreds before and after feeding, so as to establish a hyperspectral data set of the tobacco slices or leaf shreds before and after feeding, specifically includes:
[0015] Use hyperspectral industrial cameras to collect and analyze multi-spectral data, determine applicable spectral bands and signal markers of corresponding spectral bands, and establish a data set corresponding to the specified brand;
[0016] Taking the samples before entering the feeding drum and after leaving the drum as comparison, sufficient hyperspectral data was collected based on the difference in spectral images before and after feeding, and an analytical data set that can determine the feeding status was established.
[0017] As described above, the method for detecting uniformity of feeding material based on hyperspectral data set, wherein preferably, the method for constructing a feeding uniformity evaluation model based on the hyperspectral data set specifically includes:
[0018] Based on the hyperspectral data set, an intelligent interpretation method combining spectral information is adopted to construct a feeding uniformity assessment model.
[0019] As described above, the method for detecting uniformity of feeding material based on hyperspectral data set, wherein preferably, the method of constructing a feeding uniformity evaluation model by adopting an intelligent interpretation method combined with spectral information based on the hyperspectral data set specifically includes:
[0020] Using an intelligent interpretation method combined with spectral information to select key spectral features in the hyperspectral dataset;
[0021] Performing spectral end member extraction on tobacco slices or leaf shreds and liquid according to the key spectral features;
[0022] A spectral mixing model is established based on the extracted spectral endmembers;
[0023] Solving the spectral mixing model to obtain a mixing ratio coefficient of tobacco flakes or shredded tobacco leaves and liquid;
[0024] The liquid content is obtained based on the mixing ratio coefficient of the tobacco slices or leaf shreds and the liquid, thereby constructing a feeding uniformity evaluation model.
[0025] As described above, the method for detecting uniformity of feeding based on hyperspectral, wherein preferably, the feeding uniformity of the tobacco slices or tobacco shreds after feeding is evaluated based on the hyperspectral data of the tobacco slices or tobacco shreds before and after feeding and the feeding uniformity evaluation model, specifically comprising:
[0026] Use a hyperspectral industrial camera to capture the spectrum and imaging information of the tobacco leaves or shreds to be tested after adding materials;
[0027] The spectrum and imaging information of the tobacco slices or shreds to be tested after adding materials are input into the feeding uniformity evaluation model to obtain the feeding uniformity evaluation result.
[0028] The present invention also provides a feeding uniformity detection system based on hyperspectral, comprising:
[0029] A hyperspectral data set acquisition module, used to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials by using a hyperspectral imaging method;
[0030] A feeding uniformity assessment model building module is used to build a feeding uniformity assessment model based on the hyperspectral data set;
[0031] The charging uniformity evaluation module is used to evaluate the charging uniformity of the tobacco slices or tobacco shreds after charging according to the hyperspectral data of the tobacco slices or tobacco shreds before and after charging and the charging uniformity evaluation model.
[0032] The present invention provides a method for detecting uniformity of feeding based on hyperspectral technology, adopts hyperspectral technology to study the intrinsic correlation between uniformity of feeding and spectral information, establishes a characterization method of uniformity of feeding using hyperspectral technology, realizes real-time, accurate and quantitative monitoring of feeding effect, and facilitates improving the control level of feeding process link. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:
[0034] Figure 1 A flow chart of an embodiment of a method for detecting uniformity of feeding material based on hyperspectral provided by the present invention;
[0035] Figure 2 This is a structural block diagram of an embodiment of a feeding uniformity detection system based on hyperspectral provided by the present invention. DETAILED DESCRIPTION
[0036] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and is in no way intended to limit the present disclosure and its application or use. The present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to make the present disclosure thorough and complete and to fully express the scope of the present disclosure to those skilled in the art. It should be noted that unless otherwise specifically stated, the relative arrangement of the components and steps, the composition of the materials, the numerical expressions and the numerical values set forth in these embodiments should be interpreted as being merely exemplary and not as limiting.
[0037] The words "first", "second" and similar words used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. The words "include" or "comprises" and similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of including other elements. "Up", "down" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0038] In the present disclosure, when a specific component is described as being located between a first component and a second component, there may or may not be an intermediate component between the specific component and the first component or the second component. When a specific component is described as being connected to other components, the specific component may be directly connected to the other components without an intermediate component, or may not be directly connected to the other components but have an intermediate component.
[0039] All terms (including technical terms or scientific terms) used in the present disclosure have the same meanings as those understood by ordinary technicians in the field to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries, such as general dictionaries, should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined herein.
[0040] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.
[0041] At present, the industry's main method for evaluating the uniformity of cigarette fillings is YC / T 353-2010 "Determination of the Uniformity of Cigarette Fillings". This method is an offline chemical analysis method with high accuracy of test results, but it has disadvantages such as complex operation, long test cycle, and delayed feedback on results.
[0042] At present, the research methods on feeding uniformity at home and abroad also include production process precision control method, material morphology modeling and analysis method, characteristic substance detection method, near-infrared spectroscopy method, imaging method, etc.
[0043] Among them, the production process precision control method is to add various material information tracking and comparison measures to the workshop control system, and enable response mechanisms such as prohibition of start-up, automatic shutdown, alarm prompts or automatic adjustment when errors are found. This type of method is used to prevent mixed material odor and excessive application ratio problems. It has relatively mature and extensive application experience. Its main disadvantage is that it cannot cover the blind areas of stirring and moisture removal after the blades enter the feeding barrel, and it cannot track and evaluate the effective liquid content actually absorbed by the blades in real time.
[0044] The material morphology modeling analysis method models the movement and spatial distribution of finely granulated tobacco and liquid particles during the drum mixing process, thereby directly calculating the mixing ratio of tobacco leaves and liquid at different positions. This method also cannot track and evaluate the effective liquid content actually absorbed by the leaves in real time.
[0045] The characteristic substance detection method is to analyze the liquid content of the leaves leaving the drum. This type of analysis usually uses specific alcohol components in the liquid as detection markers. Although this type of method has a relatively solid theoretical basis and the test results given are of high reference value, it is difficult to achieve repeatability of the measurement data because the alcohol characteristic substances in the liquid evaporate very quickly and the data detected at different times vary greatly. In addition, this method has high requirements for instruments and experimental environment, and the analysis process may cause irreversible damage to the tobacco material, making it difficult to combine with the application environment of online detection.
[0046] Near-infrared spectroscopy and imaging methods have been used by some researchers to detect liquid in tobacco leaves. However, most of the related research methods use single-point sampling spectroscopy or imaging methods without spectral information. Single-point infrared spectroscopy has a limited number of sampling points and is sensitive to the influence of sample heterogeneity and impurities, which affects the accuracy of measurement. It is also difficult to obtain accurate liquid distribution. The overall sampling speed is slow, resulting in low detection efficiency. Due to the lack of spectral dimension information, the imaging method is difficult to detect the true concentration of the liquid, and it is also impossible to identify the liquid that has been absorbed by the tobacco leaves.
[0047] like Figure 1 As shown, the method for detecting uniformity of feeding material based on hyperspectral provided in this embodiment specifically includes the following steps during actual implementation:
[0048] Step S1, using a hyperspectral imaging method to establish a hyperspectral dataset of tobacco leaves or shreds before and after adding materials.
[0049] Specifically, in the present invention, a hyperspectral data set of tobacco leaves or shreds before and after adding materials is established by using a hyperspectral industrial camera.
[0050] In one embodiment of the method for detecting uniformity of feeding material based on hyperspectral of the present invention, the step S1 may specifically include:
[0051] Step S11, selecting tobacco leaves or shreds before adding materials, liquid materials, and materials after adding materials of a specified brand.
[0052] In some embodiments of the present invention, the spectral information dataset is no less than 4 brands.
[0053] Step S12: Using a hyperspectral industrial camera to capture the spectrum and imaging information of the tobacco slices or leaf shreds before and after adding the material, so as to establish a hyperspectral data set of the tobacco slices or leaf shreds before and after adding the material.
[0054] In one embodiment of the method for detecting uniformity of feeding material based on hyperspectral of the present invention, the step S12 may specifically include:
[0055] Step S121: Use a hyperspectral industrial camera to collect and analyze multi-spectral data, determine applicable spectral bands and signal markers of corresponding spectral bands, and establish a data set corresponding to the specified brand.
[0056] Specifically, the tobacco slices / shreds after feeding are sampled in real time through an online acquisition device, and photographed using a hyperspectral industrial camera to achieve high-frequency sampling of production line samples. In some embodiments of the present invention, after the tobacco slices are fed and removed from the barrel, they enter the hyperspectral photography device through an online sampling device to complete hyperspectral data collection and synchronously perform data analysis to obtain the feeding uniformity analysis results; after completing the hyperspectral image shooting, the samples will be recovered to the production line through the online acquisition device, and flexibly embedded in the actual production process.
[0057] Step S122: Taking the samples before entering the feeding drum and after leaving the drum as comparison, sufficient hyperspectral data is collected based on the difference in spectral images before and after feeding, and an analysis data set capable of determining the feeding status is established.
[0058] In some embodiments of the present invention, tobacco slices or leaf shreds before adding feed are manually sprayed with feed solutions of different concentrations to prepare content gradient samples of feed, and hyperspectral photography is performed using a hyperspectral industrial camera.
[0059] Specifically, a light source of appropriate wavelength (such as ultraviolet light) is selected to excite the spectral response of the test object, and the material before and after the addition is imaged by a hyperspectral camera. The material and the spice liquid mixture are accurately located based on the planar image information, and the pixel spectrum at the corresponding position is analyzed. Based on the sample spectrum before and after the addition, the mixed spectrum is decomposed. First, the interference of pseudo information in the data is eliminated, and the spectral response data sets of pure liquid, pure material, and the mixture after the addition are established according to the characteristic waveform of the leaf group formula in the spectrum. At the same time, the accuracy and consistency of the hyperspectral imaging data are analyzed by metrology, so as to evaluate the reliability of the spectral data set as a spectral dictionary for the uniformity detection of the addition.
[0060] The feeding uniformity detection algorithm adopted by the present invention (which will be described in detail in the subsequent step S2) is based on the difference in hyperspectral images before and after the feeding. Therefore, in step S1, it is necessary to collect data in different bands on a large number of tobacco slices / leaf shreds before and after the feeding and the feed liquid, and to conduct a systematic study on the correlation between the spectral characteristics and the feeding uniformity index through statistical methods, so as to establish a standard hyperspectral data set that can effectively characterize the differences in samples before and after the feeding.
[0061] Step S2: constructing a feed uniformity assessment model based on the hyperspectral data set.
[0062] In the present invention, based on the hyperspectral data set, an intelligent interpretation method combining spectral information is adopted to construct a feed uniformity assessment model.
[0063] In one embodiment of the method for detecting uniformity of feeding material based on hyperspectral of the present invention, the step S2 may specifically include:
[0064] Step S21: adopting a spatial-spectral unified intelligent interpretation method to select key spectral features from the hyperspectral data set.
[0065] Step S22: performing spectral end member extraction (EndMember Extraction) on the tobacco leaves or shredded leaves and the liquid according to the key spectral features.
[0066] Step S23: establishing a spectral unmixing model according to the extracted spectral endmembers.
[0067] Step S24, solving the spectral mixing model to obtain the mixing ratio coefficient of the tobacco leaves or shreds and the liquid.
[0068] Step S25, obtaining the liquid content based on the mixing ratio coefficient of the tobacco leaves or shreds and the liquid, thereby constructing a feeding uniformity evaluation model.
[0069] In some embodiments of the present invention, the charging uniformity evaluation model corresponds to a brand, and there are no less than 4 brands.
[0070] Step S3, evaluating the feeding uniformity of the tobacco sheets or tobacco shreds after feeding according to the hyperspectral data of the tobacco sheets or tobacco shreds before and after feeding and the feeding uniformity evaluation model.
[0071] In one embodiment of the method for detecting uniformity of feeding material based on hyperspectral of the present invention, the step S3 may specifically include:
[0072] Step S31, using a hyperspectral industrial camera to capture the spectrum and imaging information of the tobacco leaves or shreds to be tested after adding materials.
[0073] Step S32, inputting the spectrum and imaging information of the tobacco slices or shreds to be tested after adding the feed into the feed uniformity evaluation model to obtain the feed uniformity evaluation result.
[0074] Hyperspectral technology fully integrates spectral and imaging technologies, and can be combined with data models to perform direct imaging measurements on the concentration distribution of materials. It is highly feasible in solving the problems of tobacco feeding uniformity and utilization rate detection.
[0075] The hyperspectral-based feeding uniformity detection method provided in the embodiment of the present invention adopts hyperspectral technology to study the intrinsic correlation between feeding uniformity and spectral information, establishes a characterization method for feeding uniformity using hyperspectral technology, realizes real-time, accurate and quantitative monitoring of feeding effects, and facilitates improving the management and control level of the feeding process.
[0076] Correspondingly, the present invention also provides a feeding uniformity detection system based on hyperspectral, comprising:
[0077] A hyperspectral data set acquisition module 101 is used to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials by using a hyperspectral imaging method;
[0078] A feed uniformity assessment model building module 102 is used to build a feed uniformity assessment model based on the hyperspectral data set;
[0079] The charging uniformity evaluation module 103 is used to evaluate the charging uniformity of the tobacco slices or tobacco shreds after charging according to the hyperspectral data of the tobacco slices or tobacco shreds before and after charging and the charging uniformity evaluation model.
[0080] The hyperspectral-based feeding uniformity detection system provided in the embodiment of the present invention adopts hyperspectral technology to study the intrinsic relationship between feeding uniformity and spectral information, establishes a characterization method of feeding uniformity using hyperspectral technology, realizes real-time, accurate and quantitative monitoring of feeding effects, and facilitates improving the management and control level of the feeding process.
[0081] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.
[0082] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.
Claims
1. A method for detecting uniformity of feeding based on hyperspectral, It is characterized in that include: A hyperspectral imaging method is used to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials; Based on the hyperspectral data set, construct a feed uniformity assessment model; Based on the hyperspectral data of the tobacco slices or shreds before and after adding the material and the feeding uniformity evaluation model, the feeding uniformity of the tobacco slices or shreds after adding the material is evaluated.
2. The method for detecting uniformity of feeding material based on hyperspectral according to claim 1, It is characterized in that The method of using a hyperspectral imaging method to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials specifically includes: Using a hyperspectral industrial camera, a hyperspectral dataset of tobacco leaves or shreds before and after adding materials is established.
3. The method for detecting uniformity of feeding material based on hyperspectral according to claim 2, It is characterized in that The method of establishing a hyperspectral data set of tobacco leaves or shreds before and after adding materials by using a hyperspectral industrial camera specifically includes: Select the designated brand of tobacco flakes or shreds before adding the material, the liquid material, and the material after adding the material; A hyperspectral industrial camera is used to capture the spectrum and imaging information of tobacco slices or leaf shreds before and after adding materials, so as to establish a hyperspectral dataset of tobacco slices or leaf shreds before and after adding materials.
4. The method for detecting uniformity of feeding material based on hyperspectral according to claim 3, It is characterized in that The method of using a hyperspectral industrial camera to capture the spectrum and imaging information of tobacco slices or leaf shreds before and after adding materials to establish a hyperspectral data set of tobacco slices or leaf shreds before and after adding materials specifically includes: Use hyperspectral industrial cameras to collect and analyze multi-spectral data, determine applicable spectral bands and signal markers of corresponding spectral bands, and establish a data set corresponding to the specified brand; Taking the samples before entering the feeding drum and after leaving the drum as comparison, sufficient hyperspectral data was collected based on the difference in spectral images before and after feeding, and an analytical data set that can determine the feeding status was established.
5. The method for detecting uniformity of feeding material based on hyperspectral according to claim 1, It is characterized in that The step of constructing a feed uniformity assessment model based on the hyperspectral data set specifically includes: Based on the hyperspectral data set, an intelligent interpretation method combining spectral information is adopted to construct a feeding uniformity assessment model.
6. The method for detecting uniformity of feeding material based on hyperspectral according to claim 5, It is characterized in that The method of constructing a feeding uniformity evaluation model based on the hyperspectral data set and combining spectral information with an intelligent interpretation method specifically includes: Using an intelligent interpretation method combined with spectral information to select key spectral features in the hyperspectral dataset; Performing spectral end member extraction on tobacco slices or leaf shreds and liquid according to the key spectral features; A spectral mixing model is established based on the extracted spectral endmembers; Solving the spectral mixing model to obtain a mixing ratio coefficient of tobacco flakes or shredded tobacco leaves and liquid; The liquid content is obtained based on the mixing ratio coefficient of the tobacco slices or leaf shreds and the liquid, thereby constructing a feeding uniformity evaluation model.
7. The method for detecting uniformity of feed material based on hyperspectral according to claim 1, It is characterized in that The step of evaluating the uniformity of the tobacco slices or shreds after adding the feed based on the hyperspectral data of the tobacco slices or shreds before and after adding the feed and the feed uniformity evaluation model specifically includes: Use a hyperspectral industrial camera to capture the spectrum and imaging information of the tobacco leaves or shreds to be tested after adding materials; The spectrum and imaging information of the tobacco slices or shreds to be tested after adding materials are input into the feeding uniformity evaluation model to obtain the feeding uniformity evaluation result.
8. A feeding uniformity detection system based on hyperspectral, It is characterized in that include: A hyperspectral data set acquisition module, used to establish a hyperspectral data set of tobacco leaves or shreds before and after adding materials by using a hyperspectral imaging method; A feeding uniformity assessment model building module is used to build a feeding uniformity assessment model based on the hyperspectral data set; The charging uniformity evaluation module is used to evaluate the charging uniformity of the tobacco slices or tobacco shreds after charging according to the hyperspectral data of the tobacco slices or tobacco shreds before and after charging and the charging uniformity evaluation model.
