Tobacco leaf spectrum data processing method and device, electronic equipment, medium and product

Abstract

The embodiments of the present disclosure disclose a tobacco spectrum data processing method and device, electronic equipment, medium and product. The method comprises: obtaining first tobacco spectrum data of a target tobacco and second tobacco spectrum data of at least one optional tobacco; for at least one optional tobacco, determining a fractal similarity corresponding to the optional tobacco according to the second tobacco spectrum data of the optional tobacco and the first tobacco spectrum data, the fractal similarity being used to indicate the similarity of the deep spectrum structure between the first tobacco spectrum data and the second tobacco spectrum data; and determining a replaceable tobacco for replacing the target tobacco from the at least one optional tobacco according to at least one fractal similarity. The technical scheme is based on the fractal similarity calculated from the spectrum data to represent the difference of the deep spectrum structure, gets rid of the subjective influence of artificial product evaluation, objectively and accurately selects the suitable replacement tobacco, improves the reliability of the selection of the tobacco raw material replacement, and improves the efficiency of the tobacco replacement selection.

Description

Technical Field

[0001] This disclosure relates to the field of tobacco raw material quality evaluation technology, and in particular to a method, apparatus, electronic device, medium and product for processing tobacco spectral data. Background Technology

[0002] With the development of the tobacco industry towards large-scale and refined production, the contradiction between supply and demand of tobacco raw materials has become increasingly prominent. The rational allocation of stockpiled tobacco leaves, the precise substitution of high-quality tobacco leaves, and the digital upgrading of formulas have become common problems that the industry urgently needs to solve.

[0003] In related technologies, traditional tobacco substitution still relies primarily on manual judgment, based on tobacco chemical indicators combined with human experience and sensory evaluation. However, this method requires repeated manual evaluation, comparison, and adjustment, which leads to problems such as high workload, strong subjectivity, low efficiency, and difficulty in finding the optimal substitute tobacco. Summary of the Invention

[0004] This disclosure provides a method, apparatus, electronic device, medium, and product for processing tobacco leaf spectral data. It uses fractal similarity calculation based on spectral data to characterize differences in deep spectral structures, eliminating the subjective influence of manual evaluation, and objectively and accurately screening suitable alternative tobacco leaves. This improves the reliability of tobacco raw material replacement selection and increases the efficiency of tobacco leaf replacement screening.

[0005] According to one aspect of this disclosure, a method for processing tobacco leaf spectral data is provided, the method comprising: Acquire first tobacco leaf spectral data of the target tobacco leaf and second tobacco leaf spectral data of at least one optional tobacco leaf; For at least one of the optional tobacco leaves, a fractal similarity corresponding to the optional tobacco leaf is determined based on the second tobacco leaf spectral data and the first tobacco leaf spectral data of the optional tobacco leaf. The fractal similarity is used to indicate the degree of similarity of the deep spectral structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data. Based on at least one of the fractal similarities, alternative tobacco leaves for replacing the target tobacco leaf are determined from at least one of the optional tobacco leaves.

[0006] According to another aspect of this disclosure, a tobacco leaf spectral data processing apparatus is provided, the apparatus comprising: The spectral data acquisition module is used to acquire the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf. A similarity determination module is used to determine, for at least one of the optional tobacco leaves, a fractal similarity corresponding to the optional tobacco leaf based on the second tobacco leaf spectral data and the first tobacco leaf spectral data of the optional tobacco leaf, wherein the fractal similarity is used to indicate the degree of similarity of the spectral deep structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data; A tobacco leaf determination module is used to determine alternative tobacco leaves for replacing the target tobacco leaf from at least one of the optional tobacco leaves based on at least one of the fractal similarities.

[0007] According to another aspect of this disclosure, an electronic device is provided, the electronic device comprising: One or more processors; Storage device for storing one or more programs. When one or more programs are executed by one or more processors, the one or more processors implement a tobacco leaf spectral data processing method as described in any of the embodiments of this disclosure.

[0008] According to another aspect of this disclosure, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement any of the tobacco leaf spectral data processing methods in the embodiments of this disclosure.

[0009] According to another aspect of the present disclosure, a computer program product is provided, which, when executed by a processor, implements a tobacco leaf spectral data processing method as described in any of the embodiments of the present disclosure.

[0010] The technical solution of this disclosure provides objective and standardized basic data for subsequent precise matching of tobacco leaves by acquiring the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf. Furthermore, for at least one optional tobacco leaf, based on the second and first tobacco leaf spectral data, a fractal similarity is determined. This fractal similarity indicates the degree of similarity in the deep spectral structure between the first and second tobacco leaf spectral data, overcoming the limitation of only comparing apparent spectral data. By using fractal similarity to accurately quantify the intrinsic quality differences of tobacco leaves, subjective errors caused by manual tasting are effectively reduced, improving the accuracy of tobacco leaf similarity assessment. Furthermore, based on at least one fractal similarity, alternative tobacco leaves for replacing the target tobacco leaf are determined from at least one optional tobacco leaf. Relying on the quantified fractal similarity as an objective screening standard, suitable alternative tobacco leaves can be quickly and accurately screened, reducing the uncertainty of manual tasting and improving the efficiency of tobacco raw material selection and formula stability. The technical solution of this disclosure solves the problems of high workload, strong subjectivity, low efficiency and difficulty in finding the best alternative tobacco leaves caused by the manual judgment method in related technologies. It realizes tobacco leaf substitution screening based on digital comparison of deep spectral structural features, completely gets rid of the subjective limitations of traditional manual evaluation and judgment, effectively reduces the workload of manual labor, improves the efficiency of tobacco leaf substitution screening, can accurately capture the internal quality differences of tobacco leaves, improve the accuracy of tobacco leaf formula substitution screening, and provide reliable data support for intelligent and digital formulation of cigarette formulas.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating a method for processing tobacco leaf spectral data provided in this embodiment of the present disclosure; Figure 2 A flowchart illustrating another method for processing tobacco leaf spectral data provided in this embodiment of the present disclosure; Figure 3 A flowchart illustrating an optional embodiment of a tobacco leaf spectral data processing method provided in this disclosure; Figure 4This is a schematic diagram of the structure of a tobacco leaf spectral data processing device provided in an embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present disclosure.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0017] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0018] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0019] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0020] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0021] Before introducing this technical solution, we can first illustrate its application scenarios. This technical solution can be applied to scenarios where spectral data of tobacco leaves is used to screen for similar tobacco leaves to replace target tobacco leaves. For example, it is widely used in tobacco raw material quality evaluation, digital management of tobacco inventory, intelligent substitution of cigarette formulas, and optimization of leaf threshing and re-drying processes. In recent years, in the field of tobacco raw material quality evaluation technology, tobacco spectroscopy has been widely used in tobacco component detection, origin identification, and formula substitution selection due to its advantages of speed, non-destructiveness, and low cost. Its fingerprint spectrum can intuitively reflect the differences in tobacco component composition and comprehensively capture the differences in the types and contents of chemical components among tobacco leaves.

[0022] Currently, similarity analysis of tobacco leaves based on their spectra is an unsupervised task, lacking labels and making it difficult to quantitatively optimize preprocessing conditions. Many methods rely on blindly setting conditions based on supervised modeling experience. Furthermore, existing workflows often combine dimensionality reduction techniques such as principal component analysis, cluster analysis, and t-distribution random neighborhood embedding to compress the spectral dimension before performing similarity calculations in the lower-dimensional space. However, this approach has significant drawbacks: the lack of evaluation criteria for preprocessing selection easily leads to the loss of key structural features or the retention of noise; the dimensionality reduction effect is severely affected by preprocessing, making it difficult to preserve the inherent structure of the spectrum; and since tobacco leaf spectra are ordered curves with multi-scale fluctuations, traditional algorithms only compare absorbance values ​​at the same sites, discarding the spectral line arrangement order and multi-level morphological features, resulting in a one-sided and distorted similarity evaluation.

[0023] Based on this, in this embodiment, fractal similarity is introduced to characterize the deep structural similarity of the spectrum, thereby screening alternative tobacco leaves. This scheme characterizes the differences in spectral structure from the microscopic to the macroscopic level, requires no rigorous preprocessing, has a complete theory, strong noise resistance, improves the accuracy of tobacco leaf formulation substitution screening, and provides reliable data support for the intelligent and digital formulation of cigarette formulas.

[0024] Figure 1This is a flowchart illustrating a method for processing tobacco leaf spectral data according to an embodiment of this disclosure. This embodiment is applicable to scenarios where tobacco leaf spectral data is used to filter similar tobacco leaves for use as substitutes for target tobacco leaves. This method can be executed by a tobacco leaf spectral data processing device, which can be implemented in hardware and / or software and can be configured in electronic devices such as computers or servers. Figure 1 As shown, the method in this embodiment includes: S110: Obtain first tobacco leaf spectral data of the target tobacco leaf and second tobacco leaf spectral data of at least one optional tobacco leaf.

[0025] In one scenario, the target tobacco leaf can be understood as the baseline tobacco leaf raw material that needs to be replaced. The target tobacco leaf can be a tobacco leaf that is regularly used in a cigarette formula but requires a substitute due to supply imbalance. For example, a middle-grade high-quality tobacco leaf that is regularly used in a cigarette formula but needs to be replaced due to insufficient inventory; this tobacco leaf is the target tobacco leaf. The first tobacco leaf spectral data can be understood as fingerprint spectral data characterizing the intrinsic quality of the target tobacco leaf. The first tobacco leaf spectral data can be spectral information obtained through spectral detection methods that reflects the chemical composition and quality structure of the target tobacco leaf at the molecular level. The first tobacco leaf spectral data can include multiple data points, each corresponding to the spectral absorption response value (e.g., absorbance) of the target tobacco leaf at a corresponding wavenumber. The first tobacco leaf spectral data can include various spectral data that characterize the physicochemical quality of the target tobacco leaf, including spectral data obtained from detection at different wavebands. The first tobacco leaf spectral data can include first near-infrared spectral data. It can be understood that near-infrared spectroscopy, as a molecular fingerprint spectrum, can comprehensively capture subtle differences in the types and contents of chemical components among tobacco leaves. The first near-infrared spectral data can be the absorption spectrum of the target tobacco leaf collected in the near-infrared band.

[0026] In one scenario, optional tobacco leaves can be understood as alternative tobacco materials, which are available for screening and comparison and have substitutive feasibility. Optional tobacco leaves can be various types of tobacco raw materials with sufficient inventory, compliant categories, and can serve as candidates to replace the target tobacco leaf. Optionally, optional tobacco leaves can include at least one of the following: inventory tobacco leaves of the same grade but different production batches as the target tobacco leaf; idle tobacco leaves from the same production area and adjacent quality grade as the target tobacco leaf; and tobacco raw materials from other regions with similar physicochemical indicators and abundant inventory as the target tobacco leaf. For example, continuing with the previous example, if the target tobacco leaf is a consistently used medium-grade high-quality tobacco leaf in a certain cigarette formula, different batches of medium-grade tobacco leaves of the same grade or similar quality remaining in the warehouse all fall under the category of optional tobacco leaves. Secondary tobacco leaf spectral data can be understood as fingerprint spectral data characterizing the intrinsic quality of optional tobacco leaves. Secondary tobacco leaf spectral data can be spectral information obtained through spectral detection methods that can reflect the chemical composition and quality structure of optional tobacco leaves at the molecular level. The second tobacco leaf spectral data can include multiple data points, each corresponding to the spectral absorption response value (e.g., absorbance) of the selected tobacco leaf at a corresponding wavenumber. The second tobacco leaf spectral data can include various spectral data characterizing the physicochemical quality of the selected tobacco leaf, including spectral data obtained from different wavebands. Specifically, the second tobacco leaf spectral data can include second near-infrared spectral data. The second near-infrared spectral data can be the absorption spectrum of the selected tobacco leaf collected in the near-infrared band. The data structure of the second tobacco leaf spectral data (including the number of data points, wavenumber correspondence, and response value dimension) is completely matched with the first tobacco leaf spectral data, ensuring a basis for direct comparison and avoiding systematic errors introduced due to differences in acquisition conditions.

[0027] It should be noted that, to ensure complete data structure matching between the first and second tobacco leaf spectral data, the target and optional tobacco leaves can be processed using the same sample preparation standards to obtain target and optional tobacco leaf samples. Furthermore, using the same type of spectral acquisition equipment under the same spectral acquisition conditions, the target and optional tobacco leaf samples are subjected to spectral acquisition to obtain the first tobacco leaf spectral data for the target tobacco leaf and the second tobacco leaf spectral data for the optional tobacco leaf.

[0028] It should also be noted that the quality characteristics of the target tobacco leaf (such as chemical composition, sensory characteristics, and formulation suitability) are already clearly defined. Its first-stage tobacco leaf spectral data serves as a benchmark for similarity calculation and can be used to compare and analyze with the second-stage tobacco leaf spectral data of select tobacco leaves, thereby screening out suitable alternative tobacco leaves with matching performance. Using a tobacco leaf spectral similarity evaluation system to determine alternative tobacco leaves to replace the target tobacco leaf can quantify the consistency of spectral patterns at an overall level, thus achieving the transformation from "spectral similarity" to "tobacco leaf similarity."

[0029] In this embodiment, after determining at least one optional tobacco leaf, in order to collect tobacco leaf spectra from the optional tobacco leaf, the optional tobacco leaf can be prepared according to a preset sample preparation standard, and the prepared tobacco leaf sample can be collected to obtain the second tobacco leaf spectra data of the optional tobacco leaf.

[0030] Optionally, obtaining the second tobacco spectral data of at least one optional tobacco leaf includes: preparing a sample of the optional tobacco leaf to obtain a tobacco leaf sample; drying the tobacco leaf sample at a preset drying temperature and a preset drying time; grinding the dried tobacco leaf sample into powder to obtain a tobacco leaf powder sample; and using a spectral acquisition device to perform infrared scanning on the tobacco leaf powder sample to acquire spectral data in a preset spectral band to obtain the second tobacco spectral data corresponding to the optional tobacco leaf.

[0031] Sample preparation refers to a series of standardized processing operations performed on selectable tobacco leaves to obtain stable and comparable spectral data. This process can eliminate interference from differences in the physical state of selectable tobacco leaves (such as moisture content, particle size, and morphology) on spectral detection. Optionally, sample preparation includes at least one step such as stem removal, cutting and crushing, and uniform mixing. Sample preparation ensures the consistency of processing effects in subsequent drying and grinding stages, laying the foundation for the reliability of spectral data. The tobacco leaf sample can refer to an intermediate tobacco product with a uniform morphology and no obvious impurities (such as stems and debris) after the sample preparation step. The morphology of the tobacco leaf sample is usually broken tobacco leaf fragments (rather than powder), serving as a transitional form between the selectable tobacco leaves and subsequent drying and grinding steps. This ensures uniform moisture loss during drying and yields a powder sample with consistent particle size after grinding. The preset drying temperature refers to a predetermined constant temperature used for moisture control of the selectable tobacco leaf sample. Optionally, the preset drying temperature includes 30℃, 35℃, 40℃, and 45℃. Preferably, the preset drying temperature is 40℃. It should be noted that higher drying temperatures for tobacco leaf samples lead to the loss of low-boiling-point volatile aroma compounds; conversely, lower drying temperatures may fail to reduce the moisture content of the selected tobacco leaf samples. Therefore, to avoid excessively high temperatures causing the loss of low-boiling-point volatile aroma compounds while efficiently reducing the test moisture content to a stable range, a preset drying temperature of 40°C can be set. The preset drying time refers to a predetermined fixed duration for drying the tobacco leaf samples at the preset drying temperature. Optionally, the preset drying time can include 7 hours, 8 hours, 9 hours, and 10 hours, etc. Preferably, the preset drying time can be 8 hours. It should be noted that longer drying times at the preset drying temperature lead to the loss of low-boiling-point volatile aroma compounds, increasing unnecessary costs and potentially affecting the repeatability of spectral data; shorter drying times result in insufficient and uneven moisture removal, obscuring the spectral signals of the target chemical components and reducing the comparability and reliability of the spectral data. Both excessively long and short drying times will interfere with subsequent similarity assessment results. Furthermore, in order to ensure that the moisture content of tobacco leaf samples from different parts and production areas is stable within the range of 5%-6%, to minimize the interference of moisture on the spectrum, to achieve uniformity of moisture content among different types of selectable tobacco leaf samples, and to ensure the repeatability and comparability of subsequent spectral acquisition, the preset drying time can be set to 8 hours.

[0032] The dried tobacco leaf sample refers to a sample that has been dried at a preset temperature and time, resulting in a stable low moisture content (e.g., 5%-6%) without moisture fluctuations. The dried tobacco leaf sample remains in the physical state of dry tobacco leaf fragments, and its moisture content meets the requirements for tobacco spectral acquisition. It can be directly used in subsequent grinding steps, avoiding spectral data distortion due to residual moisture. The tobacco powder sample refers to a uniform powder obtained by grinding a selected dried tobacco leaf sample using a grinding device and then sieving it through a 60-mesh sieve (250μm aperture). The core characteristics of the tobacco powder sample are uniform particle size (ensuring consistent diffuse reflectance during spectral acquisition), absence of large particulate impurities, and the ability for uniform near-infrared light penetration and full utilization of the tobacco chemical components, ensuring the accuracy and representativeness of the spectral response values. The spectral acquisition equipment refers to the core detection instrument used to collect the spectral data of the tobacco powder sample. The spectral acquisition equipment may include a Fourier transform near-infrared spectrometer. The working principle of a Fourier transform near-infrared spectrometer is understandable. It generates interference light through an interferometer, illuminates the sample, and receives the reflected / transmitted light signals. The Fourier transform converts the time-domain signal into a frequency-domain (wavenumber) signal, outputting spectral response values ​​at different wavenumbers. Fourier transform near-infrared spectrometers possess stable wavenumber accuracy and response sensitivity, ensuring that the acquired spectral data accurately reflects the chemical composition characteristics of tobacco leaves. Infrared scanning refers to placing a tobacco powder sample in a sample cup, allowing it to settle naturally, and then placing the sample cup in the detection optical path of the Fourier transform near-infrared spectrometer. The spectrometer then irradiates the sample with near-infrared light and acquires signals according to preset parameters (such as the number of scans and resolution). The core of infrared scanning is to cause near-infrared light to interact with the chemical components (such as sugars, nicotine, and total nitrogen) in the tobacco powder sample, resulting in molecular vibrational absorption and generating spectral signals related to the type and content of the chemical components. The preset spectral band refers to the pre-defined wavenumber range used to acquire near-infrared spectral data. Optionally, the preset spectral band can be 3600-10000 cm⁻¹, representing the number of vibrations of light waves per centimeter. The preset spectral band covers the molecular vibrational (combination and overtone) absorption peaks of core chemical components in tobacco leaves, such as sugars, nicotine, and total nitrogen, which can comprehensively reflect the chemical composition characteristics of tobacco leaves and provide effective data support for subsequent similarity assessment.

[0033] In one embodiment, after identifying at least one selectable tobacco leaf, the selectable tobacco leaf can be processed according to a preset sample preparation step (removing stems, cutting and crushing, and uniformly mixing) to obtain a tobacco leaf sample with no obvious impurities and uniform morphology. Further, the tobacco leaf sample is placed in a preset drying temperature environment and dried for a preset drying time to ensure that the sample moisture content remains stable at a low level, resulting in a dried tobacco leaf sample. Further, the dried tobacco leaf sample is ground using a grinding device and passed through a 60-mesh sieve to obtain a tobacco leaf powder sample with uniform particle size. Further, the tobacco leaf powder sample is placed in a sample cup and naturally compacted. The sample cup is then placed in the detection optical path of a Fourier transform near-infrared spectrometer. The Fourier transform near-infrared spectrometer is used to irradiate the tobacco leaf powder sample with near-infrared light and acquire signals according to preset parameters (such as the number of scans and resolution). Spectral response data under a preset spectral band are acquired, ultimately obtaining second tobacco leaf spectral data that reflects the chemical composition characteristics of the selectable tobacco leaf.

[0034] It should be noted that the target tobacco leaf can be processed using the second tobacco leaf spectral data acquisition method described above to obtain the first tobacco leaf spectral data of the target tobacco leaf. This embodiment will not repeat the description here.

[0035] In one scenario, the provided solution involves acquiring first spectral data of the target tobacco leaf and second spectral data of at least one optional tobacco leaf, comprising at least one of the following: performing spectral scanning on the target tobacco leaf and the optional tobacco leaf to obtain first spectral data of the target tobacco leaf and second spectral data of the optional tobacco leaf; retrieving a pre-recorded spectral record of the target tobacco leaf as first spectral data of the target tobacco leaf, and retrieving a pre-recorded spectral record of the optional tobacco leaf as second spectral data of the optional tobacco leaf.

[0036] S120. For at least one optional tobacco leaf, determine the fractal similarity to the optional tobacco leaf based on the second tobacco leaf spectral data and the first tobacco leaf spectral data of the optional tobacco leaf.

[0037] In one context, fractal similarity can be understood as a quantitative evaluation index measuring the consistency of the spectral quality of two sets of tobacco leaves. Fractal similarity can be a numerical value specifically used to characterize the degree of similarity in the multi-scale deep structure and morphological features of two tobacco leaf spectral curves. Fractal similarity can be used to characterize the similarity of the intrinsic quality of tobacco leaves. It can also indicate the degree of similarity in the deep spectral structure between the spectral data of the first and second tobacco leaves. The deep spectral structure can be understood as the hidden tobacco leaf quality characteristics information within the spectral curves. This deep spectral structure can be the inherent structural features of the tobacco leaf spectral curves, such as their overall morphology and local fluctuations, corresponding to the chemical composition and overall quality of the tobacco leaves. For example, the microscopic fluctuations and overall trend differences in the spectra of different tobacco leaves all fall within the scope of the deep spectral structure.

[0038] In one scenario, the provided scheme determines the fractal similarity to the selected tobacco leaves based on the second and first tobacco leaf spectral data, including at least one of the following: processing the first and second tobacco leaf spectral data in segments to obtain multi-scale fluctuation features, resulting in a first and second fractal feature vectors; calculating the fractal similarity based on the first and second fractal feature vectors; and extracting the multi-scale feature vectors of the first and second tobacco leaf spectra respectively, and calculating their fractal similarity through vector space comparison.

[0039] In one embodiment, after obtaining the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf, for the at least one optional tobacco leaf, the first tobacco leaf spectral data and the second tobacco leaf spectral data are spectrally segmented to obtain multiple first segmented spectra corresponding to the first tobacco leaf spectral data and multiple second segmented spectra corresponding to the second tobacco leaf spectral data. Further, for each first segmented spectrum, the fluctuation difference value of the first segmented spectrum under multiple different delay step sizes is determined, and the multiple fluctuation difference values ​​are fitted to obtain fractal feature values ​​corresponding to the first segmented spectrum. Based on the fractal feature values ​​corresponding to the multiple first segmented spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined. Similarly, for each second segmented spectrum, the fluctuation difference value of the second segmented spectrum under multiple different delay step sizes is determined, and the multiple fluctuation difference values ​​are fitted to obtain fractal feature values ​​corresponding to the second segmented spectrum. Based on the fractal feature values ​​corresponding to the multiple second segmented spectra, a first fractal feature vector corresponding to the optional tobacco leaf is determined. Furthermore, based on the first fractal feature vector and the second fractal feature vector, the correlation coefficient and vector similarity between the first tobacco leaf spectral data and the second tobacco leaf spectral data are determined, and based on the correlation coefficient and vector similarity, the fractal similarity corresponding to the selectable tobacco leaf is determined.

[0040] S130. Based on at least one fractal similarity, determine alternative tobacco leaves from at least one selection of tobacco leaves to replace the target tobacco leaf.

[0041] In one context, alternative tobacco leaves can be understood as high-quality tobacco raw materials that can equivalently replace the target tobacco leaves. Alternative tobacco leaves can be optional tobacco leaves whose spectral deep structure is highly similar to that of the target tobacco leaves and whose quality matches well, making them suitable for replacing cigarette formulations. For example, among multiple optional tobacco leaves, the stock tobacco leaves with the highest fractal similarity and the best quality match to the target tobacco leaves are the finally selected alternative tobacco leaves.

[0042] In one scenario, the provided scheme, based on at least one fractal similarity, determines alternative tobacco leaves from at least one selectable tobacco leaf to replace the target tobacco leaf, which may include at least one of the following: sorting at least one selectable tobacco leaf in descending order of fractal similarity, and determining a predetermined number of selectable tobacco leaves at the top of the list as alternative tobacco leaves; comparing at least one fractal similarity with a predetermined similarity threshold, and determining selectable tobacco leaves whose fractal similarity is not less than the predetermined similarity threshold as alternative tobacco leaves.

[0043] Optionally, based on at least one fractal similarity, determining alternative tobacco leaves from at least one optional tobacco leaf to replace the target tobacco leaf includes: sorting at least one optional tobacco leaf in descending order of fractal similarity, and determining a predetermined number of optional tobacco leaves at the top of the list as alternative tobacco leaves.

[0044] In one scenario, the preset quantity can be a pre-configured value used to limit the scale of replaceable tobacco leaves. The preset quantity can be flexibly adjusted based on tobacco leaf substitution needs (such as the required amount of raw materials for the formula, the remaining inventory of existing optional tobacco leaves, and the reserve requirements for alternative formula adjustments). Optionally, the preset quantity can be 1, 2, or 3, etc.

[0045] In one implementation, after obtaining the fractal similarity corresponding to at least one optional tobacco leaf, the at least one optional tobacco leaf can be sorted in descending order of fractal similarity. Furthermore, a predetermined number of optional tobacco leaves at the top of the sorted list can be selected as alternative tobacco leaves to replace the target tobacco leaf.

[0046] Optionally, based on at least one fractal similarity, a substitute tobacco leaf for replacing the target tobacco leaf is determined from at least one optional tobacco leaf, including: comparing at least one fractal similarity with a preset similarity threshold, and determining optional tobacco leaves whose fractal similarity is not less than the preset similarity threshold as substitute tobacco leaves.

[0047] The preset similarity threshold can be understood as a pre-determined benchmark value for judging similarity compliance. The preset similarity threshold can be a pre-set acceptance standard; only tobacco leaves meeting this standard can be used as substitutes. The preset similarity threshold can be flexibly adjusted according to the needs of cigarette formulation blending. Optionally, the preset similarity threshold may include 60%, 70%, or 80%, etc.

[0048] In one implementation, after obtaining the fractal similarity corresponding to at least one optional tobacco leaf, the at least one fractal similarity can be compared with a preset similarity threshold. Further, optional tobacco leaves with fractal similarity not less than the preset similarity threshold can be determined based on the similarity comparison results, and the determined optional tobacco leaves can be used as substitutes for the target tobacco leaf.

[0049] The technical solution of this disclosure involves spectral segmentation of the first and second tobacco leaf spectral data to obtain multiple first segmented spectra corresponding to the first tobacco leaf spectral data and multiple second segmented spectra corresponding to the second tobacco leaf spectral data. Further, based on the multiple first segmented spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined, and based on the multiple second segmented spectra, a second fractal feature vector corresponding to the selectable tobacco leaf is determined. Furthermore, based on the first and second fractal feature vectors, a fractal similarity corresponding to the selectable tobacco leaf is determined. This achieves the effect of precisely capturing local fine structural information of the spectrum through segmented fractal feature extraction. By calculating similarity based on fractal feature vectors, the intrinsic quality differences of tobacco leaves can be deeply characterized, significantly improving the accuracy of tobacco leaf similarity evaluation.

[0050] Figure 2 This is a schematic flowchart illustrating another method for processing tobacco leaf spectral data provided in this embodiment. The technical solution of this embodiment can be combined with other embodiments; for the same or related parts, descriptions of other embodiments can be used, and will not be repeated here. Figure 2 As shown, the method in this embodiment may specifically include: S210, acquire the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf.

[0051] S220. For at least one optional tobacco leaf, the spectral data of the first tobacco leaf and the spectral data of the second tobacco leaf are spectrally segmented to obtain multiple first segmented spectra corresponding to the spectral data of the first tobacco leaf and multiple second segmented spectra corresponding to the spectral data of the second tobacco leaf.

[0052] In one context, spectral segmentation can be understood as a process of splitting a continuous spectrum. Spectral segmentation can be the process of dividing a complete spectrum into multiple sub-spectral segments according to wavelength ranges. For example, splitting the first tobacco leaf spectral data into multiple sub-spectral segments based on shortwave, mediumwave, and longwave. The first segmented spectrum can be understood as a sub-spectral fragment obtained after splitting the first tobacco leaf spectral data. The first segmented spectrum can also be a single-end spectrum obtained by segmenting the first tobacco leaf spectral data; for example, the various wavelength range segments obtained after segmenting the first tobacco leaf spectral data. Similarly, the second segmented spectrum can be understood as a sub-spectral fragment obtained by segmenting the second tobacco leaf spectral data. The second segmented spectrum can also be a single-end spectrum obtained by segmenting the second tobacco leaf spectral data; for example, the various wavelength range segments obtained after segmenting the second tobacco leaf spectral data.

[0053] In one scenario, the method for determining the first segment spectrum and the method for determining the second segment spectrum in the provided scheme can be the same. The specific method for determining the first segment spectrum can be explained below using the first segment spectrum as an example.

[0054] In one scenario, the provided scheme involves spectral segmentation of the first tobacco leaf spectral data to obtain multiple first segmented spectra corresponding to the first tobacco leaf spectral data. This segmentation can include at least one of the following: employing a moving window sliding interception method, setting a fixed number of window points and a sliding step size, and continuously sliding and intercepting along the entire first tobacco leaf spectrum to generate multiple first segmented spectra; rigidly segmenting the first tobacco leaf spectrum according to a fixed wavelength range, with fixed segment boundaries and no overlap between segments, resulting in multiple sets of independent first segmented spectra; or dynamically adjusting window parameters based on the position of spectral characteristic absorption peaks, refining the segmentation in feature-rich regions and increasing the segmentation span in smooth bands, adaptively dividing to obtain multiple first segmented spectra. One of these determination methods is described in detail below.

[0055] Optionally, the first tobacco leaf spectral data is segmented to obtain multiple first segmented spectra corresponding to the first tobacco leaf spectral data, including: using a preset sliding window to continuously slide and capture along the first tobacco leaf spectral data according to a preset sliding step size to obtain multiple first segmented spectra.

[0056] In one scenario, a preset sliding window can be understood as a pre-defined spectral interception unit. The preset sliding window can be a pre-defined interception interval containing a certain number of sampling points within a single spectral segment; that is, the window size of the preset sliding window can correspond to the spectral data points contained in a single segmented spectrum. For example, a pre-defined interception interval containing 200 spectral data points in a single segmented spectrum is a preset sliding window. The preset sliding step size can be a pre-determined span of single window translation. The preset sliding step size can be the number of sampling points the preset sliding window moves backward along the spectral curve each time. For example, setting the preset sliding window to move 40 spectral data points sequentially each time, these 40 spectral data points are the preset sliding step size. Continuous sliding interception can be understood as the processing action of continuously translating the preset sliding window and segmenting the spectrum. Continuous sliding interception can be the preset sliding window translating sequentially along the entire spectrum, intercepting a segment of the spectrum after each preset sliding step.

[0057] In one implementation, the number of spectral points contained in the sliding window is predetermined to obtain a preset sliding window, and the sampling point span of a single slide is predetermined to obtain a preset sliding step size. Further, starting from the starting point of the first tobacco leaf spectral data, the spectral data within the corresponding range is selected using the preset sliding window to obtain the first segment of the spectrum. Further, the preset sliding window is controlled to shift a fixed number of spectral points sequentially according to the preset sliding step size. Each time a shift is completed, a new segment of the spectrum is captured and generated. This sliding capture operation is repeated until the end of the preset sliding window reaches the end of the entire spectrum, ultimately resulting in multiple first segment spectra.

[0058] For example, the first segment of the spectrum can be determined using the following formula: in, Indicates the number of segments in the first spectrum; This indicates the total number of spectral data points contained in the first tobacco leaf spectral data; The window size represents the preset sliding window size, and the number of spectral data points contained in a single first segment spectrum. Indicates the preset sliding step size; This indicates a floor operation, ensuring that the number of segments is an integer.

[0059] S230. Based on multiple first segmented spectra, determine the first fractal feature vector corresponding to the target tobacco leaf, and based on multiple second segmented spectra, determine the second fractal feature vector corresponding to the optional tobacco leaf.

[0060] In one scenario, the first fractal feature vector can be understood as a quantized array summarizing the spectral structural features of each first segment of the target tobacco leaf. The first fractal feature vector can be a feature vector formed by integrating the fractal features extracted from each first segment spectrum, used to characterize the deep spectral structure of the first tobacco leaf spectral data. The second fractal feature vector can be understood as a quantized array summarizing the spectral structural features of each second segment of the selectable tobacco leaf. The second fractal feature vector can be a feature vector formed by integrating the fractal features extracted from each second segment spectrum, used to characterize the deep spectral structure of the second tobacco leaf data.

[0061] It should be noted that the determination methods for the first fractal eigenvector and the second fractal eigenvector are the same. The following explanation will take the first fractal eigenvector as an example to illustrate its determination method.

[0062] Optionally, based on multiple first segment spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined, including: for each first segment spectrum, determining the fluctuation difference value of the first segment spectrum under multiple different delay step sizes; performing fitting operations on multiple fluctuation difference values ​​to obtain fractal feature values ​​corresponding to the first segment spectrum; and determining the first fractal feature vector corresponding to the target tobacco leaf based on the fractal feature values ​​corresponding to multiple first segment spectra.

[0063] In one scenario, the delay step size can be understood as the selection interval parameter between spectral data points when calculating spectral fluctuations. The delay step size can be the number of data points between consecutive sampling points when performing difference calculations on spectral data. Changing various interval values ​​results in multiple different delay step sizes. For example, on the first segment of the spectrum, taking values ​​at intervals of 3, 6, and 9 data points respectively for difference calculations represents three different delay step sizes. The fluctuation difference value can be understood as the calculation result used to quantify the degree of local fluctuation changes in the segmented spectrum. The fluctuation difference value can be calculated based on the selected delay step size, showing the difference between the spectral values ​​of corresponding points in the segmented spectrum, and is used to characterize the magnitude of fluctuations in the spectral curve. The fluctuation difference value can include the semivariogram value, meaning the semivariogram function can be used to calculate the fluctuation difference value. Fitting operations can be understood as data processing operations that fit the changing patterns based on multiple sets of difference data. Fitting operations can be performing a pattern fitting on all fluctuation difference values ​​calculated at different delay step sizes for the same first segment of the spectrum. Fractal eigenvalues ​​can be understood as quantified numerical values ​​that characterize the fractal structural properties of a single segment of the spectrum. A fractal eigenvalue can be a unique feature parameter obtained through fitting operations, uniquely representing the intrinsic structural similarity of that segment of the spectrum.

[0064] Optionally, determining the fluctuation difference value of the first segmented spectrum under multiple different delay steps includes: for each delay step, determining at least one pair of spectral points of the first segmented spectrum under the delay step; wherein the pair of spectral points includes a first spectral point and a second spectral point, the second spectral point being located after the first spectral point and separated by a delay step; for at least one pair of spectral points, determining the square of the absorbance difference corresponding to the pair of spectral points based on the first absorbance value of the first spectral point and the second absorbance value of the second spectral point in the pair of spectral points; and determining the fluctuation difference value of the first segmented spectrum under the delay step based on the square of the absorbance difference corresponding to the at least one pair of spectral points.

[0065] In one scenario, a spectral point pair can be understood as a set of consecutive spectral data points paired according to a delay step size. A spectral point pair can be a data pairing unit consisting of a first spectral point positioned earlier and a second spectral point shifted by one delay step size. The first spectral point can be understood as a single spectral sampling point positioned earlier in the spectral point pair. The second spectral point can be a paired sampling point shifted backward by one delay step size in the spectral point pair. For example, assuming a delay step size of 3 data points, the first spectral point could be the 3rd spectral sampling point in the first segmented spectrum, and the second spectral point could be the 6th spectral sampling point in the first segmented spectrum. The 3rd and 6th spectral sampling points can form a spectral point pair.

[0066] In one scenario, the first absorbance value can be understood as the spectral detection value corresponding to the first spectral point in the first segmented spectrum. The first absorbance value can be the measured absorbance data corresponding to the position of the first spectral point, acquired by the spectral acquisition device. The second absorbance value can be understood as the spectral detection value corresponding to the second spectral point in the first segmented spectrum. The second absorbance value can also be the measured absorbance data corresponding to the position of the second spectral point, acquired by the spectral acquisition device. The square of the absorbance difference can be understood as the result of squaring the absorbance difference between the two spectral points.

[0067] In one implementation, for each delay step, a first spectral point located earlier in the first segmented spectrum is sequentially selected, and a corresponding second spectral point is obtained by matching points that cross the delay step backward. Two spectral points form a spectral point pair, thus obtaining at least one spectral point pair. Further, for at least one spectral point pair, a first absorbance value corresponding to the first spectral point and a second absorbance value corresponding to the second spectral point are extracted from the first segmented spectrum. Further, the square of the difference between the first and second absorbance values ​​is determined to obtain the squared absorbance difference corresponding to the spectral point pair. Further, the squared absorbance differences corresponding to at least one spectral point pair are summed to obtain a total squared difference. Based on the total squared difference and the number of spectral point pairs, the fluctuation difference value of the first segmented spectrum at that delay step is obtained.

[0068] For example, the fluctuation difference value of the first segment spectrum at the delay step can be determined by the following formula: in, Indicates the first The first segment of the spectrum, at the delay step size The fluctuation difference value below; Indicates the delay step size The number of spectral point pairs in the first segment of the spectrum; Indicates the first Within the first segment of the spectrum, the first The first absorbance value of the first spectral point; Indicates the first Within the first segment of the spectrum, from Point-to-back interval delay step The second absorbance value of the second spectral point; Indicates the delay step size.

[0069] In one implementation, for each first segment spectrum, after obtaining the fluctuation difference values ​​of the first segment spectrum under multiple different delay step sizes, the logarithm of each delay step size and the fluctuation difference values ​​under that delay step size is taken to obtain the logarithmic data corresponding to that delay step size. Further, a linear fit is performed on the logarithmic data under multiple delay step sizes to obtain the slope of the fitted line. Further, based on the slope of the fitted line and the curve topology dimension of the first tobacco leaf spectral data, fractal feature values ​​corresponding to the first segment spectrum are obtained. Further, according to the left-to-right generation arrangement order of each first segment spectrum on the first tobacco leaf spectral data, the fractal feature values ​​calculated for each first segment spectrum are sequentially arranged, integrated, and collected, thereby forming a first fractal feature vector that can comprehensively characterize the spectral structure information of the target tobacco leaf.

[0070] For example, the fractal eigenvalue corresponding to the first segment of the spectrum can be determined using the following formula: in, Indicates the first The fractal characteristic value corresponding to the first segment of the spectrum; The curve topology dimension of the first tobacco leaf spectral data is typically set to 1. This represents the slope of the fitted line, which is determined by... and The slope of the line in a linear fit. Indicates the first The first segment of the spectrum, at the delay step size The fluctuation difference value below, This indicates taking the logarithm of the fluctuation difference value. Indicates the delay step size Take the logarithm.

[0071] S240. Determine the fractal similarity to the selectable tobacco leaves based on the first fractal feature vector and the second fractal feature vector.

[0072] Optionally, based on the first fractal feature vector and the second fractal feature vector, the fractal similarity to the selectable tobacco leaves is determined, including: based on the first fractal feature vector and the second fractal feature vector, determining the correlation coefficient and vector similarity between the spectral data of the first tobacco leaf and the spectral data of the second tobacco leaf; and based on the correlation coefficient and vector similarity, determining the fractal similarity to the selectable tobacco leaves.

[0073] In one context, the correlation coefficient can be understood as a quantitative indicator measuring the degree of linear correlation between two fractal feature vectors. The correlation coefficient can be a statistical coefficient calculated based on the numerical distribution of the first and second fractal feature vectors. For example, the correlation coefficient can range from -1 to 1, with values ​​closer to 1 indicating a closer linear relationship between the two fractal feature vectors. Optionally, the correlation coefficient can include at least one of the following: Pearson correlation coefficient, Spearman's rank correlation coefficient, or Kendall's correlation coefficient. Vector similarity can be understood as a parameter measuring the overall overlap between two fractal feature vectors from the perspective of spatial angle. Vector similarity can be used to characterize the directional closeness of two fractal feature vectors in multidimensional space. Optionally, vector similarity can include at least one of the following: cosine similarity, Euclidean distance similarity, Manhattan distance similarity, or Langevin distance similarity.

[0074] In one implementation, the Pearson correlation coefficient function can be used to calculate the first and second fractal feature vectors to obtain the correlation coefficient between the first and second tobacco leaf spectral data. Furthermore, the Lambda distance similarity function can be used to calculate the vector similarity between the first and second tobacco leaf spectral data. Further, the fractal similarity to the selected tobacco leaf can be determined based on the correlation coefficient and vector similarity.

[0075] In one scenario, the provided scheme determines the fractal similarity to the selectable tobacco leaves based on the correlation coefficient and vector similarity. This can include at least one of the following: processing the correlation coefficient and vector similarity using a weighted summation operation to obtain the fractal similarity; determining the ratio between the correlation coefficient and vector similarity to obtain the fractal similarity; selecting the minimum value between the correlation coefficient and vector similarity as the fractal similarity; or calculating the arithmetic mean of the correlation coefficient and vector similarity and using the arithmetic mean as the fractal similarity. One such determination method is explained in detail below.

[0076] Optionally, based on the correlation coefficient and vector similarity, the fractal similarity corresponding to the selectable tobacco leaves is determined, including: determining the ratio between the correlation coefficient and the vector similarity to obtain the fractal similarity corresponding to the selectable tobacco leaves.

[0077] In one implementation, after obtaining the correlation coefficient and vector similarity between the first tobacco leaf spectral data and the second tobacco leaf spectral data, the ratio between the correlation coefficient and the vector similarity can be determined, and this ratio can be determined as the fractal similarity corresponding to the selected tobacco leaf.

[0078] For example, fractal similarity can be determined using the following formula: in, Represents the first fractal eigenvector Second fractal eigenvectors The Pearson correlation coefficient; Represents the first fractal eigenvector. The fractal characteristic value of the first segment of the spectrum; This indicates the number of segmented spectra, i.e., the number of the first segmented spectra or the number of the second segmented spectra; Represents the first fractal eigenvector The average value; The second fractal eigenvector represents the... The fractal characteristic value of the second segment of the spectrum; Represents the second fractal eigenvector The average value.

[0079] in, Represents the first fractal eigenvector Second fractal eigenvectors The similarity of the Lancz distance.

[0080] in, This represents fractal similarity.

[0081] S250. Based on at least one fractal similarity, determine alternative tobacco leaves from at least one selection of tobacco leaves to replace the target tobacco leaf.

[0082] The technical solution of this disclosure provides objective and standardized basic data for subsequent precise matching of tobacco leaves by acquiring the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf. Furthermore, for at least one optional tobacco leaf, based on the second and first tobacco leaf spectral data, a fractal similarity is determined. This fractal similarity indicates the degree of similarity in the deep spectral structure between the first and second tobacco leaf spectral data, overcoming the limitation of only comparing apparent spectral data. By using fractal similarity to accurately quantify the intrinsic quality differences of tobacco leaves, subjective errors caused by manual tasting are effectively reduced, improving the accuracy of tobacco leaf similarity assessment. Furthermore, based on at least one fractal similarity, alternative tobacco leaves for replacing the target tobacco leaf are determined from at least one optional tobacco leaf. Relying on the quantified fractal similarity as an objective screening standard, suitable alternative tobacco leaves can be quickly and accurately screened, reducing the uncertainty of manual tasting and improving the efficiency of tobacco raw material selection and formula stability. The technical solution of this disclosure solves the problems of high workload, strong subjectivity, low efficiency and difficulty in finding the best alternative tobacco leaves caused by the use of manual judgment in related technologies. By using tobacco leaf spectrum combined with fractal similarity to quantify the deep structural similarity of the spectrum, the subjective interference of human sensory evaluation is eliminated, and suitable alternative tobacco leaves are objectively and accurately screened, thereby improving the efficiency of tobacco leaf raw material selection and improving the stability of formula quality.

[0083] Figure 3 This is a flowchart illustrating an optional embodiment of a tobacco leaf spectral data processing method provided by this disclosure. Specific implementation details of this method can be found in the following embodiments. Technical features that are the same as or similar to those in the above embodiments will not be repeated here. See also... Figure 3 The method in this embodiment specifically includes the following steps: First, first-segment spectral data of the target tobacco leaf and second-segment spectral data of various stock tobacco leaves are collected. Further, for each type of stock tobacco leaf, a preset sliding window is used to continuously capture segments along the first-segment spectral data at a preset sliding step size to obtain multiple first-segment spectra. Similarly, a preset sliding window is used to continuously capture segments along the second-segment spectral data at a preset sliding step size to obtain multiple second-segment spectra. Further, for each first-segment spectrum, the fluctuation difference values ​​of the first-segment spectrum under multiple different delay step sizes are determined; multiple fluctuation difference values ​​are fitted to obtain fractal feature values ​​corresponding to the first-segment spectra; based on the fractal feature values ​​corresponding to the multiple first-segment spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined. Similarly, for each second-segment spectrum, the fluctuation difference values ​​of the second-segment spectrum under multiple different delay step sizes are determined; multiple fluctuation difference values ​​are fitted to obtain fractal feature values ​​corresponding to the second-segment spectra; based on the fractal feature values ​​corresponding to the multiple second-segment spectra, a second fractal feature vector corresponding to the stock tobacco leaf is determined. Furthermore, the Pearson correlation coefficient function can be used to calculate the correlation coefficient between the first and second fractal feature vectors to obtain the correlation coefficient between the first and second tobacco leaf spectral data. Further, the Lambda distance similarity function can be used to calculate the vector similarity between the first and second tobacco leaf spectral data. Further, the ratio between the correlation coefficient and the vector similarity can be determined, and this ratio can be used as the fractal similarity corresponding to the stockpiled tobacco leaves. Further, having obtained the fractal similarities corresponding to various stockpiled tobacco leaves, the various stockpiled tobacco leaves are sorted in descending order of fractal similarity, and the stockpiled tobacco leaf ranked first is determined as the substitute tobacco leaf for the target tobacco leaf.

[0084] This solution addresses three major technical shortcomings of existing near-infrared spectroscopy (NIRS) similarity analysis methods: reliance on manual judgment for tobacco leaf substitution in cigarette formulations, low efficiency, and high subjectivity; and the blindness of preprocessing, loss of information in dimensionality reduction, and neglect of spectral sequence structural features. It innovatively introduces fractal geometry theory into the field of tobacco leaf NIRS analysis. By quantitatively characterizing the multi-scale structure of molecular vibrations, hierarchical overlap of spectral bands, long-range correlation of baseline drift, and noise multifractals in the spectrum through fractal dimensions, it achieves a technological breakthrough by enabling scientific assessment of tobacco leaf similarity without complex spectral preprocessing and dimensionality reduction. This provides a more reliable and interpretable mathematical tool for digital evaluation of tobacco leaf quality and formulation substitution. This solution can be widely applied to tobacco raw material quality evaluation, digital management of tobacco inventory, intelligent substitution of cigarette formulations, and optimization of leaf threshing and re-drying processes, significantly improving the efficiency of tobacco resource allocation and the scientific nature of formulation design. With the accelerated digital transformation of the tobacco industry, this technology is expected to be deeply integrated with artificial intelligence and big data platforms to build a full-chain quality traceability and intelligent decision-making system for tobacco, showing broad application prospects.

[0085] Figure 4 This is a schematic diagram of a tobacco leaf spectral data processing device provided in an embodiment of this disclosure. Figure 4 As shown, the information generation device includes: a spectral data acquisition module 310, a similarity determination module 320, and a tobacco leaf determination module 330. The spectral data acquisition module 310 is used to acquire first tobacco leaf spectral data of the target tobacco leaf and second tobacco leaf spectral data of at least one optional tobacco leaf. The similarity determination module 320 is used to determine, for at least one of the optional tobacco leaves, a fractal similarity corresponding to the optional tobacco leaf based on the second tobacco leaf spectral data and the first tobacco leaf spectral data, wherein the fractal similarity indicates the degree of similarity in the deep spectral structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data. The tobacco leaf determination module 330 is used to determine, based on at least one of the fractal similarities, a substitute tobacco leaf for replacing the target tobacco leaf from at least one of the optional tobacco leaves.

[0086] The technical solution of this disclosure provides objective and standardized basic data for subsequent precise matching of tobacco leaves by acquiring the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf. Furthermore, for at least one optional tobacco leaf, based on the second and first tobacco leaf spectral data, a fractal similarity is determined. This fractal similarity indicates the degree of similarity in the deep spectral structure between the first and second tobacco leaf spectral data, overcoming the limitation of only comparing apparent spectral data. By using fractal similarity to accurately quantify the intrinsic quality differences of tobacco leaves, subjective errors caused by manual tasting are effectively reduced, improving the accuracy of tobacco leaf similarity assessment. Furthermore, based on at least one fractal similarity, alternative tobacco leaves for replacing the target tobacco leaf are determined from at least one optional tobacco leaf. Relying on the quantified fractal similarity as an objective screening standard, suitable alternative tobacco leaves can be quickly and accurately screened, reducing the uncertainty of manual tasting and improving the efficiency of tobacco raw material selection and formula stability. The technical solution of this disclosure solves the problems of high workload, strong subjectivity, low efficiency and difficulty in finding the best alternative tobacco leaves caused by the use of manual judgment in related technologies. By using tobacco leaf spectrum combined with fractal similarity to quantify the deep structural similarity of the spectrum, the subjective interference of human sensory evaluation is eliminated, and suitable alternative tobacco leaves are objectively and accurately screened, thereby improving the efficiency of tobacco leaf raw material selection and improving the stability of formula quality.

[0087] In some embodiments of this disclosure, optionally, the similarity determination module 320 includes: a spectral segmentation submodule, a feature vector determination submodule, and a fractal similarity determination submodule. The spectral segmentation submodule is used to perform spectral segmentation on the first tobacco leaf spectral data and the second tobacco leaf spectral data respectively, obtaining multiple first segmented spectra corresponding to the first tobacco leaf spectral data and multiple second segmented spectra corresponding to the second tobacco leaf spectral data; the feature vector determination submodule is used to determine a first fractal feature vector corresponding to the target tobacco leaf based on the multiple first segmented spectra, and to determine a second fractal feature vector corresponding to the optional tobacco leaf based on the multiple second segmented spectra; the fractal similarity determination submodule is used to determine the fractal similarity to the optional tobacco leaf based on the first fractal feature vector and the second fractal feature vector.

[0088] In some embodiments of this disclosure, optionally, the feature vector determination submodule includes: a difference value determination unit, a fractal feature value determination unit, and a feature vector determination unit. The difference value determination unit is used to determine, for each of the first segmented spectra, the fluctuation difference value of the first segmented spectrum under multiple different delay step sizes; the fractal feature value determination unit is used to perform fitting operations on the multiple fluctuation difference values ​​to obtain fractal feature values ​​corresponding to the first segmented spectrum; and the feature vector determination unit is used to determine a first fractal feature vector corresponding to the target tobacco leaf based on the fractal feature values ​​corresponding to the multiple first segmented spectra.

[0089] In some embodiments of this disclosure, optionally, the difference value determination unit is specifically configured to determine at least one pair of spectral points of the first segmented spectrum under each delay step; wherein, the pair of spectral points includes a first spectral point and a second spectral point, the second spectral point being located after the first spectral point and separated by the delay step; for at least one pair of spectral points, determining the square of the absorbance difference corresponding to the pair of spectral points based on the first absorbance value of the first spectral point and the second absorbance value of the second spectral point in the pair of spectral points; and determining the fluctuation difference value of the first segmented spectrum under the delay step based on the square of the absorbance difference corresponding to at least one pair of spectral points.

[0090] In some embodiments of this disclosure, optionally, the fractal similarity determination submodule includes: a vector similarity determination unit and a fractal similarity determination unit. The vector similarity determination unit is used to determine the correlation coefficient and vector similarity between the first tobacco leaf spectral data and the second tobacco leaf spectral data based on the first fractal feature vector and the second fractal feature vector; the fractal similarity determination unit is used to determine the fractal similarity corresponding to the optional tobacco leaf based on the correlation coefficient and the vector similarity.

[0091] In some embodiments of this disclosure, optionally, the fractal similarity determination unit is specifically used to determine the ratio between the correlation coefficient and the vector similarity to obtain the fractal similarity corresponding to the optional tobacco leaf.

[0092] In some embodiments of this disclosure, optionally, the apparatus further includes: a spectral data acquisition module, used to prepare a sample of at least one optional tobacco leaf, obtain a tobacco leaf sample, dry the tobacco leaf sample at a preset drying temperature and a preset drying time, and grind the dried tobacco leaf sample into powder to obtain a tobacco leaf powder sample; and to perform infrared scanning on the tobacco leaf powder sample using a spectral acquisition device to acquire spectral data in a preset spectral band to obtain second tobacco leaf spectral data corresponding to the optional tobacco leaf.

[0093] The tobacco leaf spectral data processing device provided in this disclosure can execute the tobacco leaf spectral data processing method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of executing the method.

[0094] It is worth noting that the various units and modules included in the above-mentioned tobacco leaf spectral data processing device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of this disclosure.

[0095] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0096] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0098] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as tobacco leaf spectral data processing methods.

[0099] In some embodiments, the tobacco leaf spectral data processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into random access memory (RAM) 13 and executed by processor 11, one or more steps of the tobacco leaf spectral data processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the tobacco leaf spectral data processing method by any other suitable means (e.g., by means of firmware).

[0100] Various implementations of the systems and techniques described above herein can be implemented in digital circuit systems, integrated circuits, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chips (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, at least one input device, and at least one output device.

[0101] Computer programs used to implement the tobacco leaf spectral data processing method of this disclosure can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0102] This disclosure provides a computer-readable storage medium storing computer instructions for causing a processor to execute a tobacco leaf spectral data processing method, comprising: acquiring first tobacco leaf spectral data of a target tobacco leaf and second tobacco leaf spectral data of at least one optional tobacco leaf; determining, for at least one optional tobacco leaf, a fractal similarity corresponding to the optional tobacco leaf based on the second tobacco leaf spectral data of the optional tobacco leaf and the first tobacco leaf spectral data, wherein the fractal similarity is used to indicate the degree of similarity in the spectral depth structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data; and determining, based on at least one fractal similarity, a substitute tobacco leaf for replacing the target tobacco leaf from at least one optional tobacco leaf.

[0103] In the context of this disclosure, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM, also known as flash memory), optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0104] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a cathode-ray tube (CRT) or liquid crystal display (LCD) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0105] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0106] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0107] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of embodiments of this disclosure.

[0108] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements a tobacco leaf spectral data processing method according to any embodiment of this disclosure.

[0109] In implementing a computer program product, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0110] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.

[0111] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for processing spectral data of tobacco leaves, characterized in that, include: Acquire first tobacco leaf spectral data of the target tobacco leaf and second tobacco leaf spectral data of at least one optional tobacco leaf; For at least one of the optional tobacco leaves, a fractal similarity corresponding to the optional tobacco leaf is determined based on the second tobacco leaf spectral data and the first tobacco leaf spectral data of the optional tobacco leaf. The fractal similarity is used to indicate the degree of similarity of the deep spectral structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data. Based on at least one of the fractal similarities, alternative tobacco leaves for replacing the target tobacco leaf are determined from at least one of the optional tobacco leaves.

2. The method for processing tobacco leaf spectral data according to claim 1, characterized in that, The step of determining the fractal similarity to the optional tobacco leaf based on the second tobacco leaf spectral data and the first tobacco leaf spectral data includes: The first tobacco leaf spectral data and the second tobacco leaf spectral data are respectively divided into spectral segments to obtain multiple first segmented spectra corresponding to the first tobacco leaf spectral data and multiple second segmented spectra corresponding to the second tobacco leaf spectral data; Based on multiple first segmented spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined, and based on multiple second segmented spectra, a second fractal feature vector corresponding to the optional tobacco leaf is determined; Based on the first fractal feature vector and the second fractal feature vector, the fractal similarity to the optional tobacco leaf is determined.

3. The method for processing tobacco leaf spectral data according to claim 2, characterized in that, The step of determining the first fractal feature vector corresponding to the target tobacco leaf based on multiple first segmented spectra includes: For each of the first segmented spectra, determine the fluctuation difference value of the first segmented spectra under multiple different delay steps; A fitting operation is performed on multiple fluctuation difference values ​​to obtain fractal feature values ​​corresponding to the first segmented spectrum; Based on the fractal feature values ​​corresponding to multiple first segmented spectra, a first fractal feature vector corresponding to the target tobacco leaf is determined.

4. The method for processing tobacco leaf spectral data according to claim 3, characterized in that, Determining the fluctuation difference values ​​of the first segmented spectrum at multiple different delay steps includes: For each delay step, at least one pair of spectral points of the first segmented spectrum is determined at the delay step; wherein the pair of spectral points includes a first spectral point and a second spectral point, the second spectral point being located after the first spectral point and separated by the delay step; For at least one of the spectral point pairs, the square of the absorbance difference corresponding to the spectral point pair is determined based on the first absorbance value of the first spectral point and the second absorbance value of the second spectral point in the spectral point pair. The fluctuation difference value of the first segmented spectrum at the delay step is determined based on the square of the absorbance difference corresponding to at least one of the spectral point pairs.

5. The method for processing tobacco leaf spectral data according to claim 2, characterized in that, The step of determining the fractal similarity to the selectable tobacco leaf based on the first fractal feature vector and the second fractal feature vector includes: Based on the first fractal feature vector and the second fractal feature vector, determine the correlation coefficient and vector similarity between the first tobacco leaf spectral data and the second tobacco leaf spectral data; Based on the correlation coefficient and the vector similarity, the fractal similarity corresponding to the optional tobacco leaf is determined.

6. The method for processing tobacco leaf spectral data according to claim 5, characterized in that, The step of determining the fractal similarity to the selectable tobacco leaf based on the correlation coefficient and the vector similarity includes: The ratio between the correlation coefficient and the vector similarity is determined to obtain the fractal similarity corresponding to the optional tobacco leaf.

7. The method for processing tobacco leaf spectral data according to claim 1, characterized in that, The steps for acquiring the second tobacco leaf spectral data include: For at least one optional tobacco leaf, a sample is prepared from the optional tobacco leaf to obtain a tobacco leaf sample. The tobacco leaf sample is dried at a preset drying temperature and a preset drying time, and the dried tobacco leaf sample is ground into powder to obtain a tobacco leaf powder sample. The tobacco powder sample is scanned by infrared using a spectral acquisition device to collect spectral data in a preset spectral band, so as to obtain the second tobacco spectral data corresponding to the optional tobacco leaf.

8. A tobacco leaf spectral data processing device, characterized in that, include: The spectral data acquisition module is used to acquire the first tobacco leaf spectral data of the target tobacco leaf and the second tobacco leaf spectral data of at least one optional tobacco leaf. A similarity determination module is used to determine, for at least one of the optional tobacco leaves, a fractal similarity corresponding to the optional tobacco leaf based on the second tobacco leaf spectral data and the first tobacco leaf spectral data of the optional tobacco leaf, wherein the fractal similarity is used to indicate the degree of similarity of the spectral deep structure between the first tobacco leaf spectral data and the second tobacco leaf spectral data; A tobacco leaf determination module is used to determine alternative tobacco leaves for replacing the target tobacco leaf from at least one of the optional tobacco leaves based on at least one of the fractal similarities.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the tobacco leaf spectral data processing method as described in any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the tobacco leaf spectral data processing method according to any one of claims 1-7.

Images