Medical material processing and production digital visual supervision system and method based on cloud platform

Abstract

The invention relates to the technical field of digital supervision, in particular to a medical material processing and production digital visual supervision system and method based on a cloud platform. The method comprises the following steps: in a medicinal material processing process, collecting the nodule shape, size and distribution characteristics of a medicinal material rhizome to obtain medicinal material image information; identifying a medicinal material form change trend according to the medicinal material image information, and adjusting the slice thickness and the drying time of medicinal material processing equipment to obtain form characteristic supervision data; in the medicinal material processing process, the change range of the medicinal materials from faint yellow when the medicinal materials are fresh to yellowish brown after the medicinal materials are processed is collected, and spectral information of the medicinal materials is obtained. According to the method, the operation parameters of the processing equipment are precisely regulated and controlled by collecting the images, the spectrum and the mechanical response information of the medicinal materials and combining the environmental parameters, digital supervision of medicinal material processing is achieved, and therefore the medicinal material processing quality stability and the production efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of digital supervision technology, and in particular to a digital visual supervision system and method for medicinal material processing and production based on a cloud platform. Background Art

[0002] Polygonatum sibiricum is the dried rhizome of Polygonatum yunnanensis, Polygonatum sibiricum, or Polygonatum multiflorum, all plants in the Liliaceae family. Its rhizomes vary in morphology and are rich in starch and moisture. If not processed promptly, they are prone to mold. Traditional processing of Polygonatum sibiricum often results in varying degrees of drying due to inconsistent slice areas during slicing and drying, impacting the quality of the medicinal material. Specifically, Polygonatum sibiricum has complex morphological characteristics, with nodules on the rhizome exhibiting uneven shapes, sizes, and distribution. Existing image recognition technologies struggle to accurately identify these subtle features. The size and shape of individual nodules can change during processing, depending on operations such as slicing and drying. During processing, the morphology dynamically changes with each step (e.g., steaming, slicing, and drying). Existing digital visual monitoring systems lack the real-time monitoring capabilities to detect these dynamic changes, making it difficult to detect any anomalies during processing. Furthermore, Polygonatum sibiricum exhibits a wide range of color variations, ranging from light yellow when fresh to yellow-brown after processing. Furthermore, unstable lighting conditions in processing workshops can affect color recognition accuracy. Furthermore, Polygonatum sibiricum's texture characteristics (such as firmness and elasticity) require strict control during processing. The existing digital visual supervision system lacks direct means of texture detection and mainly relies on manual touch and experience judgment, which is inefficient and highly subjective. Summary of the Invention

[0003] Based on this, it is necessary to provide a digital visual supervision system and method for medicinal material processing and production based on a cloud platform to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a digital visual supervision method for medicinal material processing and production based on a cloud platform is provided, the method comprising the following steps: Step S1: During the medicinal material processing process, the shape, size, and distribution characteristics of the nodules of the medicinal material rhizome are collected to obtain medicinal material image information; based on the medicinal material image information, the morphological change trend of the medicinal material is identified, and the slice thickness and drying time of the medicinal material processing equipment are adjusted to obtain morphological characteristic supervision data; Step S2: During the medicinal material processing process, the color change range of the medicinal material from light yellow when fresh to yellow-brown after processing is collected to obtain medicinal material spectral information; the light intensity and ambient temperature of the processing workshop are monitored; the medicinal material spectral information is analyzed based on the light intensity and ambient temperature to determine the medicinal material color change trend, and the steaming time and steaming temperature of the medicinal material processing equipment are adjusted to obtain color characteristic supervision data; Step S3: During the medicinal material processing, the firmness and elasticity changes of the medicinal material are collected to obtain the mechanical response information of the medicinal material; by analyzing the mechanical response information, the texture change trend of the medicinal material during the processing is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic supervision data; Step S4: Upload the morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, intelligently adjust the operating parameters of the processing equipment to perform digital visual supervision operations for medicinal material processing and production.

[0005] Preferably, in step S1, during the medicinal material processing, the shape, size and distribution characteristics of the nodules of the medicinal material rhizomes are collected, including: Place the rhizome of the medicinal material on the collection platform, adjust the angle and position of the collection platform, and make sure the nodule part of the rhizome is completely within the collection field of view; Images were collected from the front, middle, and end of the rhizome of the medicinal material, and the focal length of the acquisition device was adjusted each time; The collected images are segmented and only the image information of the nodule part is retained; The segmented nodule images are normalized to adjust all nodule images to a uniform size ratio; Performing grayscale processing on the standardized nodule image to obtain a nodule grayscale image; The nodule grayscale image is subjected to contrast enhancement processing to highlight the outline of the nodule to obtain a nodule enhanced image; Delineate the nodule enhancement image and identify the nodule boundary contour by pixel-by-pixel scanning; The nodule boundary contours were quantitatively analyzed, and the area and perimeter of each nodule were calculated to determine the nodule size; Measure the aspect ratio and roundness of the nodule boundary outline to determine the shape characteristics of the nodule; The nodule distribution in the nodule-enhanced image was counted, the distance between adjacent nodules was measured, and the distribution density of nodules on the rhizome was calculated to determine the nodule distribution characteristics.

[0006] Preferably, in step S1, identifying the morphological change trend of the medicinal material according to the medicinal material image information and adjusting the slice thickness and drying time of the medicinal material processing equipment includes: The nodule shapes in the medicinal material images are detected frame by frame, and the complexity of the medicinal material shape is determined based on the tortuosity and number of branches of the nodule contour. The complexity of the medicinal material shape is correlated and compared with the processing time series to obtain the trend of nodule shape changes. The average area of ​​the nodule size in each frame of the image is calculated, and the change trend of the nodule size over the processing time series is determined based on the average area of ​​the nodule size to obtain the change trend of the nodule size; Count the nodule distribution density in each frame of the image; compare the nodule distribution density with the processing time series to obtain the nodule uniformity change trend; If the nodule shape change trend shows that the shape complexity increases over time, and the nodule size change trend shows that the average area increases over time, the slice thickness of the medicinal material processing equipment will be gradually reduced, and each adjustment amount is 5% of the initial slice thickness; If the nodule shape change trend shows that the shape complexity decreases over time, and the nodule size change trend shows that the average area decreases over time, the slice thickness of the medicinal material processing equipment is gradually increased, with each adjustment amount being 10% of the initial slice thickness; If the trend of nodule uniformity shows that the distribution density increases over time, the drying time of the medicinal material processing equipment will be gradually extended, with each adjustment amount being 20% ​​of the initial drying time; If the trend of nodule uniformity changes shows that the distribution density decreases over time, the drying time of the medicinal material processing equipment will be gradually shortened, with each adjustment amount being 25% of the initial drying time.

[0007] Preferably, in the step S2, during the medicinal material processing, the range of changes in the color of the collected medicinal materials from light yellow when fresh to yellow-brown after processing includes: S21: During the medicinal material processing, the top, middle and bottom of the medicinal material surface, as well as the edge of the medicinal material nodules are selected as selected areas; for each selected area, spectral reflectance data is collected respectively; S22: Perform multi-dimensional quantitative processing on the spectral reflectance data, calculate the ratio of the reflectance at each wavelength point to the reflectance of standard white, and obtain the color index of each segment; average the color index of each segment to obtain the average color index of the segment; calculate the standard deviation of the color index of each segment to obtain the color variation coefficient of the segment; S23: Divide the color change range of the medicinal material during processing into multiple segments according to the color variation coefficient, each segment corresponds to a specific interval of the medicinal material color change, and obtain the color segment of the medicinal material; among them, the first segment is the light yellow interval when the medicinal material is fresh, the second segment is the intermediate transition color interval during the medicinal material processing, and the third segment is the yellow-brown interval after the medicinal material processing is completed.

[0008] Preferably, in step S2, analyzing the medicinal material color change trend of the medicinal material spectrum information by light intensity and ambient temperature includes: According to the spectrum information of medicinal materials, the spectrum intensity, spectrum reflectance and absorption peak intensity of each segment corresponding to the specific wavelength range are extracted to obtain the spectrum characteristic parameters; Based on the influence of light intensity and ambient temperature on the spectral characteristic parameters, the change parameters of the spectral characteristic parameters of the medicinal materials in each segment are calculated respectively under different light intensities and ambient temperatures. The influence coefficients of light intensity and ambient temperature on the color change of the medicinal materials are determined and recorded as light intensity-temperature influence coefficients; The light intensity-temperature influence coefficient is used to detect the influence of changes in the color characteristics of the medicinal materials in the medicinal material spectrum information to obtain the color change trend of the medicinal materials.

[0009] Preferably, adjusting the steaming time and steaming temperature of the medicinal material processing equipment in step S3 includes: The color change rate of medicinal materials is obtained by calculating the difference between the color change trends of the medicinal materials; If the difference in the rate of change of the medicinal material color is greater than 10, it is determined that the medicinal material is in the stage of rapid color change; if the difference in the rate of change of the medicinal material color is less than 5, it is determined that the medicinal material is in the stage of short color change; Adjust the steaming time and temperature of the medicinal material processing equipment according to the color change stage of the medicinal material and the color segment of the medicinal material; When the medicinal material is in the first stage and the color changes briefly, extend the steaming time by 5 minutes and keep the steaming temperature unchanged; When the medicinal materials are in the second stage and the color changes rapidly, reduce the steaming temperature by 5°C and keep the steaming time unchanged; When the medicinal material is in the third section and the color change is close to the target value, the steaming time is shortened by 3 minutes and the steaming temperature is lowered by 3°C.

[0010] Preferably, in step S3, during the medicinal material processing, collecting the changes in firmness and elasticity of the medicinal material comprises the following steps: During the medicinal material processing, a pressure sensor is used to collect the firmness data of the medicinal material. The pressure sensor has a measurement range of 0-100 Newtons and an accuracy of 0.1 Newtons. At the same time, a displacement sensor is used to collect the elastic change data of the medicinal material. The displacement sensor has a measurement range of 0-10 mm and an accuracy of 0.01 mm. At the initial stage of medicinal material processing, the compression amount of the medicinal material when a pressure of 10 Newtons is applied is collected through a pressure sensor and recorded as the initial compression amount; at the same time, the rebound amount of the medicinal material after the pressure is released is recorded as the initial elastic response data; During the mid-stage of medicinal material processing, a 20 Newton pressure was repeatedly applied, and the compression and rebound amounts of the medicinal materials were collected and recorded as mid-stage compression and mid-stage elastic response data, respectively. The compression and rebound amounts in the initial stage and the mid-stage were compared, and the change rates of the compression and rebound amounts were calculated to obtain the mid-stage mechanical response data. In the later stage of medicinal material processing, a higher pressure of 30 Newtons is applied, and the compression and rebound amounts of the medicinal materials are collected and recorded as the late compression amount and late elastic response data, respectively; the compression and rebound amounts in the middle stage and the late stage are compared, and the change rate of the compression amount and the change rate of the rebound amount are further calculated to obtain the late mechanical response data.

[0011] Preferably, in step S3, identifying the texture change trend of the medicinal material during the processing by analyzing the mechanical response information and adjusting the pressing force parameters of the medicinal material processing equipment include: The average compression and average rebound of the medicinal materials were extracted from the mid-term mechanical response data as mid-term compression characteristics and mid-term rebound characteristics, respectively. The average compression and average rebound of the medicinal materials were extracted from the late mechanical response data as late compression characteristics and late rebound characteristics, respectively. The difference between the late compression characteristic and the mid-term compression characteristic is calculated and recorded as the compression difference; the difference between the late rebound characteristic and the mid-term rebound characteristic is calculated and recorded as the rebound difference; If the compression difference is less than -10% and the rebound difference is less than -10%, the drug material texture is determined to be hard and the elasticity is weakened; if the compression difference is greater than 10% and the rebound difference is greater than 10%, the drug material texture is determined to be soft and the elasticity is enhanced; if the compression difference is between -10% and 10% and the rebound difference is between -10% and 10%, the drug material texture is determined to be stable; When the texture of the medicinal material becomes harder and its elasticity weakens, the pressing force parameters of the medicinal material processing equipment will be reduced by 20%; when the texture of the medicinal material becomes softer and its elasticity increases, the pressing force parameters of the medicinal material processing equipment will be increased by 20%; when the texture of the medicinal material is stable, the pressing force parameters of the medicinal material processing equipment will be kept unchanged.

[0012] Preferably, uploading the morphological feature supervision data, the color feature supervision data, and the texture feature supervision data to the cloud platform in step S4 includes: Convert morphological feature supervision data into morphological visualization images and number them according to time series; Convert the color feature supervision data into a structured data table of RGB values, record the color data collected each time and its corresponding timestamp, and draw a curve chart of the RGB value change of the color feature; where the timestamp is the horizontal axis and the RGB value is the vertical axis; The texture feature monitoring data is converted into a numerical data table, and the compression force and deformation data collected each time and their corresponding timestamps are recorded. A compression force-deformation relationship diagram of the texture feature is drawn, with the compression force as the horizontal axis and the deformation as the vertical axis. The morphological visualization image, RGB value change curve and compression force-deformation relationship diagram are encrypted and uploaded to the cloud platform for storage to form full-link supervision data.

[0013] The technical beneficial effects of the present invention are as follows: by collecting the shape, size and distribution characteristics of the nodules of the rhizomes of the medicinal materials, the medicinal materials image information is obtained, and then the trend of the morphological changes of the medicinal materials is identified according to the medicinal materials image information, and the slice thickness and drying time of the medicinal materials processing equipment are adjusted accordingly to obtain the morphological characteristic supervision data. This process can achieve precise control of the morphological changes in the medicinal materials processing process. Specifically, based on the recognition and analysis of the medicinal materials image information, the adjustment of the slice thickness and drying time can be made to be more in line with the actual morphological characteristics of the medicinal materials, avoiding the problems of unstable medicinal materials processing quality due to excessive or too small slice thickness, and the decline in medicinal materials quality due to insufficient or too long drying time, thereby effectively improving the morphological consistency and quality stability of medicinal materials processing, laying a good foundation for subsequent processing links, and ensuring the accuracy and reliability of the medicinal materials processing process in terms of morphological control. During the medicinal material processing process, spectral information is collected, measuring the range of the medicinal material's color from pale yellow when fresh to yellow-brown after processing. Simultaneously, the light intensity and ambient temperature in the processing workshop are monitored. The spectral information is used to analyze the color change trends of the medicinal materials. This information is then used to adjust the steaming time and temperature of the medicinal material processing equipment, generating color characteristic monitoring data. This process enables precise monitoring and control of medicinal material color changes. By collecting spectral information and combining it with comprehensive analysis of processing environment parameters, the color change trends of the medicinal materials can be accurately identified and adjusted accordingly. This effectively prevents excessive color deviation caused by improper steaming parameters, ensures that the color of the processed medicinal materials meets quality standards, improves the stability of the processed medicinal material's appearance quality, and provides a scientific basis and precise control method for color quality control during the medicinal material processing process. During the medicinal material processing process, the firmness and elasticity of the medicinal materials are collected to obtain mechanical response information. This mechanical response information is analyzed to identify the changing trends of the medicinal material's texture during processing. The pressing force parameters of the medicinal material processing equipment are adjusted accordingly to generate texture characteristic monitoring data. This step can accurately monitor and control changes in the texture of medicinal materials. Based on the analysis of the mechanical response information of medicinal materials, the texture change trend of medicinal materials can be accurately identified, and then the pressing force parameters can be reasonably adjusted to effectively avoid problems such as medicinal materials being too hard or too soft due to unreasonable pressing force, ensuring that the texture of the medicinal materials after processing meets the quality requirements, ensuring the stability of the medicinal ingredients of the medicinal materials and the safety of use, providing strong guarantees for the texture quality control during the medicinal materials processing process, and improving the overall quality control level of medicinal materials processing. The morphological feature supervision data, color feature supervision data, and texture feature supervision data are uploaded to the cloud platform to form full-link supervision data, and the operating parameters of the processing equipment are intelligently adjusted based on the full-link supervision data to perform digital visual supervision operations for medicinal materials processing and production. This process realizes comprehensive, real-time, and precise supervision of the medicinal materials processing process. By integrating and uploading multi-dimensional supervision data to the cloud platform and building a full-link supervision data system, digital and visual management of the entire process of medicinal materials processing can be achieved.Intelligent regulation based on full-link supervision data can dynamically adjust the operating parameters of processing equipment according to real-time data feedback, ensuring that the medicinal materials processing process is always in the best state, further improving the stability and consistency of medicinal materials processing quality, reducing production costs, and improving production efficiency. At the same time, it is convenient for regulatory authorities to trace and supervise the medicinal materials processing process, ensure the quality and safety of medicinal materials products, and promote the digital and intelligent development of the medicinal materials processing industry.

[0014] In this specification, a digital visual supervision system for medicinal material processing and production based on a cloud platform is provided, which is used to implement the above-mentioned digital visual supervision method for medicinal material processing and production based on a cloud platform. The digital visual supervision system for medicinal material processing and production based on a cloud platform includes: The morphological feature monitoring module is used to collect the shape, size, and distribution characteristics of the nodules of the medicinal material rhizomes during the medicinal material processing process to obtain medicinal material image information; based on the medicinal material image information, it identifies the trend of medicinal material morphological changes and adjusts the slice thickness and drying time of the medicinal material processing equipment to obtain morphological feature monitoring data; The color feature monitoring module is used to collect the color variation range of the medicinal materials from light yellow when fresh to yellow-brown after processing during the medicinal material processing process to obtain the medicinal material spectral information; monitor the light intensity and ambient temperature of the processing workshop; analyze the medicinal material color variation trend based on the light intensity and ambient temperature of the medicinal material spectral information, and adjust the steaming time and steaming temperature of the medicinal material processing equipment to obtain color feature monitoring data; The texture characteristic monitoring module is used to collect changes in the firmness and elasticity of medicinal materials during the processing of medicinal materials to obtain the mechanical response information of the medicinal materials. By analyzing the mechanical response information, the changing trend of the medicinal material texture during the processing of the medicinal materials is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic monitoring data. The intelligent control module for processing equipment is used to upload morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, the operating parameters of the processing equipment are intelligently controlled to perform digital visual supervision operations for medicinal material processing and production.

[0015] The system of the present invention uses a morphological feature supervision module to accurately collect medicinal material image information and adjust the slice thickness and drying time to generate morphological feature supervision data; a color feature supervision module collects medicinal material spectral information and controls the steaming time and temperature in combination with environmental parameters to generate color feature supervision data; a texture feature supervision module collects medicinal material mechanical response information and adjusts the pressing force parameters to generate texture feature supervision data. These data are uploaded to the cloud platform to form full-link supervision data, and the processing equipment intelligent control module intelligently controls the processing equipment operating parameters based on this data to achieve digital and visual supervision of medicinal material processing. The system can comprehensively and accurately control the changes in morphology, color and texture during the processing of medicinal materials, ensure stable processing quality, reduce production costs, improve production efficiency, and facilitate supervision and traceability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the steps of a digital visual supervision method for medicinal material processing and production based on a cloud platform; Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG. The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0017] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0018] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0019] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0020] To achieve this, please refer to Figures 1 to 2 A digital visual supervision method for medicinal material processing and production based on a cloud platform, the method comprising the following steps: Step S1: During the medicinal material processing process, the shape, size, and distribution characteristics of the nodules of the medicinal material rhizome are collected to obtain medicinal material image information; based on the medicinal material image information, the morphological change trend of the medicinal material is identified, and the slice thickness and drying time of the medicinal material processing equipment are adjusted to obtain morphological characteristic supervision data; In an embodiment of the present invention, during the medicinal material processing process, a high-resolution industrial camera is first used to capture image information of the medicinal material rhizomes. This industrial camera has an imaging capability of at least 12 megapixels and is equipped with a macro lens, enabling clear capture of the shape, size, and distribution characteristics of the nodules on the medicinal material rhizomes. The camera's shooting parameters are set to: exposure time of 1 / 200 second, aperture value of F4, and ISO value of 200 to ensure image clarity and detail. The captured medicinal material images are stored in a local server's image database in JPEG format. Each image file is named "Medicinal Material Image_X," where X is the acquisition sequence number. Subsequently, the captured medicinal material images are transmitted to a deep learning-based image recognition system. This system utilizes a pretrained convolutional neural network (CNN) architecture to identify the morphological characteristics of medicinal material rhizomes through transfer learning. The system performs image preprocessing, including grayscale conversion, binarization, and edge detection, to highlight the shape and distribution of nodules. The recognition algorithm analyzes parameters such as nodule outline, area, and spacing in the image to determine the changing trends of the medicinal material morphology. For example, when the spacing between nodules is small, the system determines the herb's morphology to be dense; when the nodule area is large, the system determines the herb's morphology to be coarse. Recognition results are output as structured data, including information such as nodule shape (round, oval, irregular), size (diameter range), and distribution density (number of nodules per unit area). This data is stored in the cloud platform's morphological feature database. The data table is named "Herbal Morphological Feature Table," and the fields include "Nodule Shape," "Nodule Size," "Distribution Density," and "Collection Time." Based on the herb's morphological feature recognition results, the cloud platform's automated control system adjusts the parameters of the herb processing equipment. For herb slicing equipment, the slice thickness is adjusted based on nodule size. For nodule diameters less than 5 mm, the slice thickness is set to 1 mm; for nodule diameters between 5 and 10 mm, the slice thickness is set to 2 mm; and for nodule diameters greater than 10 mm, the slice thickness is set to 3 mm. The blade speed of the slicing equipment is fixed at 3000 rpm, and the feed speed is dynamically adjusted according to the slice thickness. When the slice thickness is 1 mm, the feed speed is 10 cm / min, when the slice thickness is 2 mm, the feed speed is 8 cm / min, and when the slice thickness is 3 mm, the feed speed is 6 cm / min. For medicinal material drying equipment, the drying time is adjusted according to the density of the medicinal material morphology. When the nodule distribution density is greater than 10 per square centimeter, the drying time is set to 4 hours; when the nodule distribution density is between 5 and 10 per square centimeter, the drying time is set to 3 hours; when the nodule distribution density is less than 5 per square centimeter, the drying time is set to 2 hours. The temperature of the drying equipment is fixed at 60 degrees Celsius, and the wind speed is 2 meters per second.The adjusted equipment parameters are transmitted to the medicinal material processing equipment's control unit via the Industrial Internet of Things (IIoT) protocol and recorded in real time in the cloud platform's equipment parameter database. This data table, named "Equipment Parameter Adjustment Record Table," includes fields such as "Equipment Type," "Slice Thickness," "Feed Speed," "Drying Time," and "Adjustment Time." Finally, the operating parameters of the medicinal material processing equipment and the results of the medicinal material morphological feature recognition are integrated into morphological feature supervision data and stored in the cloud platform's supervision data warehouse. This data table, named "Medicinal Material Processing Morphological Feature Supervision Table," includes fields such as "Medicinal Material Image Number," "Nodule Shape," "Nodule Size," "Distribution Density," "Slice Thickness," "Feed Speed," "Drying Time," and "Supervision Time." Through the cloud platform's visual interface, users can view real-time changes in morphological features and equipment parameter adjustments during the medicinal material processing process, enabling digital, visual supervision of medicinal material processing and production.

[0021] Step S2: During the medicinal material processing process, the color change range of the medicinal material from light yellow when fresh to yellow-brown after processing is collected to obtain medicinal material spectral information; the light intensity and ambient temperature of the processing workshop are monitored; the medicinal material spectral information is analyzed based on the light intensity and ambient temperature to determine the medicinal material color change trend, and the steaming time and steaming temperature of the medicinal material processing equipment are adjusted to obtain color characteristic supervision data; In this embodiment of the present invention, during the medicinal material processing process, a fiber optic spectrometer is first used to collect spectral information on the color changes of the medicinal materials during processing. The spectrometer's wavelength range covers the visible light region (400 to 700 nanometers) with a resolution of 1 nanometer. During the acquisition process, the spectrometer scans the medicinal materials' spectrum every 30 seconds, monitoring the color change from the fresh pale yellow (wavelength of approximately 580 nanometers) to the processed yellow-brown (wavelength of approximately 590 to 610 nanometers). The collected spectral data is stored in a local data storage unit in CSV format, with the file name "Medicinal Material Spectral Data_X," where X is the acquisition batch number. Simultaneously, an environmental monitoring sensor is used to collect the light intensity and ambient temperature of the processing workshop. The light intensity sensor, based on the photodiode principle, has a measurement range of 0 to 100,000 lux and an accuracy of ±5%. The ambient temperature sensor, a thermistor-type temperature sensor, has a measurement range of 0°C to 50°C and an accuracy of ±0.5°C. Light intensity sensors and ambient temperature sensors collect data every minute and transmit it to the cloud platform's environmental monitoring database via wireless communication. The data table, named "Processing Workshop Environmental Data Table," includes fields such as "Light Intensity," "Ambient Temperature," and "Collection Time." The cloud platform's data analysis module then correlates the medicinal material's spectral information with the environmental data. The analysis module determines the rate and direction of the medicinal material's color change based on the range of light intensity and ambient temperature, combined with the changing trends in the medicinal material's spectral data. When light intensity is between 30,000 and 50,000 lux and the ambient temperature is between 25°C and 30°C, the medicinal material's color transition from light yellow to yellow-brown is rapid. When light intensity is below 20,000 lux or the ambient temperature is below 20°C, the color change is slower. Based on the analysis results, the system automatically adjusts the steaming time and temperature of the medicinal material processing equipment. When the medicinal material's color changes rapidly, the steaming time is shortened to 45 minutes, and the steaming temperature is set to 95°C. When the medicinal material's color changes slowly, the steaming time is extended to 60 minutes, and the steaming temperature is set to 100°C. The adjusted equipment parameters are updated in real time to the medicinal material processing equipment via the industrial automation control system and simultaneously recorded in the cloud platform's color feature monitoring database. The data table is named "Medicinal Material Color Feature Monitoring Table" and includes fields such as "Medicinal Material Spectral Data Number," "Light Intensity," "Ambient Temperature," "Steaming Time," "Steaming Temperature," and "Monitoring Time."

[0022] Step S3: During the medicinal material processing, the firmness and elasticity changes of the medicinal material are collected to obtain the mechanical response information of the medicinal material; by analyzing the mechanical response information, the texture change trend of the medicinal material during the processing is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic supervision data; In one embodiment of the present invention, a dynamic mechanical analyzer (DMA) is used to collect information on changes in the firmness and elasticity of medicinal materials during processing. This analyzer applies cyclic deformation stress and measures the strain response to the medicinal materials, thereby obtaining the mechanical response characteristics of the medicinal materials. Specifically, the medicinal material sample is secured in the DMA fixture, and the test mode is set to dynamic frequency sweep, with a frequency range of 0.1 Hz to 10 Hz and a strain amplitude of 0.1% to 1%. The temperature is maintained at a constant temperature of 25°C. During processing, the medicinal material samples are mechanically tested every 10 minutes, and changes in the modulus (E) and loss factor (tan δ) are recorded. The modulus E reflects the firmness of the medicinal material, with higher values ​​indicating stronger materials. The loss factor (tan δ) reflects the elasticity of the medicinal material, with higher values ​​indicating less elastic materials. The collected mechanical response data is stored in XML format in the mechanical data storage module of the cloud platform. The data file is named "Medicinal Material Mechanical Response Data_X," where X is the test batch number. The collected mechanical response information is then processed by the cloud platform's data analysis module. The analysis module identifies changes in the medicinal material's texture during processing based on the changing trends of its modulus E and loss factor tanδ. When the modulus E gradually decreases from an initial value of 100 MPa to 80 MPa and the loss factor tanδ increases from 0.05 to 0.1, the medicinal material's texture transitions from firm to soft. When the modulus E remains above 120 MPa and the loss factor tanδ is below 0.03, the medicinal material's texture remains firm. Based on these texture trends, the system automatically adjusts the pressing force parameters of the medicinal material processing equipment. When the medicinal material's texture becomes softer, the pressing force is reduced to 100 Newtons; when the medicinal material's texture remains firm, the pressing force is maintained at 200 Newtons. The adjusted pressing force parameters are transmitted in real time to the pressing device of the medicinal material processing equipment via the industrial automation control system and simultaneously recorded in the texture characteristic monitoring database on the cloud platform. The data table, named "Medicinal Material Texture Characteristic Monitoring Table," contains fields such as "Mechanical Response Data Number," "Modulus E," "Loss Factor tanδ," "Pressing Force," and "Monitoring Time."

[0023] Step S4: Upload the morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, intelligently adjust the operating parameters of the processing equipment to perform digital visual supervision operations for medicinal material processing and production.

[0024] In this embodiment of the present invention, morphological, color, and texture characteristic monitoring data are first uploaded in batches from a local data storage unit to a cloud platform via an Industrial Internet of Things (IIoT) gateway. This data is organized in a structured JSON format and includes fields such as "herbal medicine image number," "nodule shape," "nodule size," "distribution density," "slice thickness," "feed rate," "drying time," "herbal medicine spectral data number," "light intensity," "ambient temperature," "steaming time," "steaming temperature," "mechanical response data number," "modulus E," "loss factor tan δ," and "pressing force." Data is uploaded every 30 minutes and encrypted using the HTTPS protocol to ensure data security and integrity. The uploaded data is stored in the cloud platform's full-link monitoring database in a table named "Full-Link Monitoring Data Table." The data fields include all of the aforementioned monitoring data fields, as well as "upload time" and "data integrity check code." The cloud platform's intelligent control module then performs a comprehensive analysis of the full-link monitoring data. Based on a pre-set rule engine, this module dynamically adjusts the operating parameters of processing equipment, combining herbal medicine processing requirements and historical data. For example, if morphological data indicates dense nodules and thin slices, combined with color data indicating a long steaming time, the intelligent control module will appropriately increase the drying time of the processing equipment to ensure uniform drying. If texture data indicates a soft texture and a modulus E below a set threshold, the intelligent control module will reduce the pressing force of the processing equipment to prevent excessive compaction and quality issues. The adjusted operating parameters are sent to the processing equipment's control system via the cloud platform's device command distribution module using the OPC UA (Open Platform Communications Unified Architecture) protocol, ensuring that the equipment operates according to the optimized parameters. The adjusted parameters are also stored in the "Equipment Operating Parameter Adjustment Record Table" of the full-chain monitoring database. Fields include "Device Type," "Adjustment Parameter Name," "Parameter Value Before Adjustment," "Parameter Value After Adjustment," "Adjustment Time," and "Basis for Adjustment." Finally, the cloud platform's visual interface displays full-chain monitoring data and the operating status of the processing equipment in real time, providing comprehensive production process monitoring and decision support for production managers, enabling digital, visual oversight of medicinal material processing.

[0025] Preferably, in step S1, during the medicinal material processing, the shape, size and distribution characteristics of the nodules of the medicinal material rhizomes are collected, including: Place the rhizome of the medicinal material on the collection platform, adjust the angle and position of the collection platform, and make sure the nodule part of the rhizome is completely within the collection field of view; Images were collected from the front, middle, and end of the rhizome of the medicinal material, and the focal length of the acquisition device was adjusted each time; The collected images are segmented and only the image information of the nodule part is retained; The segmented nodule images are normalized to adjust all nodule images to a uniform size ratio; Grayscale processing is performed on the standardized nodule image to obtain a nodule grayscale image; The nodule grayscale image is subjected to contrast enhancement processing to highlight the outline of the nodule to obtain a nodule enhanced image; Delineate the nodule enhancement image and identify the nodule boundary contour by pixel-by-pixel scanning; The nodule boundary contours were quantitatively analyzed, and the area and perimeter of each nodule were calculated to determine the nodule size; Measure the aspect ratio and roundness of the nodule boundary outline to determine the shape characteristics of the nodule; The nodule distribution in the nodule-enhanced image was counted, the distance between adjacent nodules was measured, and the distribution density of nodules on the rhizome was calculated to determine the nodule distribution characteristics.

[0026] In this embodiment of the present invention, the medicinal herb rhizomes are placed on a high-precision, electrically adjustable acquisition platform with multi-dimensional angle adjustment. By controlling the platform's tilt angle and horizontal position, the rhizome's nodular portion is completely within the acquisition field of view. The platform's positioning sensor precisely adjusts the platform's position to within a ±0.1 mm error range. A high-resolution industrial camera is used as the acquisition device to capture images from the front, middle, and end of the rhizomes. Before each acquisition, the camera's focal length is adjusted using an autofocus system to ensure that the captured image clarity meets the 20 / 20 pixel resolution standard. The camera's aperture is set to F8, the shutter speed to 1 / 250 second, and the ISO value to 200 to ensure image quality. The captured raw images are transmitted to an image processing server and segmented using a deep learning-based image segmentation algorithm. A pretrained segmentation model identifies and retains only the image information of the rhizome nodular portion, while removing background and other non-nodular areas. The confidence threshold of the segmentation algorithm is set to 0.95 to ensure segmentation accuracy. The segmented nodule images were normalized using a bilinear interpolation algorithm to resize all nodule images to a uniform size of 256 × 256 pixels. The images were scaled by calculating the ratio factor between the original and target sizes to ensure that the geometric features of the nodule images remained consistent during subsequent processing. The normalized nodule images were grayscaled using the RGB-to-grayscale formula Y = 0.299R + 0.587G + 0.114B to obtain the grayscale images. The grayscaled images were converted to pixel values ​​ranging from 0 to 255 to reduce image data size and highlight structural features. The grayscaled images were contrast-enhanced using a histogram equalization algorithm to adjust the image histogram distribution and highlight the nodule outline. The contrast-enhanced images achieved a uniform histogram distribution with a contrast enhancement factor of 1.5, ensuring clear nodule outlines. The contrast-enhanced nodule images were then contoured, and the Canny edge detection algorithm was used to extract the nodule boundary. By scanning pixel by pixel, the pixel coordinates of the nodule boundary contour are identified, and these coordinates are stored as a two-dimensional array. The nodule boundary contour is quantitatively analyzed, and the size and shape characteristics of the nodule are determined by calculating the geometric parameters of the contour. The specific operation is: using the coordinates of the contour pixel points, the area and perimeter of each nodule are calculated. The area is calculated using the polygon area formula, and the perimeter is calculated by accumulating the Euclidean distance of the contour pixel points. The calculation accuracy of the area and perimeter reaches three decimal places. The aspect ratio and roundness of the nodule boundary contour are measured to determine the shape characteristics of the nodule. The aspect ratio is obtained by calculating the aspect ratio of the minimum circumscribed rectangle of the nodule contour; the nodule distribution of the nodule enhanced image is counted, and the distance between adjacent nodules is measured by the identified nodule boundary contour coordinates.The distance between adjacent nodule centers was calculated using the Euclidean distance formula, with a measurement accuracy of 0.01 mm. The distribution density of nodules on the rhizome was also calculated by counting the number of nodules per unit length of rhizome. The data on nodule size, shape, and distribution characteristics obtained from this analysis were stored in a cloud platform database in a structured format, including fields such as nodule number, area, perimeter, aspect ratio, circularity, distance between adjacent nodules, and distribution density.

[0027] Preferably, in step S1, identifying the morphological change trend of the medicinal material according to the medicinal material image information and adjusting the slice thickness and drying time of the medicinal material processing equipment includes: The nodule shapes in the medicinal material images are detected frame by frame, and the complexity of the medicinal material shape is determined based on the tortuosity and number of branches of the nodule contour. The complexity of the medicinal material shape is correlated and compared with the processing time series to obtain the trend of nodule shape changes. The average area of ​​the nodule size in each frame of the image is calculated, and the change trend of the nodule size over the processing time series is determined based on the average area of ​​the nodule size to obtain the change trend of the nodule size; Count the nodule distribution density in each frame of the image; compare the nodule distribution density with the processing time series to obtain the nodule uniformity change trend; If the nodule shape change trend shows that the shape complexity increases over time, and the nodule size change trend shows that the average area increases over time, the slice thickness of the medicinal material processing equipment will be gradually reduced, and each adjustment amount is 5% of the initial slice thickness; If the nodule shape change trend shows that the shape complexity decreases over time, and the nodule size change trend shows that the average area decreases over time, the slice thickness of the medicinal material processing equipment is gradually increased, with each adjustment amount being 10% of the initial slice thickness; If the trend of nodule uniformity shows that the distribution density increases over time, the drying time of the medicinal material processing equipment will be gradually extended, with each adjustment amount being 20% ​​of the initial drying time; If the trend of nodule uniformity changes shows that the distribution density decreases over time, the drying time of the medicinal material processing equipment will be gradually shortened, with each adjustment amount being 25% of the initial drying time.

[0028] In an embodiment of the present invention, images are extracted frame by frame from a video stream of the medicinal material processing process. The nodule contours in each frame are extracted using an edge detection algorithm (such as the Canny algorithm). The degree of tortuosity of the nodule contour is quantified by calculating the rate of change of the contour curvature. The curvature rate is obtained by solving the derivative of the contour curvature with respect to the arc length. Simultaneously, the number of branches in the nodule contour is counted using a connectivity analysis algorithm. The shape complexity of the medicinal material is calculated based on the tortuosity and number of branches of the nodule contour. The complexity is determined by multiplying the tortuosity by a weighting factor of 0.6 and adding the number of branches multiplied by a weighting factor of 0.4. The complexity value for each frame is stored in a cloud platform database. The shape complexity of the medicinal material is correlated and compared with the processing time series, and the trend of complexity change over processing time is calculated using a time series analysis algorithm (such as the sliding window averaging method). If the complexity increases over time, it is recorded as a positive trend; if the complexity decreases over time, it is recorded as a negative trend. After contour extraction of nodules in each image frame, the area of ​​each nodule was calculated using the polygon area formula, and the average area of ​​nodule sizes in each image frame was calculated. The average area was determined by summing the areas of all nodules and dividing by the number of nodules. The trend of nodule size variation over the processing time sequence was determined based on the average area. A time series analysis algorithm was used to calculate the trend of the average area over processing time; if the average area increased over time, it was recorded as a positive trend; if the average area decreased over time, it was recorded as a negative trend. The distribution density of nodules in each image frame was calculated by calculating the number of nodules per unit area. The distribution density was determined by dividing the number of nodules by the image area. The nodule distribution density was correlated with the processing time sequence, and the trend of the distribution density over processing time was calculated using a time series analysis algorithm. If the distribution density increased over time, it was recorded as a positive trend; if the distribution density decreased over time, it was recorded as a negative trend. If the nodule shape change trend shows an increase in shape complexity over time (a positive trend), and the nodule size change trend shows an increase in average area over time (a positive trend), instructions are sent to the medicinal material processing equipment via the cloud platform control interface to gradually reduce the slice thickness. Each adjustment is 5% of the initial slice thickness. The adjusted slice thickness is the initial slice thickness multiplied by 0.95. If the nodule shape change trend shows a decrease in shape complexity over time (a negative trend), and the nodule size change trend shows a decrease in average area over time (a negative trend), instructions are sent to the medicinal material processing equipment via the cloud platform control interface to gradually increase the slice thickness. Each adjustment is 10% of the initial slice thickness. The adjusted slice thickness is the initial slice thickness multiplied by 1.10. If the nodule uniformity trend shows an increase in distribution density over time (a positive trend), instructions are sent to the medicinal material processing equipment via the cloud platform control interface to gradually increase the drying time. Each adjustment is 20% of the initial drying time. The adjusted drying time is the initial drying time multiplied by 1.20.If the nodule uniformity trend shows a decrease in distribution density over time (a negative trend), instructions are sent to the medicinal material processing equipment through the cloud platform control interface to gradually shorten the drying time. Each adjustment is 25% of the initial drying time. The adjusted drying time is the initial drying time multiplied by 0.75.

[0029] As an example of the present invention, refer to Figure 2 As shown, in step S2, during the medicinal material processing, the range of changes in the collected medicinal materials from light yellow when fresh to yellow-brown after processing includes: S21: During the medicinal material processing, the top, middle and bottom of the medicinal material surface, as well as the edge of the medicinal material nodules are selected as selected areas; for each selected area, spectral reflectance data is collected respectively; S22: Perform multi-dimensional quantitative processing on the spectral reflectance data, calculate the ratio of the reflectance at each wavelength point to the reflectance of standard white, and obtain the color index of each segment; average the color index of each segment to obtain the average color index of the segment; calculate the standard deviation of the color index of each segment to obtain the color variation coefficient of the segment; S23: Divide the color change range of the medicinal material during processing into multiple segments according to the color variation coefficient, each segment corresponds to a specific interval of the medicinal material color change, and obtain the color segment of the medicinal material; among them, the first segment is the light yellow interval when the medicinal material is fresh, the second segment is the intermediate transition color interval during the medicinal material processing, and the third segment is the yellow-brown interval after the medicinal material processing is completed.

[0030] In this embodiment of the present invention, during the medicinal material processing process, the top, middle, and bottom portions of the medicinal material surface, as well as the edges of the medicinal material nodules, are selected as selected areas. Spectral reflectance data for each selected area is collected using a high-precision spectrometer. The spectrometer's wavelength range is set to 400-700 nm, covering the visible light region, and the wavelength resolution is set to 1 nm. During the acquisition process, the spectrometer's integration time is set to 100 ms to ensure data stability and accuracy. The collected spectral reflectance data undergoes multi-dimensional quantification. First, the ratio of the reflectance at each wavelength point to the reflectance of standard white is calculated to obtain the color index of each segment. The standard white reflectance is obtained by measuring a standard white calibration plate under the same conditions, with a reflectance value of 95%. The color index is calculated as follows: Color Index = (Reflectance of Selected Area / Reflectance of Standard White) × 100. Next, the color index of each segment is averaged to obtain the average color index for that segment. The average color index is obtained by taking the arithmetic mean of the color indices for all wavelength points within each segment. Finally, the standard deviation of the color index of each segment is calculated to obtain the color variation coefficient of that segment. The color variation coefficient is determined by calculating the ratio of the standard deviation of the color index within each segment to the average color index and is used to measure the color uniformity of that segment. Based on the color variation coefficient, the color variation range of the medicinal material during processing is divided into multiple segments, each corresponding to a specific range of the medicinal material's color variation. The specific division criteria are as follows: the first segment is the light yellow range of the fresh medicinal material, with a color variation coefficient of less than 0.1 and an average color index between 60-70; the second segment is the intermediate transition color range during the medicinal material processing, with a color variation coefficient between 0.1 and 0.3 and an average color index between 40-60; the third segment is the yellow-brown range after the medicinal material is processed, with a color variation coefficient greater than 0.3 and an average color index less than 40. Using these division criteria, the color segment of the medicinal material is determined, thereby achieving real-time monitoring and accurate assessment of color changes during the medicinal material processing.

[0031] Preferably, in step S2, analyzing the medicinal material color change trend of the medicinal material spectrum information by light intensity and ambient temperature includes: According to the spectrum information of medicinal materials, the spectrum intensity, spectrum reflectance and absorption peak intensity of each segment corresponding to the specific wavelength range are extracted to obtain the spectrum characteristic parameters; Based on the influence of light intensity and ambient temperature on the spectral characteristic parameters, the change parameters of the spectral characteristic parameters of the medicinal materials in each segment are calculated respectively under different light intensities and ambient temperatures. The influence coefficients of light intensity and ambient temperature on the color change of the medicinal materials are determined and recorded as light intensity-temperature influence coefficients; The light intensity-temperature influence coefficient is used to detect the influence of changes in the color characteristics of the medicinal materials in the medicinal material spectrum information to obtain the color change trend of the medicinal materials.

[0032] In this embodiment of the present invention, a high-precision spectrometer is used to collect spectral information from medicinal materials. The spectrometer's wavelength range is set to 400-700 nm, covering the visible light region, with a wavelength resolution of 1 nm. During the acquisition process, the spectrometer's integration time is set to 100 ms. From the collected spectral information, the spectral intensity, spectral reflectance, and absorption peak intensity within specific wavelength intervals are extracted for each segment. Spectral intensity is obtained by measuring the light intensity at each wavelength. Spectral reflectance is calculated by calculating the ratio of reflected light to incident light on the medicinal material surface using the formula: reflectance = (reflected light intensity / incident light intensity) × 100%. Absorption peak intensity is obtained by identifying the absorption peak position in the spectral curve and measuring the corresponding light intensity. The specific wavelength intervals selected are 450-500 nm (blue region) and 600-700 nm (red region), which are more sensitive to medicinal material color changes. During the medicinal material processing process, light intensity and ambient temperature are monitored using a light sensor and a temperature sensor, respectively. The light sensor has a measurement range of 0-100,000 lux with an accuracy of ±1%. The temperature sensor has a measurement range of -20°C to 60°C with an accuracy of ±0.5°C. The change parameters of the spectral characteristic parameters of the medicinal material in each segment are calculated under different light intensities and ambient temperatures. The change parameters are determined by comparing the difference between the current and initial measurements. For example, the change parameter for spectral intensity is calculated as: spectral intensity change parameter = current spectral intensity - initial spectral intensity; the change parameter for spectral reflectance is calculated as: spectral reflectance change parameter = current spectral reflectance - initial spectral reflectance; and the change parameter for absorption peak intensity is calculated as: absorption peak intensity change parameter = current absorption peak intensity - initial absorption peak intensity. Based on the change parameters, the influence coefficients of light intensity and ambient temperature on the medicinal material color change are determined. The influence coefficients are calculated by comparing the change parameter with the change in the corresponding environmental parameter (light intensity or temperature). For example, the formula for calculating the light intensity influence coefficient is: Light intensity influence coefficient = (spectral characteristic parameter change parameter / light intensity change); the formula for calculating the ambient temperature influence coefficient is: Ambient temperature influence coefficient = (spectral characteristic parameter change parameter / temperature change). The calculated influence coefficient is recorded as the light intensity-temperature influence coefficient and stored in the cloud platform database. The light intensity-temperature influence coefficient is used to detect the impact of changes in the color characteristics of the medicinal material in the medicinal material spectral information. Based on the light intensity-temperature influence coefficient, the change pattern of the medicinal material color characteristics under different light intensities and ambient temperatures is analyzed. For example, if the light intensity influence coefficient is large and positive, it means that an increase in light intensity will cause the medicinal material color characteristics to change in a certain direction; if the ambient temperature influence coefficient is large and negative, it means that a decrease in temperature will cause the medicinal material color characteristics to change in another direction. Based on the change impact detection results, the trend of medicinal material color change is determined.Color trends are determined by analyzing the relationship between the light intensity-temperature influence coefficient and the medicinal material's color characteristics. For example, if light intensity gradually increases during processing and the light intensity influence coefficient is positive, while the ambient temperature remains stable, the medicinal material's color will gradually change in a specific direction, such as from light yellow to yellow-brown.

[0033] Preferably, adjusting the steaming time and steaming temperature of the medicinal material processing equipment in step S3 includes: The color change rate of medicinal materials is obtained by calculating the difference between the color change trends of the medicinal materials; If the difference in the rate of change of the medicinal material's color is greater than 10, it is determined that the medicinal material is in the stage of rapid color change; if the difference in the rate of change of the medicinal material's color is less than 5, it is determined that the medicinal material is in the stage of short color change; Adjust the steaming time and temperature of the medicinal material processing equipment according to the color change stage of the medicinal material and the color segment of the medicinal material; When the medicinal material is in the first stage and the color changes briefly, extend the steaming time by 5 minutes and keep the steaming temperature unchanged; When the medicinal materials are in the second stage and the color changes rapidly, reduce the steaming temperature by 5°C and keep the steaming time unchanged; When the medicinal material is in the third section and the color change is close to the target value, the steaming time is shortened by 3 minutes and the steaming temperature is lowered by 3°C.

[0034] In this embodiment of the present invention, color change trend data for medicinal materials over consecutive monitoring cycles is obtained from a cloud platform database, and the color change trend difference between adjacent monitoring cycles is calculated. The formula for calculating the color change trend difference is: Difference = Current cycle color change trend value - Previous cycle color change trend value. Based on the calculated color change trend difference, the medicinal material's color change stage is determined. If the difference is greater than 10, the medicinal material is determined to be in the rapid color change stage; if the difference is less than 5, the medicinal material is determined to be in the transient color change stage. When the medicinal material is in the first stage and is determined to be in the transient color change stage, a command is sent to the medicinal material processing equipment via the cloud platform control interface to extend the steaming time by 5 minutes while maintaining the steaming temperature unchanged. Specifically, the current steaming time parameter is read from the cloud platform database, increased by 5 minutes, and the updated steaming time parameter is sent to the medicinal material processing equipment. When the medicinal material is in the second stage and is determined to be in the rapid color change stage, a command is sent to the medicinal material processing equipment via the cloud platform control interface to reduce the steaming temperature by 5°C while maintaining the steaming time unchanged. The specific operation is as follows: the current steaming temperature parameters are read from the cloud platform database, reduced by 5°C, and the updated steaming temperature parameters are sent to the medicinal material processing equipment. When the medicinal material is in the third segment and the color change approaches the target value, a command is sent to the medicinal material processing equipment through the cloud platform control interface to shorten the steaming time by 3 minutes and reduce the steaming temperature by 3°C. The specific operation is as follows: the current steaming time and steaming temperature parameters are read from the cloud platform database, the steaming time is reduced by 3 minutes and the steaming temperature is reduced by 3°C, and the updated steaming time and steaming temperature parameters are sent to the medicinal material processing equipment.

[0035] Preferably, in step S3, during the medicinal material processing, collecting the changes in firmness and elasticity of the medicinal material comprises the following steps: During the medicinal material processing, a pressure sensor is used to collect the firmness data of the medicinal material. The pressure sensor has a measurement range of 0-100 Newtons and an accuracy of 0.1 Newtons. At the same time, a displacement sensor is used to collect the elastic change data of the medicinal material. The displacement sensor has a measurement range of 0-10 mm and an accuracy of 0.01 mm. At the initial stage of medicinal material processing, the compression amount of the medicinal material when a pressure of 10 Newtons is applied is collected through a pressure sensor and recorded as the initial compression amount; at the same time, the rebound amount of the medicinal material after the pressure is released is recorded as the initial elastic response data; During the mid-stage of medicinal material processing, a 20 Newton pressure was repeatedly applied, and the compression and rebound amounts of the medicinal materials were collected and recorded as mid-stage compression and mid-stage elastic response data, respectively. The compression and rebound amounts in the initial stage and the mid-stage were compared, and the change rates of the compression and rebound amounts were calculated to obtain the mid-stage mechanical response data. In the later stage of medicinal material processing, a higher pressure of 30 Newtons is applied, and the compression and rebound amounts of the medicinal materials are collected and recorded as the late compression amount and late elastic response data, respectively; the compression and rebound amounts in the middle stage and the late stage are compared, and the change rate of the compression amount and the change rate of the rebound amount are further calculated to obtain the late mechanical response data.

[0036] In an embodiment of the present invention, a pressure sensor and a displacement sensor are installed on the medicinal material processing equipment. The pressure sensor has a measurement range of 0-100 Newtons with an accuracy of 0.1 Newtons; the displacement sensor has a measurement range of 0-10 millimeters with an accuracy of 0.01 millimeters. After installation, the sensors are calibrated. During the initial processing phase, a pressure of 10 Newtons is applied to the medicinal material using the pressure sensor, while the compression of the medicinal material is recorded using the displacement sensor. This compression is recorded as the initial compression. The pressure is then released, and the displacement of the medicinal material after rebound is recorded. This displacement is recorded as the initial elastic response data. Both the initial compression and initial elastic response data are collected by the sensors and transmitted in real time to a cloud platform for storage. During the mid-stage of medicinal material processing, a pressure of 20 Newtons is repeatedly applied, and the compression of the medicinal material is recorded using the displacement sensor. This is recorded as the mid-stage compression. After the pressure is released, the rebound of the medicinal material is recorded as the mid-stage elastic response data. The mid-stage compression and mid-stage elastic response data are also collected by the sensors and transmitted to the cloud platform. The compression and rebound values ​​at the initial and mid-stages were compared, and the rates of change of compression and rebound values ​​were calculated. The compression rate was calculated as follows: Compression rate = (mid-stage compression - initial compression) / initial compression × 100%; the rebound rate was calculated as follows: Rebound rate = (mid-stage elastic response data - initial elastic response data) / initial elastic response data × 100%. The calculated results were stored in the cloud platform database as mid-stage mechanical response data. In the later stages of medicinal material processing, a higher pressure of 30 Newtons was applied, and the compression of the medicinal material was measured using a displacement sensor, recorded as the late-stage compression. After the pressure was released, the rebound value was recorded as the late-stage elastic response data. The late-stage compression and late-stage elastic response data were collected by sensors and transmitted to the cloud platform. The compression and rebound values ​​at the mid-stage and late stages were compared, and the rates of change of compression and rebound values ​​were further calculated. The formula for calculating the compression rate is: Compression rate = (late compression - mid-term compression) / mid-term compression × 100%. The formula for calculating the rebound rate is: Rebound rate = (late elastic response data - mid-term elastic response data) / mid-term elastic response data × 100%. The calculation results are stored as late mechanical response data in the cloud platform database.

[0037] Preferably, in step S3, identifying the texture change trend of the medicinal material during the processing by analyzing the mechanical response information and adjusting the pressing force parameters of the medicinal material processing equipment include: The average compression and average rebound of the medicinal materials were extracted from the mid-term mechanical response data as mid-term compression characteristics and mid-term rebound characteristics, respectively. The average compression and average rebound of the medicinal materials were extracted from the late mechanical response data as late compression characteristics and late rebound characteristics, respectively. The difference between the late compression characteristic and the mid-term compression characteristic is calculated and recorded as the compression difference; the difference between the late rebound characteristic and the mid-term rebound characteristic is calculated and recorded as the rebound difference; If the compression difference is less than -10% and the rebound difference is less than -10%, the drug material texture is determined to be hard and the elasticity is weakened; if the compression difference is greater than 10% and the rebound difference is greater than 10%, the drug material texture is determined to be soft and the elasticity is enhanced; if the compression difference is between -10% and 10% and the rebound difference is between -10% and 10%, the drug material texture is determined to be stable; When the texture of the medicinal material becomes harder and its elasticity weakens, the pressing force parameters of the medicinal material processing equipment will be reduced by 20%; when the texture of the medicinal material becomes softer and its elasticity increases, the pressing force parameters of the medicinal material processing equipment will be increased by 20%; when the texture of the medicinal material is stable, the pressing force parameters of the medicinal material processing equipment will be kept unchanged.

[0038] In an embodiment of the present invention, mid-term mechanical response data is retrieved from the cloud platform database, and the average values ​​of mid-term compression and rebound are calculated as mid-term compression characteristics and mid-term rebound characteristics, respectively. The mid-term compression characteristic calculation formula is: mid-term compression characteristic = sum of mid-term compression amounts / mid-term measurement times; the mid-term rebound characteristic calculation formula is: mid-term rebound characteristic = sum of mid-term rebound amounts / mid-term measurement times. Late-term mechanical response data is retrieved from the cloud platform database, and the average values ​​of late-term compression and rebound are calculated as late-term compression characteristics and late-term rebound characteristics, respectively. The late-term compression characteristic calculation formula is: late-term compression characteristic = sum of late-term compression amounts / late-term measurement times; the late-term rebound characteristic calculation formula is: late-term rebound characteristic = sum of late-term rebound amounts / late-term measurement times. The difference between the late-term compression characteristic and the mid-term compression characteristic is calculated, recorded as the compression difference; the difference between the late-term rebound characteristic and the mid-term rebound characteristic is calculated, recorded as the rebound difference. The formula for calculating the compression difference is: Compression difference = (late compression characteristic - mid-term compression characteristic) / mid-term compression characteristic × 100%; the formula for calculating the rebound difference is: Rebound difference = (late rebound characteristic - mid-term rebound characteristic) / mid-term rebound characteristic × 100%. If both the compression difference and the rebound difference are less than -10%, the medicinal material texture is considered to have hardened and its elasticity has decreased. If both the compression difference and the rebound difference are greater than 10%, the medicinal material texture is considered to have softened and its elasticity has increased. If both the compression difference and the rebound difference are between -10% and 10%, the medicinal material texture is considered to be stable. When the medicinal material texture has hardened and its elasticity has decreased, a command is sent to the medicinal material processing equipment via the cloud platform control interface to reduce the pressing force parameter by 20%. The specific operation is to read the current pressing force parameter from the cloud platform database, calculate 20% of it, subtract it from the current parameter, and send the adjusted pressing force parameter to the medicinal material processing equipment. When the medicinal material softens and becomes more elastic, a command is sent to the medicinal material processing equipment through the cloud platform control interface to increase the pressing force parameter by 20%. Specifically, the current pressing force parameter is read from the cloud platform database, 20% of the value is calculated and added to the current parameter, and the adjusted pressing force parameter is sent to the medicinal material processing equipment. When the medicinal material texture stabilizes, the pressing force parameter of the medicinal material processing equipment remains unchanged, and no adjustments are made.

[0039] Preferably, uploading the morphological feature supervision data, the color feature supervision data, and the texture feature supervision data to the cloud platform in step S4 includes: Convert morphological feature supervision data into morphological visualization images and number them according to time series; Convert the color feature supervision data into a structured data table of RGB values, record the color data collected each time and its corresponding timestamp, and draw a curve chart of the RGB value change of the color feature; where the timestamp is the horizontal axis and the RGB value is the vertical axis; The texture feature monitoring data is converted into a numerical data table, and the compression force and deformation data collected each time and their corresponding timestamps are recorded. A compression force-deformation relationship diagram of the texture feature is drawn, with the compression force as the horizontal axis and the deformation as the vertical axis. The morphological visualization image, RGB value change curve and compression force-deformation relationship diagram are encrypted and uploaded to the cloud platform for storage to form full-link supervision data.

[0040] In this embodiment of the present invention, data on the morphological characteristics of medicinal materials during processing, including information such as their size and shape, is obtained from a cloud platform database. Image processing technology is used to convert this data into visual morphological images. Specifically, the morphological data is rendered using image processing software (such as OpenCV) to generate intuitive images of the medicinal material's morphology. The visual morphological images are numbered according to a time sequence using the "YYYYMMDDHHMMSS" numbering scheme, where YYYY represents year, MM represents month, DD represents day, HH represents hour, MM represents minute, and SS represents second. A timestamp is recorded for each image and stored in the cloud platform database. Data on the color characteristics of medicinal materials during processing, including information such as spectral reflectance, is obtained from the cloud platform database. The spectral reflectance data is converted into RGB values ​​using a color conversion algorithm. Specifically, based on the spectral reflectance data, the spectral data is converted into corresponding RGB values ​​using a table lookup or calculation formula. The converted RGB values ​​and their corresponding timestamps are recorded in a structured data table. The table fields include timestamp, R value, G value, and B value. Use a data visualization tool (such as Matplotlib or Excel) to plot a graph of the RGB values ​​of the color characteristics, with the timestamp on the horizontal axis and the RGB values ​​on the vertical axis. The plotting parameters include: the horizontal axis time format is set to "YYYY-MM-DDHH:MM:SS", the vertical axis RGB value range is set to 0-255, and the curve colors are red (R), green (G), and blue (B). Obtain data on the texture characteristics of medicinal materials during processing from a cloud platform database, including compression force and deformation data. This data is converted into a numerical data table with fields including timestamp, compression force, and deformation. Use a data visualization tool to plot the compression force-deformation relationship of the texture characteristics, with compression force on the horizontal axis and deformation on the vertical axis. The plotting parameters include: the compression force range is set to 0-100 Newtons on the horizontal axis and the deformation range is set to 0-10 mm on the vertical axis. The data points are presented as a scatter plot, and a trend line is added to illustrate the relationship between compression force and deformation. The generated morphological visualization images, RGB value change curves, and compression force-deformation relationship diagrams are encrypted. The AES encryption algorithm is used with a key length of 256 bits. The encryption process includes: converting the image and chart files into binary data, encrypting the binary data using the AES encryption algorithm, and generating encrypted data files. The encrypted data files are uploaded to the cloud platform for storage. The upload process is carried out using the HTTPS protocol to ensure the security of data transmission. After the upload is completed, the cloud platform generates a data storage path and stores the path information in the database, forming full-link supervision data.

[0041] Of particular importance is the intelligent regulation of the operating parameters of the processing equipment based on the full-link supervision data in step S4 to implement digital visual supervision of medicinal material processing and production. The operations include: Obtain encrypted full-link supervision data from the cloud platform, including morphological, color, and texture characteristics; Use the pre-stored key to decrypt the data, restore the original data format, and parse to obtain a morphological visualization image, an RGB value change curve, and a compression force-deformation relationship diagram; Adjust the cutting parameters of the processing equipment based on the morphological visualization image; if the morphological characteristics of the medicinal material indicate that the medicinal material is large, increase the blade speed of the cutting equipment to 1500 rpm and increase the feed speed to 50 mm / s; if the morphological characteristics of the medicinal material indicate that the medicinal material is small, reduce the blade speed to 1200 rpm and the feed speed to 30 mm / s; Adjust the drying parameters of the processing equipment according to the RGB value change curve; if the color characteristics of the medicinal material indicate that the medicinal material is lighter in color, increase the drying temperature to 60 degrees Celsius and extend the drying time to 30 minutes; if the color characteristics of the medicinal material indicate that the medicinal material is darker in color, reduce the drying temperature to 50 degrees Celsius and shorten the drying time to 20 minutes; Adjust the pressing force parameters of the processing equipment according to the pressing force-deformation relationship diagram; if the texture characteristics of the medicinal material show that the medicinal material is hard, reduce the pressing force to 50 Newtons; if the texture characteristics of the medicinal material show that the medicinal material is soft, increase the pressing force to 80 Newtons.

[0042] In an embodiment of the present invention, encrypted full-link regulatory data, including morphological, color, and texture characteristics, is retrieved from the cloud platform database through a cloud platform interface. The data is stored as an encrypted binary file. A pre-stored key is used to decrypt the encrypted data. The AES encryption algorithm with a 256-bit key length is employed. The decryption process includes reading the binary data from the encrypted file, decrypting the data using the AES decryption algorithm and the pre-stored key, and restoring the original data format. The decrypted data includes a morphological visualization image, a graph of RGB value variations, and a graph showing the relationship between the pressing force and deformation. The morphological visualization image is parsed to analyze the morphological characteristics of the medicinal material, particularly its size. Image processing techniques (such as OpenCV) are used to analyze the morphological visualization image and extract the dimensional information. If the morphological characteristics of the medicinal material indicate a large size (e.g., exceeding 10 cm in length), a command is sent to the cutting device through the cloud platform control interface to adjust the blade speed to 1500 rpm and increase the feed rate to 50 mm / s. If the herbal material's morphological characteristics indicate a small size (e.g., less than 5 cm in length), a command is sent to the cutting equipment via the cloud platform control interface to reduce the blade speed to 1200 rpm and the feed speed to 30 mm / s. The RGB value curve is analyzed to analyze the herbal material's color characteristics, particularly the changing trends of the RGB values. This RGB value curve is analyzed using data processing tools (such as Excel or Python) to extract the average RGB value of the herbal material. If the herbal material's color characteristics indicate a light color (e.g., the R, G, and B components of the average RGB value are all greater than 150), a command is sent to the drying equipment via the cloud platform control interface to increase the drying temperature to 60°C and extend the drying time to 30 minutes. If the herbal material's color characteristics indicate a dark color (e.g., the R, G, and B components of the average RGB value are all less than 100), a command is sent to the drying equipment via the cloud platform control interface to reduce the drying temperature to 50°C and shorten the drying time to 20 minutes. The compression force-deformation relationship diagram is analyzed to analyze the herbal material's texture characteristics, particularly the relationship between compression force and deformation. Data processing tools analyze the relationship between pressing force and deformation to extract information about the medicinal material's hardness. If the material's texture indicates a hard material (for example, a pressing force greater than 70 Newtons and a deformation less than 5 mm), a command is sent to the processing equipment via the cloud platform's control interface to reduce the pressing force to 50 Newtons. If the material's texture indicates a soft material (for example, a pressing force less than 30 Newtons and a deformation greater than 8 mm), a command is sent to the processing equipment via the cloud platform's control interface to increase the pressing force to 80 Newtons.

[0043] Of particular importance is that in step S4, the operation of intelligently adjusting the operating parameters of the processing equipment based on the full-link supervision data to implement digital visual supervision of medicinal material processing and production also includes: Convert the motor speed setting value into a pulse frequency signal, the temperature setting value into a voltage signal, and the pressure setting value into a current signal, ultimately generating a composite control signal; Through industrial communication protocols and redundant transmission mechanisms, the composite control signal is sent to the control system of the processing equipment through two independent communication links; After receiving the control signal, the control system of the processing equipment automatically analyzes the signal content and transmits the set values ​​of motor speed, drying temperature and pressing force to the corresponding hardware driver module; During the adjustment process, the equipment control system collects the actual operating parameters of motor speed, drying temperature and pressing force at fixed time intervals and compares them with the set values; if the deviation exceeds the preset threshold, the secondary adjustment mechanism is automatically triggered until the operating parameters are consistent with the set values.

[0044] In an embodiment of the present invention, the motor speed setting value is converted into a pulse frequency signal. Assuming the motor speed setting value is 1500 rpm, the corresponding pulse frequency is calculated based on the parameter requirements of the motor driver. For example, if the pulse equivalent of the motor driver is 1000 pulses / rev, the pulse frequency calculation formula is: pulse frequency = (1500 rpm × 1000 pulses / rev) / 60 seconds = 25000 pulses / second. The drying temperature setting value is converted into a voltage signal. Assuming the drying temperature setting value is 60°C, the corresponding voltage value is calculated based on the parameter requirements of the temperature controller. For example, if the output range of the temperature controller is 0-10V and the corresponding temperature range is 0-100°C, the voltage signal calculation formula is: voltage signal = (60°C / 100°C) × 10V = 6V. The pressing force setting value is converted into a current signal. Assuming the pressing force setting value is 50 Newtons, the corresponding current value is calculated based on the parameter requirements of the pressure controller. For example, if the pressure controller's output range is 4-20mA, corresponding to a pressure range of 0-100 Newtons, the current signal calculation formula is: Current Signal = (50 Newtons / 100 Newtons) × (20mA - 4mA) + 4mA = 14mA. The converted pulse frequency, voltage, and current signals are integrated to generate a composite control signal. The composite control signal format includes a signal header, motor speed pulse frequency, drying temperature voltage value, pressing force current value, and a signal tail. Signal transmission utilizes an industrial standard communication protocol (such as Modbus TCP / IP or Profibus). Furthermore, a redundant transmission mechanism is employed, sending the composite control signal to the processing equipment's control system via two independent communication links. Each communication link has an independent physical connection and communication interface. The composite control signal is transmitted simultaneously over both communication links. During transmission, frame synchronization and checksum addition are performed to prevent errors and interference during signal transmission. For example, a cyclic redundancy check (CRC) algorithm is used to calculate the checksum and append a checksum to the signal tail. After receiving the composite control signal, the control system of the processing equipment automatically analyzes the signal content. The control system first identifies the signal header and tail to confirm signal integrity. It then extracts the motor speed pulse frequency, drying temperature voltage, and pressing force current. The parsed motor speed setpoint is transmitted to the motor driver module, the drying temperature setpoint to the temperature control module, and the pressing force setpoint to the pressure control module. Each hardware driver module configures its parameters based on the received setpoints and prepares to perform the corresponding control task. During the adjustment process, the equipment control system collects the actual operating parameters of the motor speed, drying temperature, and pressing force at regular intervals (e.g., every 1 second). This collection process is accomplished through sensors, which convert the actual operating parameters into electrical signals and transmit them to the control system. The control system then compares the collected actual operating parameters with the setpoints.If the deviation exceeds a preset threshold (for example, motor speed deviation exceeding ±10 rpm, drying temperature deviation exceeding ±2°C, or pressing force deviation exceeding ±2 Newtons), a secondary adjustment mechanism is automatically triggered. Based on the deviation, the control system adjusts the corresponding hardware driver modules until the operating parameters are consistent with the set values. For example, if the motor speed is lower than the set value, the pulse frequency of the motor driver is increased; if the drying temperature is higher than the set value, the output voltage of the temperature controller is reduced.

[0045] In this specification, a digital visual supervision system for medicinal material processing and production based on a cloud platform is provided, which is used to implement the above-mentioned digital visual supervision method for medicinal material processing and production based on a cloud platform. The digital visual supervision system for medicinal material processing and production based on a cloud platform includes: The morphological feature monitoring module is used to collect the shape, size, and distribution characteristics of the nodules of the medicinal material rhizomes during the medicinal material processing process to obtain medicinal material image information; based on the medicinal material image information, it identifies the trend of medicinal material morphological changes and adjusts the slice thickness and drying time of the medicinal material processing equipment to obtain morphological feature monitoring data; The color feature monitoring module is used to collect the color variation range of the medicinal materials from light yellow when fresh to yellow-brown after processing during the medicinal material processing process to obtain the medicinal material spectral information; monitor the light intensity and ambient temperature of the processing workshop; analyze the medicinal material color variation trend based on the light intensity and ambient temperature of the medicinal material spectral information, and adjust the steaming time and steaming temperature of the medicinal material processing equipment to obtain color feature monitoring data; The texture characteristic monitoring module is used to collect changes in the firmness and elasticity of medicinal materials during the processing of medicinal materials to obtain the mechanical response information of the medicinal materials. By analyzing the mechanical response information, the changing trend of the medicinal material texture during the processing of the medicinal materials is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic monitoring data. The intelligent control module for processing equipment is used to upload morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, the operating parameters of the processing equipment are intelligently controlled to perform digital visual supervision operations for medicinal material processing and production.

[0046] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0047] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A digital visual supervision method for medicinal material processing and production based on a cloud platform, characterized in that: The following steps are involved: Step S1: During the medicinal material processing process, the shape, size, and distribution characteristics of the nodules of the medicinal material rhizome are collected to obtain medicinal material image information; based on the medicinal material image information, the morphological change trend of the medicinal material is identified, and the slice thickness and drying time of the medicinal material processing equipment are adjusted to obtain morphological characteristic supervision data; Step S2: During the medicinal material processing process, the color change range of the medicinal material from light yellow when fresh to yellow-brown after processing is collected to obtain medicinal material spectral information; the light intensity and ambient temperature of the processing workshop are monitored; the medicinal material spectral information is analyzed based on the light intensity and ambient temperature to determine the medicinal material color change trend, and the steaming time and steaming temperature of the medicinal material processing equipment are adjusted to obtain color characteristic supervision data; Step S3: During the medicinal material processing, the firmness and elasticity changes of the medicinal material are collected to obtain the mechanical response information of the medicinal material; by analyzing the mechanical response information, the texture change trend of the medicinal material during the processing is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic supervision data; Step S4: Upload the morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, intelligently adjust the operating parameters of the processing equipment to perform digital visual supervision operations for medicinal material processing and production.

2. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 1 is characterized in that: In step S1, during the medicinal material processing, the shape, size and distribution characteristics of the nodules of the medicinal material rhizomes are collected, including: Place the rhizome of the medicinal material on the collection platform, adjust the angle and position of the collection platform, and make sure the nodule part of the rhizome is completely within the collection field of view; Images were collected from the front, middle, and end of the rhizome of the medicinal material, and the focal length of the acquisition device was adjusted each time; The collected images are segmented and only the image information of the nodule part is retained; The segmented nodule images are normalized to adjust all nodule images to a uniform size ratio; Grayscale processing is performed on the standardized nodule image to obtain a nodule grayscale image; The nodule grayscale image is subjected to contrast enhancement processing to highlight the outline of the nodule to obtain a nodule enhanced image; Delineate the nodule enhancement image and identify the nodule boundary contour by pixel-by-pixel scanning; The nodule boundary contours were quantitatively analyzed, and the area and perimeter of each nodule were calculated to determine the nodule size; Measure the aspect ratio and roundness of the nodule boundary outline to determine the shape characteristics of the nodule; The nodule distribution in the nodule-enhanced image was counted, the distance between adjacent nodules was measured, and the distribution density of nodules on the rhizome was calculated to determine the nodule distribution characteristics.

3. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 2 is characterized in that: In step S1, identifying the morphological change trend of the medicinal material according to the medicinal material image information and adjusting the slice thickness and drying time of the medicinal material processing equipment include: The nodule shapes in the medicinal material images are detected frame by frame, and the complexity of the medicinal material shape is determined based on the tortuosity and number of branches of the nodule contour. The complexity of the medicinal material shape is correlated and compared with the processing time series to obtain the trend of nodule shape changes. The average area of ​​the nodule size in each frame of the image is calculated, and the change trend of the nodule size over the processing time series is determined based on the average area of ​​the nodule size to obtain the change trend of the nodule size; Count the nodule distribution density in each frame of the image; compare the nodule distribution density with the processing time series to obtain the nodule uniformity change trend; If the nodule shape change trend shows that the shape complexity increases over time, and the nodule size change trend shows that the average area increases over time, the slice thickness of the medicinal material processing equipment will be gradually reduced, and each adjustment amount is 5% of the initial slice thickness; If the nodule shape change trend shows that the shape complexity decreases over time, and the nodule size change trend shows that the average area decreases over time, the slice thickness of the medicinal material processing equipment is gradually increased, with each adjustment amount being 10% of the initial slice thickness; If the trend of nodule uniformity shows that the distribution density increases over time, the drying time of the medicinal material processing equipment will be gradually extended, with each adjustment amount being 20% ​​of the initial drying time; If the trend of nodule uniformity changes shows that the distribution density decreases over time, the drying time of the medicinal material processing equipment will be gradually shortened, with each adjustment amount being 25% of the initial drying time.

4. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 1 is characterized in that: In step S2, during the medicinal material processing, the range of changes in the collected medicinal materials from light yellow when fresh to yellow-brown after processing includes: During the medicinal material processing, the top, middle and bottom of the medicinal material surface, as well as the edge of the medicinal material nodules are selected as selected areas; for each selected area, its spectral reflectance data is collected separately; Perform multi-dimensional quantitative processing on the spectral reflectance data, calculate the ratio of the reflectance at each wavelength point to the standard white reflectance, and obtain the color index of each segment; average the color index of each segment to obtain the average color index of the segment; calculate the standard deviation of the color index of each segment to obtain the color variation coefficient of the segment; According to the color variation coefficient, the color change range of the medicinal material during processing is divided into multiple segments. Each segment corresponds to a specific interval of the medicinal material's color change, and the color segment of the medicinal material is obtained; among them, the first segment is the light yellow interval when the medicinal material is fresh, the second segment is the intermediate transition color interval during the medicinal material processing, and the third segment is the yellow-brown interval after the medicinal material processing is completed.

5. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 1 is characterized in that: Analyzing the color change trend of the medicinal material by using the light intensity and ambient temperature in step S2 includes: According to the spectrum information of medicinal materials, the spectrum intensity, spectrum reflectance and absorption peak intensity of each segment corresponding to the specific wavelength range are extracted to obtain the spectrum characteristic parameters; Based on the influence of light intensity and ambient temperature on the spectral characteristic parameters, the change parameters of the spectral characteristic parameters of the medicinal materials in each segment are calculated respectively under different light intensities and ambient temperatures. The influence coefficients of light intensity and ambient temperature on the color change of the medicinal materials are determined and recorded as light intensity-temperature influence coefficients; The light intensity-temperature influence coefficient is used to detect the influence of changes in the color characteristics of the medicinal materials in the medicinal material spectrum information to obtain the color change trend of the medicinal materials.

6. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 4 is characterized in that: Adjusting the steaming time and steaming temperature of the medicinal material processing equipment in step S3 includes: The color change rate of medicinal materials is obtained by calculating the difference between the color change trends of the medicinal materials; If the difference in the rate of change of the medicinal material color is greater than 10, it is determined that the medicinal material is in the stage of rapid color change; if the difference in the rate of change of the medicinal material color is less than 5, it is determined that the medicinal material is in the stage of short color change; Adjust the steaming time and temperature of the medicinal material processing equipment according to the color change stage of the medicinal material and the color segment of the medicinal material; When the medicinal material is in the first stage and the color changes briefly, extend the steaming time by 5 minutes and keep the steaming temperature unchanged; When the medicinal materials are in the second stage and the color changes rapidly, reduce the steaming temperature by 5°C and keep the steaming time unchanged; When the medicinal material is in the third section and the color change is close to the target value, the steaming time is shortened by 3 minutes and the steaming temperature is lowered by 3°C.

7. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 1 is characterized in that: In step S3, during the medicinal material processing, collecting the changes in the firmness and elasticity of the medicinal material includes the following steps: During the medicinal material processing, a pressure sensor is used to collect the firmness data of the medicinal material. The pressure sensor has a measurement range of 0-100 Newtons and an accuracy of 0.1 Newtons. At the same time, a displacement sensor is used to collect the elastic change data of the medicinal material. The displacement sensor has a measurement range of 0-10 mm and an accuracy of 0.01 mm. At the initial stage of medicinal material processing, the compression amount of the medicinal material when a pressure of 10 Newtons is applied is collected through a pressure sensor and recorded as the initial compression amount; at the same time, the rebound amount of the medicinal material after the pressure is released is recorded as the initial elastic response data; During the mid-stage of medicinal material processing, a 20 Newton pressure was repeatedly applied, and the compression and rebound amounts of the medicinal materials were collected and recorded as mid-stage compression and mid-stage elastic response data, respectively. The compression and rebound amounts in the initial stage and the mid-stage were compared, and the change rates of the compression and rebound amounts were calculated to obtain the mid-stage mechanical response data. In the later stage of medicinal material processing, a higher pressure of 30 Newtons is applied, and the compression and rebound amounts of the medicinal materials are collected and recorded as the late compression amount and late elastic response data, respectively; the compression and rebound amounts in the middle stage and the late stage are compared, and the change rate of the compression amount and the change rate of the rebound amount are further calculated to obtain the late mechanical response data.

8. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 7 is characterized in that: In step S3, the trend of the texture change of the medicinal material during the processing is identified by analyzing the mechanical response information, and the pressing force parameters of the medicinal material processing equipment are adjusted, including: The average compression and average rebound of the medicinal materials were extracted from the mid-term mechanical response data as mid-term compression characteristics and mid-term rebound characteristics, respectively. The average compression and average rebound of the medicinal materials were extracted from the late mechanical response data as late compression characteristics and late rebound characteristics, respectively. The difference between the late compression characteristic and the mid-term compression characteristic is calculated and recorded as the compression difference; the difference between the late rebound characteristic and the mid-term rebound characteristic is calculated and recorded as the rebound difference; If the compression difference is less than -10% and the rebound difference is less than -10%, the drug material texture is determined to be hard and the elasticity is weakened; if the compression difference is greater than 10% and the rebound difference is greater than 10%, the drug material texture is determined to be soft and the elasticity is enhanced; if the compression difference is between -10% and 10% and the rebound difference is between -10% and 10%, the drug material texture is determined to be stable; When the texture of the medicinal material becomes harder and its elasticity weakens, the pressing force parameters of the medicinal material processing equipment will be reduced by 20%; when the texture of the medicinal material becomes softer and its elasticity increases, the pressing force parameters of the medicinal material processing equipment will be increased by 20%; when the texture of the medicinal material is stable, the pressing force parameters of the medicinal material processing equipment will be kept unchanged.

9. The digital visual supervision method for medicinal material processing and production based on a cloud platform according to claim 1 is characterized in that: Uploading the morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform in step S4 includes: Convert morphological feature supervision data into morphological visualization images and number them according to time series; Convert the color feature supervision data into a structured data table of RGB values, record the color data collected each time and its corresponding timestamp, and draw a curve chart of the RGB value change of the color feature; where the timestamp is the horizontal axis and the RGB value is the vertical axis; The texture feature monitoring data is converted into a numerical data table, and the compression force and deformation data collected each time and their corresponding timestamps are recorded. A compression force-deformation relationship diagram of the texture feature is drawn, with the compression force as the horizontal axis and the deformation as the vertical axis. The morphological visualization images, RGB value change curves and compression force-deformation relationship diagrams are encrypted and uploaded to the cloud platform for storage to form full-link supervision data.

10. A digital visual supervision system for medicinal material processing and production based on a cloud platform, characterized in that: For executing the digital visual supervision method for medicinal material processing and production based on a cloud platform as claimed in claim 1, the digital visual supervision system for medicinal material processing and production based on a cloud platform comprises: The morphological feature monitoring module is used to collect the shape, size, and distribution characteristics of the nodules of the medicinal material rhizomes during the medicinal material processing process to obtain medicinal material image information; based on the medicinal material image information, it identifies the trend of medicinal material morphological changes and adjusts the slice thickness and drying time of the medicinal material processing equipment to obtain morphological feature monitoring data; The color feature monitoring module is used to collect the color variation range of the medicinal materials from light yellow when fresh to yellow-brown after processing during the medicinal material processing process to obtain the medicinal material spectral information; monitor the light intensity and ambient temperature of the processing workshop; analyze the medicinal material color variation trend based on the light intensity and ambient temperature of the medicinal material spectral information, and adjust the steaming time and steaming temperature of the medicinal material processing equipment to obtain color feature monitoring data; The texture characteristic monitoring module is used to collect changes in the firmness and elasticity of medicinal materials during the processing of medicinal materials to obtain the mechanical response information of the medicinal materials. By analyzing the mechanical response information, the changing trend of the medicinal material texture during the processing of the medicinal materials is identified, and the pressing force parameters of the medicinal material processing equipment are adjusted to obtain the texture characteristic monitoring data. The intelligent control module for processing equipment is used to upload morphological feature supervision data, color feature supervision data, and texture feature supervision data to the cloud platform to form full-link supervision data; based on the full-link supervision data, the operating parameters of the processing equipment are intelligently controlled to perform digital visual supervision operations for medicinal material processing and production.

Images